Big Prisons, Big Dreams
Big Prisons, Big Dreams
Big Prisons, Big Dreams
Big Prisons,
Big Dreams
Crime and the Failure of
Ame ricas Pe nal System
Michae l J. Lynch
Rutg e r s U n ive r s i ty P re s s
New Brunswick, New Jersey, and London
C onte nt s
ix
Preface
20
49
82
110
146
173
203
A Consuming Culture
220
Notes
References
Index
229
241
253
vii
P re face
This book was written in response to troubling trends in American society. Since 9/11, America has been edging closer and closer to a limited
democracy that accepts the curtailment of freedom and the enhancement of governmental power and control as the price for safety. This
movement, however, has been underway for decades in the way America
responds to crime, especially street crimes, or those offenses most likely
to be engaged in by the lower classes and Americans of color. It is no
accident that these crimes, more so than the more harmful behaviors of
corporate and government ofcials, are the prime subject of crime control, and that the prime suspects are those unlike usthey represent
economic decay and difference.
It is also no accident that Americas use of imprisonment has grown
so dramatically in recent decades, and that the prison targets the poor
and minorities. This is true despite the fact that they also do not represent the greatest threat to our health and well-being. Rather, it is the
corporate criminal who pollutes the environment, uses his economic
and political power to alter the course of American politics and law,
who poses the greatest threat to the average American. But this book is
not about them; it is about the runaway train that has become Americas
penal system.
Today the average citizen regards the prison as an appropriate response to crime; and so too do Americas politicians. As a result, the rate
of imprisonment in the United States has expanded exponentially since
1973. Since then, the number of inmates imprisoned in the United States
has grown each and every year. More than thirty years later, our prison
system is the biggest in the world, in terms of both raw numbers and
rates. And, contrary to popular opinion, the United States has the longest
average prison sentences of any nation in the world. And still, we have a
substantial level of crime.
ix
Preface
These few facts, which are examined in detail in the pages that
follow, make it clear that the U.S. penal system is the harshest in the
world. Add to these facts the observation that the United States, unlike any other Western democracy, also employs the death penalty for
criminals, and the picture of an extensively repressive penal system is
nearly complete. These drastic measures, however, have not lowered our
rate of crime.
To round out this picture, we must add that the people subject to
this form of repressive control are the poor and the minorities in our
nation, the least well-off, those who have the fewest quality choices to
make during their life courses. In contrast, the well-off get away with
their crimes, or, if punished, are treated rather kindly in comparison.
Not only is this system of punishment repressive, but it fails at its mission of reducing crime. The balance of evidenceand we should make
it clear here that we mean the balance of scientic evidence produced by
independent social scientists who are not supported by grants, stipends,
or salaries from conservative think tanks, and who have not produced
pro-prison research as part of their governmental dutiesillustrates that
prisons are not an effective crime control response. Much of this book is
dedicated to demonstrating this point.
At this stage in history, it is also time to recognize that there looms
on the horizon important environmental and energy problems that must
be addressed now, which also have important implications for the future
of imprisonment in America. In contrast to the position taken by the
Bush White House, scientists around the world and leaders of the majority of other nations have come to recognize that the most important issues facing the world today with respect to long-term survival are global
warming and the end of oil. How will Americas big prison system fare
in, or respond to, a world where oil is becoming more and more scarce,
and where burning oil produces global warming? When will U.S. criminal justice policy experts recognize the end of oil and global warming
as signicant issues that should affect criminal justice policy? When will
these issues become so important that they will alter the practice of imprisonment in America? These issues are examined in this book. To my
knowledge, outside of the few mentions I have made of these themes
elsewhere, this is the rst book on criminal justice issues and policy to
make environmental issues a major theme.
Preface
xi
I expect that this work will be controversial. Without such controversy, there is little hope of progress on the issues examined here.
I owe thanks to several people who contributed to the writing of
this book. More than anyone else, I must thank Graeme R. Newman,
Distinguished Teaching Professor at the School of Criminal Justice, State
University of New York at Albany. Newman recruited me to be his graduate
student at a time when the radical ideas I was expressing in class were
not so popular among the schools student body or faculty. Without his
encouragement, my graduate career would not have lasted long. I have
known Newman for two decades now, and during that period his role
in my life has been transformed from mentor to friend. Anyone familiar
with Newmans work will see the many contrasting views we hold on
issues, and that I am certainly not repeating the words or views of my
mentor in writing this book. Yet, despite the differences in our views,
Newman has always been supportive of my workand the work of all
of his studentswhich has made him an invaluable mentor to numerous
students with divergent points of view. In addition, it would be impossible for me to write a book that deals with the topic of punishment
without thinking constantly of Newmans widely respected book, The
Punishment Response, which analyzes the history and philosophy of punishment. In order to fully recognize the importance of Newmans scholarshipand friendshipto my own work, I have chosen to dedicate this
book to him as teacher, scholar, and friend.
I also owe thanks to Raymond Michalowski, the series editor, who,
despite having worked with me on other projects, read my manuscript
and encouraged me to submit it to Rutgers University Press for review.
Ray was also available for comment as I wrote this manuscript. Despite my inability to publish any of my previous work on the topic of
global warming, the end of oil, and criminal justice, thanks are due to
Todd Clear for encouraging me in this endeavor, and being among the
only criminologists I know to see the relevance of this issue. Thanks are
also due to Adi Hovav, social science and religion editor for Rutgers
University Press. Adis comments have been most useful in preparing this
book for press. I would also like to thank the manuscript reviewers for
Rutgers University Press, Jeffrey Ian Ross and a second, anonymous reviewer. Special thanks are owed to several of my colleagues who provided
xii
Preface
useful comments and criticisms on ideas I bounced off them from time
to time: John Cochran, Thomas Mieczkowski, Wilson Palacios, Herman
Schwendinger, Shayne Jones, and Thomas Kovandzic. Thanks are also due
to Elizabeth S. Cass (Ph.D.), my wife, who manages to take the more conservative view against which I can test my ideas, and who can still put up
with me after twenty years together. Finally, I owe thanks to the University
of South Florida, which granted me a sabbatical during which I wrote the
main portion of this book.
Chap te r 1
Introduction
Big, Dark Secrets and
A m e ricas Prison System
The prison, the darkest region in the apparatus of justice, it is the place where the power to punish, which
no longer dares to manifest itself openly, silently . . .
function[s]. Michel Foucault, Discipline and Punish
B i g P ri s on s, B i g D ream s
Introduction
Hi di ng B e hi nd Bi g ne s s
Behind Americas big prison system lurk dark secrets hidden by assumptions about the nature of punishment and the ability of imprisonment to reduce crime. These dark secrets are known to scholars, though
many other scholars have played a part in keeping those secrets hidden
and in supporting Americas massive prison expansion. Nor have policy
makers shared these secrets with the general public, which, for its part,
continues to believe that crime will be controlled if the system of punishment becomes more severe and the prison system gets bigger. The
secrets about the failures of imprisonment as an effective crime control
strategy are also obscured by common American cultural themes, including the belief that bigger is better. Americans lust after big cars,
such as four-wheel drive SUVs, even though the majority of these vehicles never leave suburban or urban terrains, negating the need for
an SUVs special capabilities. They crave the security of the SUV, even
though these vehicles have proven to be more unsafe than cars in realworld conditions. Americans continue to drive SUVs, even though the
majority of these vehicles produce more pollution and consume greater
quantities of gas, edging us closer to catastrophic greenhouse effects and
the depletion of the worlds oil reserve. The American tendency to consume too much isnt limited to cars. Americans desire big-screen TVs so
they can watch the big game. They crave big bucks and big houses on
a big lot as evidence of the status they hold. They supersize their meals
at chain food restaurants and shop at superstores, while watching their
weight and cholesterol rise, standing by as superstores crush small shop
owners. Americans have big dreams and the big debts that go along with
the big American dream of consumption. They consume oil at a rate
that is ve times the size of the populationone-quarter of the oil produced in the world each year is consumed in America. Americans also
have a big crime rate. More than that, they also have the worlds biggest
correctional system.
This desire for bigness has created a big problem in America. This
big problem is what Austin and Irwin (2003) call Americas imprisonment binge, which we can dene as the tendency for Americas prison
population to continue to expand. This expansion has been underway since 1973, through Republican and Democratic administrations,
B i g P ri s on s, B i g D ream s
through good and bad economic times, unabated through the dawning of a new century. The imprisonment binge is a big problem to the
extent that imprisonment fails to accomplish its mission of reducing
crime; to the extent that it has had large negative effects on American
minorities and the poor; to the extent that it negatively affects not only
those upon whom it is imposed, but their families and communities as
well; and to the extent that its costs (nancial, social, and environmental)
far exceed its worth. And, rather than own up to this problem, Americas
political and policy leaders have declared that the real problem is that
prisons havent yet gotten big enough to serve as an effective crime
control mechanism.
The chapters that follow examine the growth of the American prison
system over the past three decades, how much Americans spend on imprisonment, whether imprisonment is an effective crime control strategy,
and whether America can continue to operate its big prison system in
an oil-strapped world. In examining these questions, this book challenges
many widely held beliefs about imprisonment, though this is not the
rst book to do so. In writing this book, I am adding my voice to the
list of scholars who have discovered that our big prison system doesnt
accomplish the big goal it set for itself. Before beginning this journey, let
us turn to a review of the literature on imprisonment.
P ri or R e search
There is a signicant scholarly literature on prisons. Surprisingly,
only a small portion of that literature actually examines the question of
prison effectiveness. With respect to crime control, existing studies examine a variety of questions such as: Do prisons deter other criminals?
Does a rise in the rate of imprisonment lead to a reduction in crime?
When the rate of imprisonment increases, are the crime reduction effects seen across all crimes? Are the incapacitative effects of imprisonment large enough to reduce crime? When current criminal populations are incarcerated, are they replaced by new offenders?
There is no consistent answer to the question of whether prisons reduce crime. From a scientic and statistical perspective, this uncertainty
alone suggests that prisons probably are an ineffective mechanism for
reducing crime.
Introduction
B i g P ri s on s, B i g D ream s
Introduction
groups, there is some support in academic literature for the idea that
prisons reduce crime. Historically this idea has been promoted by the
research of economists, most notably Nobel Prizewinner Gary Becker
(1968). In the economic view, human beings are viewed as rational actors
(this assumption is also found in numerous criminological studies). As
such, their behavior is seen as responsive to external conditions such as
punishment. Rational actors, in other words, respond to punishment by
avoiding criminal behavior.
Using attitudinal surveys, some criminologists have found that sentence severity marginally increases deterrence (Grasmick and Bryjack,
1980; Kleeper and Nagin, 1989). The problem such research fails to acknowledge is the t between attitudes and behavior or the attitude-behavior consistency problem (Bandura, 1977, 1986). As an example, health
researchers have discovered that attitude-behavior consistency or the state
of rational thinking can be disrupted by, among other things, incomplete
knowledge (Sapp, 2001). As Sapp notes, a lack of knowledge may make it
impossible for an individual to act rationally with respect to their beliefs.
Regardless of this caveat, others have found that perceived severity is not
related to deterrence (Paternoster and Iovanni, 1986). Thus, the evidence
on this issue has been mixed.
Support for the deterrent effects of imprisonment have also been
found for specic crime types. For instance, Weinrath and Gartell (2001)
found specic deterrence effects for increased sentence lengths among
a sample of offenders sentenced for drunk driving in Alberta, Canada.
Prison sentence length was associated with the tendency toward diminished recidivism. It should be noted, however, that sentences for drunk
driving in this Canadian study were rather short, especially compared
to U.S. standards. Indeed, drunk drivers were deterred by sentences that
were signicantly shorter than those handed out for other offenses.
Weinrath and Gratell argued that longer prison sentenceswhich in this
case amounted to six months or more in lengthproduced a greater deterrent effect than shorter sentences. Thus, compared to U.S. sentencing
practices, longer Canadian sentences are still quite lenient.
The crime suppression effect of imprisonment has also been studied
with respect to homicide offenders. In a recent study, Kovandzic et al.
(2004) discovered that prison population growth greatly reduced homicide rates. The authors report that the reduction was on the order of 1
B i g P ri s on s, B i g D ream s
less homicide for every 200 additional people incarcerated. The authors,
like others who have performed similar research, failed to explore the
policy implications of this empirical result. The problem is that reducing
homicides through incarceration would require an extensive expansion
of the prison population. Consider, for example, that there were approximately 14,200 homicides in the United States in 2004, the year of the
Kovandzic study. Using the Kovandzic studys estimate of 1 less homicide
for every 200 additional offenders incarcerated, the elimination of homicide through incarceration aloneand assuming there are no other causes
of homicide beside the level of punishmentwould require locking up
an additional 2.8 million people. Doing so would more than double the
already exaggerated level of imprisonment in the United States.
Similar evidence of a deterrent effect of imprisonment rates on
homicide rates has been found in several other studies. Using data for
the entire United States for the period 19301994, Marvel and Moody
(1997) reported that a 10 percent increase in the imprisonment rate corresponded with a 13 percent reduction in homicides. While this seems
like a signicant trade-off because crime is lowered by more than the
increase in punishment, currently a 10 percent increase in incarceration
amounts to an additional 220,000 new inmates. This would lower the
number of homicides (14,200 * .13) by 1,846. Assuming that the number of homicide offenders deterred or incapacitated by an expansion
in the use of imprisonment remains constant, and assuming no other
factors cause homicides, eliminating the homicide problem through incarceration would require incarcerating nearly 1.7 million more inmates,
a somewhat lower estimate than derived from the Kovndzic study, but
still signicantly large. In an earlier study, Devine, Shelley, and Smith
(1988) used U.S. imprisonment and homicide rate data for 19481985
and found that a 10 percent increase in the incarceration rate would
lower the homicides rate by 15 to 19 percent (the variation is due to the
specic variables included in the models specied in this study).
The deterrence and incapacitation effects of imprisonment on homicide reported in the above research are questionable, however. For
example, consider that in 1973, at the beginning of the imprisonment
binge, there were 19,640 homicides in the United States (Fox and Zawitz, 2006). In 2004, the number of homicides had fallen by 3,503 to
16,137. During that same period, the number of incarcerated offenders
Introduction
10
B i g P ri s on s, B i g D ream s
that result from the average property crime. Yet, these results could also
be questioned. In 1971 there were 198,061 inmates in U.S. prisons, and
by 1989, this gure had increased by 482,846 to 680,907. If Marvel and
Moody are correct, and assuming that no other factors caused crime to
rise or fall, the number of offenses suppressed by the increase in incarceration by 1989 should be more than 8.2 million (if each incarcerated
offender suppressed the number of crimes by 17, then 17 * 482, 846
should yield the number of suppressed crimes). In 1971, there were
slightly less than 5.4 million crimes known to police; by 1989 that gure had increased to slightly more than 12 million. This trend, although
it only measures known crimes, seem inconsistent with the claim that
an increased rate of imprisonment during this period led to a substantial
reduction in crime.
Summarizing a number of studies produced prior to the mid-1980s,
Visher (1987) noted that assessments of enhanced sentencing practices
enacted during the 1970s and early 1980s produced crime reduction estimates that varied from 10 to 30 percent. The variations related not only
to differences in the type of sentencing practice examined, but also to the
selection of the research method and variables employed in each study. In
reviewing these studies, Visher astutely noted that while sentencing reforms seemed to reduce crime, they did so only by substantially increasing the size of the incarcerated population. In light of this observation,
and in an effort to avoid the problem of extensive prison growth, Visher
recommended the use of selective incapacitation strategies that carefully
targeted specic kinds of offenders.
Literature That Fails to Support Prison Effectiveness
The literature reviewed above supports the idea that prisons are an
effective means for controlling crime. Indeed, if one accepts the validity of these studies, then the number of crimes reduced by expanded
imprisonment appears signicantly large. The results reported above are,
however, challenged by the results from a number of other studies.
In 1998, one of the leading criminal justice policy experts in the
United States, Alfred Blumstein, argued that even though incarceration
rates increased steadily and are now almost quadruple what they were
20 years ago, most crime rates have remained conned within a fairly
narrow range, with no strong trend. Perhaps most strikingly, we have
Introduction
11
not seen the anticipated downward trend in crime rates that might have
been expected as a result of the growth in incarceration (127). Using
data from 1972 through 1995, Blumstein argued that crime was constant overall, with both upward and downward uctuations. Examining
crime and imprisonment data from 1980 through 1994, Blumstein shows
a long-term increase in crime at the same time that imprisonment rates
were rising.
Blumsteins conclusions are supported by several other studies. DeFina and Arvanites employed state level data covering 19711998 to
study the impact of incarceration rates on the seven major crime types.
While some crimes showed evidence of a deterrent effect from rising
rates of incarceration, most did not, leading the researchers to conclude
that it was inappropriate to assume that imprisonment affected crime
at the national level.
Johnston (1999) examined data on offender recidivism rates after
their release from prison. He concluded that unless researchers determine and society changes the causes of what makes some individuals offend, the reactive approaches currently used will continue and the crime
rate will not improve (1). A similar conclusion was reached by Smith
(1997), who studied the impact of various crime control policies on homicide rates in the United States between 1976 and 1990. Moreover,
Smith found that the effect of prison population size on the homicide
rate was the opposite of that predicted by deterrence theory.
In contrast to several of their other studies, Marvel and Moodys
(1995) study of the impact of enhanced sentencing for gun-related
crimes found limited evidence of deterrence in a few states. The effects
were so sporadic that the policy of expanded prison terms as a deterrent
for gun crime could not be supported from the results.
In their review of recent literature on sentence severity and crime,
Doob and Webster (2003) concluded that sentence length had no appreciable effect on crime. Moreover, the signicant disagreement in this
literature was sufcient to lead them to argue that the null hypothesis
that sentence severity has no impact on crimeshould now be accepted.
This conclusion is supported by Lippkes (2002) review of the impact of
prison sentence length on crime.
Several studies of more specic issues also refute the idea that imprisonment reduces crime. In a study of three-strike legislation in twenty-four
12
B i g P ri s on s, B i g D ream s
states, Austin et al. (1999) concluded that, with the exception of California, tougher sentencing practices have had little impact on crime.
Weisberg, Waring, and Chayet (1995) examined the impact of specic
deterrence on white-collar offenders. The signicance of this research is
the assumption that white-collar offenses serve as an excellent example
of behaviors that are rational, thus tting the expectations of deterrence
theory. In other words, among a variety of offenders, white-collar offenders would be assumed to be more amenable to the effects of deterrence. Their study compared reoffending among a matched sample of
white-collar offenders sentenced to prison versus those who received
other sanctions. No difference was found between these two groups with
respect to recidivism.
Perhaps one of the most important recent studies of the relationship
between imprisonment and crime was published by Clear, Rose, and Waring in 2003. This study provides a partial test of Rose and Clears (1998)
earlier theory that high incarceration rates have negative effects on local
communities by undermining informal social control and other community networks that prevent offending. Thus, contrary to deterrence or
incapacitation arguments, these researchers suggest that incarceration destabilizes communities and enhances local social disorganization, which
in turns expands the problem of crime. Their study conrmed these assumptions. They discovered that prison admissions had a small impact on
crime in communities with a low level of crime. In communities where
a larger proportion of residents are removed via incarceration, they found
that crime increased.
It is also useful to consider the comparison made by Ouimet (2002)
of crime and penal trends in the United States and Canada during the
1990s. Ouimet notes that both nations experienced a signicant decline
in crime during this period. In contrast to the United States, however, in
Canada there was no enhanced policing and no expansion of sentence
lengths or in the use of imprisonment. As a result, Ouimet concludes that
increased punitive responses cannot be the cause of the crime reduction
in the United States.
In sum, the empirical literature examining the impact of imprisonment on crime leads to no clear conclusion concerning the relationship
between these two social events. From a scientic vantage point, the evidence is not compelling and the degree of uncertainty is sufcient at this
Introduction
13
14
B i g P ri s on s, B i g D ream s
to be squeezing this issue into the present work, and, in my view, would
appear to minimize the importance of Clears argument.
No full-blown effort to describe the trends in imprisonment in
America from an historical perspective would be adequate if it failed
to address the goal or purpose behind the use of prisons. Over the past
two decades, the public conversation that has been carried out on the
topic of prison and punishment in America (and by public conversation
I mean the messages the news media reports, and those of the spokespersons it has allowed to occupy broadcast time and craft the content of
news) has been captured by a conservative position that supports our big
prison system as necessary to crime reduction. Theoretically, these positions are consistent with incapacitation and deterrence strategies. The
public conversation has omitted a useful discussion of the rehabilitative
goal and potential of the prison. We seem to have forgotten that the
idea behind the founding of prisons in the United States was the rehabilitation and reform of inmates. Indeed, the portion of utilitarian philosophy devoted to the idea of reform and rehabilitation informed the
development of Americas prison system for more than a century after
its founding. In the 1960s and 1970s, rehabilitation made a short-lived
return, though it failed largely because of inadequate implementation
and faulty scal investment strategies. In hindsight, the required nancial
investment needed to operate a rehabilitative prison system is rather insignicant in comparison to the investment we have made in our current
prison system infrastructure based on the bigger is better approach. A
number of scholars, however, have continued to promote rehabilitation
as an appropriate prison goal. In addition to scholars such as Todd Clear,
one of the most prominent scholars in this area in the contemporary
period is Francis Cullen. To some extent, it is impossible to understand
the growth of Americas prison system over the past thirty years without
having some knowledge of rehabilitative strategies, how they have been
used, and why they failed. It was this failure that helped stimulate the
growth of our present prison system, fueling its focus on deterrence, incapacitation, and an exaggerated form of retribution. Thus, the work of
scholars including Cullen is a necessary supplement to the present book,
especially if the goal of such a work is to suggest some alternative to the
current lock em up approach. It is not my intention, however, to offer
an alternative to the current system, and I leave that task to those who
Introduction
15
are better informed about such issues (Austin, Irwin, Clear, and Cullen,
to name but a few).
The view of the prison as industry developed in the present book
was inuenced by Nils Christies Crime Control as Industry: Towards Gulags,Western Style (1994). Though Christies view is summarized later, one
point that he made in his book which merits discussion here is captured
in the following quotation: The major dangers of crime in modern societies is not the crimes, but that the ght against them may lead society
towards totalitarian developments (16). The problem, from my perspective, is that the issue Christie raisesthat American societys punishment
response is moving us toward a totalitarian regimeis sufciently signicant and real enough to require an independent analysis. Doing so would
require subjecting various philosophies of punishment to in-depth analysis and critique. But the analysis could not stop there, and would need
to explore the t between political preferences for democracy and the
implementation of various penal strategies. The question such an analysis
would be required to address would be, How does each penal strategy
t with democratic principles? More importantly, we must ask questions
that policy makers have not asked when selecting penal responses: How
would the implementation of a specic penal philosophy be modied
by the assumptions of a democratic form of governance? How do we
balance our emphasis on freedom and democracy with the goals and
outcomes of our penal process? To make a long story short, this balancing effort presents a real problem in a democracy. How do we determine
which principlespenal or democraticare to be privileged? From a
purely philosophic position, the short answer in the United States isor
rather should bethat principles of democracy ought to outweigh penal
philosophies. That is to say, democratic principles should be employed to
temper penal reactions whenever penal approaches become overly oppressive. Clearly, such a dialogue has not occupied much time or space
in the American prison debate. And, as a result, as Christie suggests, we
are coming very close to establishing the grounds for totalitarian rather
than democratic penal responses. This tendency presents a greater threat
to America than terrorism, though few Americans would be willing to
make or support such an argument.
In my view, it is also impossible to think about our current state of
punishment without reecting on the importance of punishment and
16
B i g P ri s on s, B i g D ream s
Introduction
17
us down the wrong path as far as crime control is concerned. To see this
dilemma, we must rst understand the nature of the order upon which
society is built. This is no simple task, and to do so requires a great deal
of reection. In short, when we reect on the issue of crime and punishment, we are again led to the conclusion that the relationship between
the two is not clear-cut or straightforward, and that no policy that takes
the relationship to be simple can offer an appropriate solution. Elements
of this argument will become more apparent as the evidence in this book
is presented and examined.
Finally, we must consider Graeme R. Newmans (1985) history of
punishment, from which we can extract the idea that it is the orienting
penal philosophy of society (or those charged with directing a society)
that affects the direction of its system of punishment or its response to
crime. The choice a society makes with respect to how it will employ
its penal system is, as Newman illustrates, connected to historical forces
that affect how different philosophies of punishment are interpreted and
put into action.
S um mary of What Fol lows
The goal of this book is to review the rapid growth of the U.S.
prison system that has been underway and unimpeded since 1973. Later
in this book, data for the U.S. prison system complied from the fty state
systems will analyzed to examine whether that system suppresses crime.
These data will be be analyzed over time and across regions to make
specic points about the growth of prisons. The growth of imprisonment
will also be assessed by comparing growth rates of the prison system to
the growth rate of crime. It will also be relevant to this discussion to look
at the predicted costs of our nations continued push to expand the use
of imprisonment as a response to crime. In addition, this book will examine who is sent to prison. The common assumption is that our growing prison system has and is being used to lock up the most dangerous of
societys offenders. As the data in this book reveal, this assumption is not
very accurate. An unfortunate dimension of the American prison system
has always been its use as part of the war on the lower classes, which in
the modern era has taken the guise of a war on primarily lower-class
drug users and their drugs of preference. Prisons, in other words, are
most likely to hold the poor, and as Ronald Goldfarb argued in the
18
B i g P ri s on s, B i g D ream s
late 1960s, prisons are Americas poor houses. Since the civil rights era,
imprisonment has also come to be part of the race war, and the number
of minorities held in prisons has escalated at a much faster rate than the
incarceration of the non-Hispanic, white population.
I will endeavor to keep these arguments simple, avoiding complex
mathematical models that are difcult for the ordinary person to comprehend. Fancy mathematics wont make the argument any more persuasive; simple models, where needed, will be persuasive enough in my
opinion. This is not to suggest that there wont be any data or models to
examine; to be sure, the story I am about to tell is one found in largely
in numbers.
The effort to comprehend what is happening with the modern-day
system of imprisonment is, in my view, one that needs to be told. It needs
to be told so that American people can make informed decisions about
the direction their society is or should be taking; so that Americans can
have the kind of information they need to participate in our democratic
form of government.
The story of the modern prison that will be told here cannot be
told with numbers alone. In order to interpret data, the person doing the
interpreting must employ some frame of reference or theoretical background in order to make sense of what the numbers say. These assumptions appear throughout this book, but are reviewed more extensively in
chapter 5. The view I present in chapter 5 is not the only one that leads
to the conclusions that I draw from the data. And, for those who really
dont care about the perspective I employed to reach these conclusions,
this material could be easily skipped without compromising the rest of
my argument. As an academic, however, I feel compelled to provide readers with these background theoretical issues.
The theoretical perspective that informs my work is grounded in
materialism. In simple terms, materialism is the idea that the social world
around us must be interpreted relative to the way in which a society is
designed. That means connecting social events, such as imprisonment
or the growth of the prison system (or any other topic), to material
conditions of society, such as its economic and political systems, its class
structure, and other important hierarchies that tell us about a society,
such as its race relationships. Thus, to understand why the U.S. prison
system has grown so dramatically over the past three decades, we need
Introduction
19
Chapte r 2
This chapter provides an overview of numerous issues that need to be considered in an examination of prison systems.
These issues include exploring how philosophies, policies, politics, and
economic factors drive the growth of prison systems, and whether there
is a relationship between the growth of prisons and a reduction in crime.
Before we can begin to examine whether a bigger prison system is better, we have to dene what we mean when we ask whether one type of
prison system is better than another.
For the purposes of this book, a better prison system is one that has
a crime suppression effect, or one that reduces the level of crime in society. After all, isnt this why societies build and employ prison systems: to
reduce crime? To be sure, crime reduction, whether through rehabilitation, deterrence, or incapacitation, has always provided the philosophical
underpinnings for imprisonment (except if one adopts a pure retributive perspective, Newman, 1985). Indeed, if all society wanted to achieve
was the simple punishment of criminal offenders, there would be better
options or at least other alternatives to consider (Newman, 1985). In
any event, the rst criterion of a better prison system is that it should
reduce crime. Thus, we can ask two questions about Americas prison
system. First, does the big prison system currently operating in America
fulll this goal? And second, has making the prison system bigger led to
continual reductions in crime? We could add additional questions here,
perhaps addressing the marginal gain in crime reduction that would occur for every 100 persons sent to prison. At this point, any additional
questions can wait since asking and answering them is contingent on
establishing the answer to the rst two questions.
In addition to reducing crime, being better also implies that bigger
prisons should do their task more efciently than some other alternative.
20
21
22
B i g P ri s on s, B i g D ream s
to the late eighteenth century, while uniform crime data are available
from the 1930s on. Likewise, the data needed to answer this question from
a cross-cultural perspective are even more problematic in terms of availability. One need only check the data maintained by the United Nations
Criminal Justice Network (UNCJIN) to establish this fact. Even if all the
cross-cultural prison and crime data we desired were available, undertaking a study comparing the impact of imprisonment on crime would be a
tremendously difcult task given the many factors that differ across nations
that would need to be addressed because they might account for the presence or absence of a crime suppression effect of imprisonment or other
crime control strategies. Researchers have identied numerous factors that
affect the causes of crime, from personality constructs, to biological predispositions, to environmental pollutants, to family and friendship patterns,
to economic, social, and political conditions that include unemployment,
general economic factors such as recessions and stages of economic development, and economic inequality, to name but a few (Vold, Bernard, and
Snipes, 2002). Research on the causes of crime has not, to date, produced
denitive answers about the causes of crime, and thus testing the vast number of opposing viewpoints would be a daunting task.
Therefore it is important to limit the scope of this investigation into
the effectiveness of prison to one location. And, while it is useful to limit
the scope of the investigation to make it manageable, it will still be necessary to use data from other nations as the basis for making comparisons
that illustrate the relative size the American correctional system. Despite
the need for such comparisons, it is not the intent of this work to explain
the factors that affect the size of prison systems across the various nations
used for the purpose of comparison.
Narrowi ng the S cope of th e Inve sti gati on
Given the large number of potential variables, and the extraordinary
degree of variation in factors that might impact the relationship between
crime and punishment across societies, history, and even individuals, it is
necessary to simplify the task. Here, three strategies were employed to
simplify the analysis of whether a big prison system reduced crime.
First, this analysis will not examine the impact of imprisonment on
individuals. Imprisonment may affect individuals differently depending
on a number of factors that vary across individuals, on how those factors
23
interact, and on the strength of each factor (Agnew, 2005). For example,
how a person was raised and socialized, an individuals personality traits,
the strength of an individuals familial and community ties, her level of
education, and so on, may all inuence the effect of punishment on a
particular individual. In addition to difculties associated with accurately
identifying and measuring these individual traits, it would be necessary to
describe how these traits interact, and how the strength of different traits
might affect the outcome. This would, at best, become cumbersome, and
most likely would produce rather weak statistical models predicting the
effect of imprisonment on an individuals potential future behavior. Such a
study would also require enormous nancial resources to accomplish, and
a signicant investment of time. Moreover, a big prison system might affect
individuals differently than a smaller prison system, or some alternative to
prison. Ideally, we would want to follow a sample of people over time as
a prison system expanded. But it is now thirty years since the U.S. prison
system began expanding, and we cant go back in time to construct such a
study. The best we could do would be to examine the impact of different
size prison systems on offenders using data from different states.
It is not only difcult to construct a study that could address all the
potential sources of variation that may inuence how imprisonment affects an individual, it is, more importantly, not necessary to know the
answer. Rather, if a big prison system suppresses crime, evidence of a
suppression effect ought to be seen in the aggregate level of crime in
society; that is, in the summing up of all crimes committed by people
in a society over a period of time. If imprisonment reduces crime, we
can see this impact without needing to measure or know its impact on
specic individuals. If we omit individual measurement as necessary, we
can create a simple hypothesis stating that as prisons get bigger and more
people are incarcerated, crime should decline in a society. Thus, it makes sense to
determine if the level of imprisonment within a nation has a signicant
impact on that societys level of crime independent of the question of
how imprisonment might affect specic individuals.
Second, the problem of controlling for a multitude of inuences can
be simplied by examining the aggregate relationship between crime and
imprisonment in one historical period. By examining one time period,
a number of historically contingent factors can be ignored because we
eliminate their effects by restricting the scope of the study. Nevertheless,
24
B i g P ri s on s, B i g D ream s
this strategy cannot remove the impact of all the factors that might inuence the studys validity and generalizability. In many of the analyses that
follow, a host of factors that may affect crime (e.g., employment rates, opportunity structures, etc.) are purposefully omitted, which creates a validity problem that must be acknowledged.Yet, if the level of punishment is
the most important factor affecting the level of crime, it is not necessary
to identify or control for alternative explanations since we are not interested in their effect; we are only attempting to establish whether the basic
assertion that a big prison system controls crime could be true.
Third, the task of studying the relationship between crime and imprisonment can be simplied by restricting the study to one nation. To
accomplish this goal, the United States was selected as the focus of this
study. There are several factors that inuenced this decision.
First, as an American researcher, I am more familiar with the U.S.
prison system than with the prison system in any other nation. Second, as
a citizen of that nation, I am concerned with the problems of crime and
punishment that affect me most directly. Third, America presents a unique
condition useful for testing the relationship between expanding levels of
imprisonment and its impact on the level of crime. That condition stems
from the fact that the level of imprisonment in the United States has expanded
each and every year from 1973 through to the present. In theory, then, we have
nearly perfect conditions to determine whether a consistent increase in
imprisonment has an equally consistent impact on reducing crime. Indeed,
these conditions are quasi-experimental to the extent that one condition of the investigationthe U.S. imprisonment ratemoves or trends
only upward. As a result, if an expanding imprisonment rate acts as expected, then the crime rate should always trend downward. Finally, America
has the worlds largest prison system. If having a big prison system reduces
crime, then this effect ought to be most evident in the United States.
It was noted that the United States has the worlds biggest system of
incarceration. But how big is the American prison system? It is useful at
this point to introduce some data that speak to this issue.
H ow B i g I s A m e ri ca s P ri s on Syste m ?
In order to comprehend the size of Americas prison system, we need
something against which to compare it. We could use historical data for
this purpose, and show the growth of Americas prison system over time.
25
(2)
Number of
Inmates
2,033,331
874,300
55,156
158,858
176,893
10,630
34,000
7,256
163,256
83,113
47,406
33,098
54,288
16,643
248,989
114,765
Country
United States
Russian Federation
Belarus
Ukraine
South Africa
Panama
Swaziland
Mongolia
Iran
Poland
Romania
Chile
Morocco
Czech Republic
Brazil
Mexico
(1)
701
611
554
413
400
367
359
279
226
215
212
204
191
162
161
156
(3)
Imprisonment
Rate**
1,709
1,010
37
172
241
73
12
28
184
156
43
145
47
34
922
44
(4)
Number
Institutions
Table 2.1
1,190
866
1,491
924
734
146
2,833
259
887
533
1,103
228
1,155
490
270
2,608
(5)
Avg. Size
Institution
2003
2002
2003
2003
2002
2002
2002
2002
2002
2003
2002
2002
2000
2003
2003
2000
(6)
Data
Year
(continued)
1.00
1.15
1.26
1.70
1.72
1.91
1.95
2.51
3.10
3.26
3.30
3.44
3.67
4.33
4.35
4.50
(7)
U.S.
Multiplier***
Lesotha
England/Wales#
Spain#
Columbia
China
Canada#
Australia#
Kenya
38,604
56,574
81,176
7,745
55,382
14,968
64,173
3,378
8,500
3,000
73,040
54,341
54,034
1,512,914
36,024
21,819
35,278
Country
Argentina#
Italy#
Germany#
Austria#
France#
Netherlands#
Turkey
Ireland#
Greece
(2)
Number of
Inmates
(1)
107
100
98
95
93
93
90
86
80
143
139
133
126
117
116
112
111
(3)
Imprisonment
Rate**
116
205
222
29
185
79
562
17
25
14
138
85
168
698
211
98
92
(4)
Number
Institutions
333
276
366
267
299
189
114
199
340
214
529
639
321
2,168
171
237
383
(5)
Avg. Size
Institution
2001
2002
2003
2003
2003
2001
2003
2002
2002
2002
2003
2003
2001
2002
2003
2002
2003
(6)
Data
Year
(continued)
6.55
7.01
7.15
7.38
7.54
7.54
7.79
8.15
8.76
4.90
5.04
5.27
5.56
5.99
6.04
6.26
6.32
(7)
U.S.
Multiplier***
19,554
6,506
3,617
4,987
3,150
2,666
67,255
39,368
304,893
Country
Venezuela#
Sweden#
Finland#
Switzerland#
Denmark#
Norway#
Japan#
Nigeria
India
180.2
76
73
70
68
59
59
53
33
29
(3)
Imprisonment
Rate**
228.6
29
83
28
167
56
43
190
147
1,058
(4)
Number
Institutions
588.4
674
78
129
30
56
62
354
268
288
(5)
Avg. Size
Institution
2003
2003
2002
2003
2001
2001
2002
2002
2002
(6)
Data
Year
9.22
9.60
10.01
10.31
11.88
11.88
13.23
21.24
24.17
(7)
U.S.
Multiplier***
# The nations marked with a # are more comparable to the United States from an economic, social, and cultural perspective.
*** The U.S. multiplier compares the U.S. rate of imprisonment to the rate of imprisonment in each nation. It is derived by dividing the U.S. imprisonment rate by a specic countrys imprisonment rate. This number tells you how many times larger the U.S. rate of imprisonment is compared
to a specic country.
** The rate of imprisonment is a standardized measure that allows for direct comparison of the size of a prison system. The imprisonment rate is
per 100,000 population within a country, and is calculated as follows: (prison population/country population) 100,000.
* Data for this table were extracted from the International Centre for Prison Studies website (http://www.prisonstudies.org). These data include both
prison and jail populations. These data (see column 6) represent the most recent year for which data were available at the time this table was created.
Mean/Avg.
(2)
Number of
Inmates
(1)
29
30
B i g P ri s on s, B i g D ream s
different nations. Instead of comparing the United States to the overall average, let us instead compare its rate of imprisonment to a more
equivalent sample of nations: modernized (and democratic) nations.
The sample of nations in table 2.1 contains a subsample of 18 modernized nations (the nations marked with a # in column 1) that are more
directly comparable to the United States economically, socially, and politically. Compared to this subsample, the U.S. rate of imprisonment is
nearly 8 times (7.8) higher. This gure indicates two facts. First, the U.S.
rate of imprisonment is not similar to the rates of imprisonment found
in a sample of similar nations. Second, it should be clear that the U.S.
rate of imprisonment is more similar to the rate of imprisonment found in
less developed nations. These facts indicate that the reliance on incarceration found in the United States is at odds with the use of imprisonment
found in similar nations of the world. Further, if we take the rate of imprisonment as an indication of a dimension of social repression, then the
United States is well ahead of its peers.
The Number and Size of Prisons
The size of a prison system could also be represented by comparing
the number of prisons within nations. These data, displayed in column 4,
indicate that the United States also has substantially more prisons than
any other nation. For example, the United States has 69 percent more
prisons than the nation with the next largest number of prisons, the
Russian Federation.
The number of prisons within a nation may not be the best measure of the size of its prison system. Again, the number of prisons within
a nation will have some relationship to how many people are incarcerated and perhaps to the number of people living within that nation.
To take these differences into account, we could create a standardized
measure of this indicator as well. Instead, let us examine an equally useful, alternative indicator of the size of a prison system that might also
contain some clues about how prisons are used within different nations,
and the philosophies that guide the use of prisons across these nations.
That indicator is the average size of a prison within each nation. Data
on the average size of a prison for each sampled nation is found in column 5. These data show that the average prison in the United States
housed nearly 1,200 inmates. On this indicator, the United States ranks
31
fth, behind Swaziland, Mexico, China, and Belarus, nations that do not
compare favorably to the United States in terms of economic development, living standards, or even in terms of political systems. Overall, the
average prison in the overall sample holds approximately 588 inmates,
meaning that the average U.S. prison is more than twice as large as the
average prison found in other nations. We might also want to ask how
the U.S. average prison size compared to the size of prisons in the subsample of modernized nations. This latter comparison group has an
average prison size around 300, indicating that prisons in the United
States are typically 4 times larger than prisons found in nations with
equivalent social, economic, and political conditions.
Calculating the U.S. Multiplier
Finally, column 7 presents a statistic I call the U.S. multiplier. The
U.S. multiplier is derived by dividing a countrys rate of imprisonment
by the U.S. rate of imprisonment. A country that had an imprisonment
rate equal to the United States would have a multiplier of 1.00. Countries with larger rates of imprisonment would have multipliers smaller
than 1.00, while countries with smaller prison systems will have multipliers greater than 1.00. Since the United States has the highest rate of
imprisonment, this multiplier indicates how many times larger the U.S.
rate of imprisonment is than the rate of imprisonment in each individual
nation. Thus, for example, if the U.S. multiplier is 5.00, then the rate
of imprisonment in the United States is 5 times higher than the rate
of imprisonment in the particular country being examined. (Note: the
U.S. multiplier can be converted into a percentage by multiplying it by
100.) As can be seen in column 7, the U.S. multipliers range is between
1.15 and 24.17. Thus, some countries come close to the United States
in terms of prison size. The vast majority, however, have much smaller
prison systems.
The data in table 2.1 should cause us to pause and ask, Why is the
U.S. use of imprisonment so much higher than in the rest of the world?
Much more data is needed to address this question.What we know at this
point is that the U.S. prison system is big: the biggest in the world. If you
believe in ideas like deterrence or incapacitation, then the relationships
noted in table 2.1 might cause you to assume that the rate of criminal
offending in the United States must, because of all this punishment, be
32
B i g P ri s on s, B i g D ream s
much lower than the rate of crime found in other nations. Without going into great detail and producing all the crime data for each nation, let
me simply indicate at this point that this assumption is not correct (for
more extensive discussion, see Marc Mauers 2003 paper Comparative
International Rates of Incarceration). This is an important assumption
to expose because Americans commonly assume that having a big prison
system accomplishes the goal of reducing crime. This crime reduction
assumption is shared by many policy makers who occupy decision-making positions through which they are able to inuence the goals and
growth of the American prison system.
Having reviewed some data that provides a picture of how large
Americas prison system is, we can begin to examine the assumption that
increasing the rate of incarceration should cause a reduction in crime.
Com mon A ssum p ti on s M ade
A bout P uni shm e nt and C ri m e
In this section, some of the more common assumptions that link
punishment and crime are examined. Particular attention is paid to assumptions linking the growth and use of imprisonment to reductions in
crime through deterrence or incapacitation.
Why Should Locking Up More People Reduce Crime?
Longstanding theories of behavior assert that punishment can change
behavior. Since the late eighteenth and early nineteenth century, these
theories have occupied a central place in criminological thinking. In the
early nineteenth century, Jeremy Bentham asserted that increasing the
costs of crime could deter criminals. Punishment is one such cost. This
idea has popular appeal, and many Americans believe that a bigger prison
system that locks up more offenders reduces crime. In theoretical terms,
the crime reduction that occurs through imprisonment involves either
incapacitation or deterrence. Technically, incapacitation and deterrence
are two different effects that can result from imprisonment, though they
may overlap and be difcult to distinguish.
Incapacitation is the term used to describe the effect of preventing offenders (usually repeat offenders since they are often the targeted
group) from committing crimes while incarcerated. That is, being locked
up prevents the offender from preying upon people in the free world.
33
34
B i g P ri s on s, B i g D ream s
35
deter? Will incapacitation and deterrence effects emerge regardless of social conditions? Or are deterrent and incapacitative effects conditioned
by social and economic conditions? And, if the effect appears to be tied
to social or economic conditions, can we be sure that the observed effect
can be attributed to the deterrent or incapacitative effects of incarceration instead of social and economic trends?
The problem of predicting the effect of locking up more inmates
is complex, and would require an elaborate statistical model. The results
derived from such a model would also vary depending on the data that
were used for calculating the outcome, and the assumptions about behavior those statistical models included. For example, a simple model
with only a few variables may show that incarceration has a large effect on the rate of criminal offending. It is quite possible, however, that
in a more complex model that adds the effect of social and economic
factors, the effect of incarceration on the rate of crime might become
attenuated. Further, the more of these other variables that get added
to the model, the smaller the effect of incarceration on crime might
become. If this were indeed the case, then we would be able to show
that there are social and economic factors that override the effects of
incarceration on crime rates.
Variation in Effects over Time and Place. We also need to consider that research results that provide evidence of a deterrent or incapacitative effect may produce this result because they examine a very
limited time frame, or data from specic locations. For instance, research
studies that have employed data from the 1990s can demonstrate a small
deterrent or incapacitative effect on crime (see chapter 1). In recent
years, these kinds of studies have been taken as valid evidence that deterrence and incapacitation work. Likewise, data from some states, but
not from others, will show a deterrent or incapacitative effect. Longer
trend data, however, as this book will demonstrate, does not show the
same deterrent or incapacitative effects of imprisonment. Over the long
run, or since 1973, the large prison system America has built has not
produced consistent deterrent or incapacitative effects. At best, the results are mixed. These mixed results are not often acknowledged. In fact,
the assumption that imprisoning offenders reduces crime is so widely
accepted that it is rarely challenged. Likewise, if asked to come up with
36
B i g P ri s on s, B i g D ream s
suggestions for reducing crime, many members of the general population would respond that more police, longer prison sentences, and sending more people to prison would do the trick. What most people fail
to understand is that punishment, or lack thereof, isnt necessarily the
most important determinant of crime, and that the effects of incarceration may not be uniform or have similar effects across time, places, or
people. In other words, the fact that imprisonment effects vary and are
inconsistent is not considered by deterrence research.
Take politicians as an example. You never hear a politician say: It
is my intention to push for an expansion of our prison system because
there is a 50 percent chance that doing so will reduce crime. That also
means that there is a 50 percent chance my plan wont work. Rather,
what politicians often do is suggest that they are in favor of raising the
rate of incarceration because it will lower crime. What they are offering the
public is what they believe the public wants to hearthat putting more
people behind bars will make the public safer. Likewise, you never hear
a politician who is attempting to gain your vote by offering his (and its
usually a man) views on how to reduce crime tell you exactly how much
crime will be reduced by following his plan. Has anyone specically said:
I am going to send 20 percent more people to prison because this will
reduce the crime rate by 30 percent? No, because, as I will demonstrate,
there are no clear-cut answers such as this when we attempt to rely on the
prison as a cure for crime. As a criminologist, I can tell you that we cant
predict how much raising the rate of imprisonment will lower crime.
Why? Well, we dont really know why exactly. What we do know is that
the relationship between crime and punishment doesnt work the way we
expect it to work; crime isnt simply a response to how much or how little
punishment there is in a society. In fact, as research indicates, fear of punishment is only a minor source of compliance with the law (Tyler, 1990).
Crime is the result of the interaction of numerous processes. The majority of factors that can be identied as causes of crimepoor economic
conditions, deteriorated neighborhood conditionswill be unaffected by
how much punishment society applies. That is, raising the rate of incarceration is unlikely to lower crime if crime is caused by factors that incarceration does not or can not address. Punishment doesnt reduce poverty, create
jobs, or eliminate the problems of impoverished neighborhoods where we
see the kinds of crimes that punishment is supposed to address.
37
ity to predict what will happen when we raise the rate of imprisonment,
consider what actually happened to crime in the United States between
1973 and 2002, the thirty years when imprisonment grew every year.
During this period, crime increased and decreased almost an equivalent
number of times as the imprisonment rate rose. This simple observation,
that crime can and has risen even while rates of incarceration are on the
rise, leads to the conclusion that either outcomea crime reduction or
an increase in crimewas equally likely. In fact, as will be demonstrated
in a later chapter, the relationship between increased rates of incarceration and crime resembles a coin toss: there are two outcomes, and neither is statistically more likely. We can see a similar pattern when we look
across states or regions within the United States.
Why cant we predict the outcome when it seems so logical that
crime should decline when we incarcerate more criminals? Consider
again the simple idea of incapacitation: the more criminals we put behind bars, the fewer crimes they can commit. If this is true, then big
prisons are an easy solution to the crime problem to the extent that all
we need to do is lock up all the criminals. The appeal of such a straightforward and simple solution to crime is extremely compelling. Following
the logic of the incapacitation strategy, the increase in incarceration that
occurred in the United States since the early 1970s should have made
crime drop like a rock falling from the Empire State Building, since we
now incarcerate over 1 million more people than we did in the early
1970s. To illustrate this point, let us consider a ctional example.
Say that in a given year, the state I live in locked up 10,000 criminals
who had committed an average of 5 crimes apiece. It seems logical that
I could expect the number of crimes to be reduced in the next year by
50,000 (5 * 10,000) as long as I dont release anyone from prison, since
I have incapacitated so many repeat offenders. But incapacitation never
works this way, or this well, because there are two primary aws in this
approach to crime control.
First, we dont lock up criminals for life, meaning the suppression
effect can only last as long as a prison sentence. So, why not lock up
criminals and throw away the key? The answer is simple. First, not all the
crimes that are committed in a society are severe enough to warrant such
a weighty penalty. Second, doing so would be nancially prohibitive and
38
B i g P ri s on s, B i g D ream s
create an even bigger prison system than we have now. In fact, if society
were to carry out such a program, it would soon nd that more Americans would be behind bars than would be free. We probably cant lock
up all the criminals because, as self-report studies indicate, the majority
of people in society have committed a crime at one time or another,
and we would end up with 100 million people or more behind bars.
Thus, we have to limit our lock em up aspirations, and concentrate
on the serious, repeat offender (as you will nd out later, our big prison
system holds its fair share of people who do not t this category, but,
rather, holds many more rst-time offenders than might be expected).
As a result, the incapacitative effects of an imprisonment strategy can
only go so far.
Second, this strategy doesnt work in the long run because when
society incarcerates some offenders, others come along to take their
place. In other words, there appears to be a replacement effect. Why?
Because incapacitation does not address the social and economic conditions that inuence people to commit crime in the rst place. If, for
instance, criminals turn to crime because they cannot nd meaningful work in their community, and incapacitation does not enhance the
prospects for employment within the community (in fact, to reduce the
unemployment rate signicantly, imprisonment rates would need to be
several times higher than they are now), then how does incapacitation
address factors that generate criminal populations? You could select any
potential cause of crime you like and substitute it into this example:
poor parenting; lack of strong nuclear families; poor social bonding;
high dropout rates and poor school performance, and so on. None of
these problems are addressed by locking up more people in prison. Or,
we might want to think about the problem this way. For the purposes of
this example, assume any cause for drug use you prefer, and assume that
policy makers attempt to reduce drug use through interdiction strategies and aggressive policing tactics that target drug sellers. When society
attempts to lower the rate of drug use crimes by locking up more drug
trafckers, we can expect that someone will take the trafckers place,
especially if the demand for drugs remains high and nothing has been
done to combat the factors that caused the users behavior. Likewise interdiction doesnt reduce the demand for drugs, since interdiction does not
address the causes of drug use. The point: punishment does not alter the
39
general behaviors associated with the use of drugs, or the illegal sale of
drugs, because the punishment does not change the conditions associated with either behavior.
How Much Should Crime Decline? Think about this: according to
the International Centre for Prison Studies, in 2003 the U.S. system of
incarceration became the rst modern penal system to incarcerate over
2 million people. In contrast, in the early 1970s, the U.S. prison system
incarcerated less than 200,000 people. It would appear obvious that if the
incapacitation approach were the answer to the problem of crime, and
we now incarcerate ten times as many people as we once did, the level
of crime should be very low today compared to the early 1970s. Exactly
how much lower should the crime rate be today compared to the early
1970s as a result of locking up so many more people might be hard to
pinpoint. Theoretical positions would lead us to believe that the reduction should be quite substantial, especially given the size of the increase
in imprisonment that has occurred in the United States over the past
thirty years. Typically, however, we can only know the answer to this
question by looking at empirical data that compares trends in crime and
punishment. So, what do these empirical data have to say?
Here, for example, is what we do know from police arrest data: in
1971 there were 897.1 crimes per 100,000 people in the United States,
while the rate of incarceration was 95 per 100,000. In 1981 there were
1,070 crimes per 100,000 American citizensa 16 percent increase
while the incarceration rate had increased by about 62 percent to 154 per
100,000. By 1991 the crime rate had increased to 1,198.8 per 100,000
people, which amounted to a 12 increase since 1981, and a 33.6 percent
increase since 1971. At the same time, the imprisonment rate rose to 313
per 100,000, constituting a 103 percent increase since 1981, and a 229
percent increase over 1971 levels (incarceration data appear in table 3.1;
for arrest data, see table 4.2004 in the Sourcebook of Criminal Justice
Statistics online). Thus, crime was rising along with substantial increases
in the use of imprisonment. Let us fast-forward to 2001, when the arrest
index was 807.3 per 100,000 people, or 10 percent lower than the 1971
rate where we began this example. The imprisonment rate, however,
stood at 498 per 100,000, an increase of 424 percent since 1971. In other
words, over these three decades, multiplying the imprisonment rate by
40
B i g P ri s on s, B i g D ream s
more than 4 times reduced crime by only 10 percent. The biggest crime
reduction over this period, which were especially evident in the 1990s,
was a reduction in property crimes, which decreased from 721.4 per
100,000 in 1971 to 581.8 per 100,000 by 2001. Over the same time period, there were more violent crimes28 percent more, in fact. So, prisons, which are supposed to protect us from the worst offenders through
incapacitation or deterrence, did not seem to work on the population
of the worst offenders. Overall, it took a large increase in imprisonment
over this time spanover 400 percentto lower the aggregate crime
rate by just 10 percent, which comprised two opposing trends: a reduction in property crime and a rise in the rate of violent crime.
In sum, increasing the rate of incarceration (by more than 400 percent) between 1971 and 2001 eventually appeared to reduce crime by 10
percent from its 1971 level. At the same time, we must consider that the
crimes most affected were property crimes, not violent crimes. These
simple facts indicate that increasing the number of people we incarcerate is not the answer to the problem of crime: very large increases in
imprisonment over an extended period seemed to lower the crime rate
slightly, assuming no other conditions that may cause crime had changed.
These simple facts also seemed to indicate that we have been mislead
by following this course of actiona course of action that claims that a
rising rate of imprisonment signicantly reduced crime (again, perhaps
because the causes of crime seem to lie somewhere outside the crimepunishment nexus).
What I have left out of the discussion above is the question of
whether prisons deter potential criminals. All we need to do is substitute
the word deterrence into the few paragraphs above to get our answer.
A big prison system ought to, in theory, deter a lot more people than a
small prison system because the threat of punishment has increased along
with the size of the prison system. If we take the rate of incarceration as a
measure of the odds of imprisonment, we can see that the increased odds
of incarceration that accompany a big prison system had little impact on
crime. The increase in the imprisonment rate from 1971 through 2001
of 424 percent (that is, an increase in the odds of being incarcerated for
the U.S. population) is related to only a 10 percent reduction in crime.
Looked at another way, we could say that: (1) locking up 403 more people out of every 100,000 people reduced crime by 90 per 100,000; or
41
42
B i g P ri s on s, B i g D ream s
43
44
B i g P ri s on s, B i g D ream s
45
46
B i g P ri s on s, B i g D ream s
rise because we have put more people in prison and someday they will
get outwe may begin to see that the crime rate also rises as a larger
number of ex-inmates fail to be reintegrated into society.
The big spaces big prisons take up become problematic for other
reasons that are not often considered. As the number of prison facilities
grow, nding suitable locations for these facilities becomes more difcult. Responding to environmental issues, there is now greater emphasis
on protecting the remaining natural areas in the United States. This fact,
coupled with the expansion of the population and its broader distribution
over the American landscape makes it harder and harder to nd isolated
locations to build large prisons. As we continue to expand the number of
prisons, we can expect to encroach on either wilderness areas or the space
taken up by existing communities. In time, the not in my backyard
public attitude that is commonly voiced concerning the location of other
noxious facilities, such as landlls, waste sites, and facilities that treat, store,
and dispose or produce hazardous materials, will be imported into the
correctional system debate. To be sure, the middle class, with its heightened fear of crime, and the group most often associated with supporting
get tough on crime approaches, may continue to call for prisons, while
simultaneously arguing that these facilities not be placed in locations that
violate their sensibilities, or lower their property values.
Fue l l i ng the Syste m
Policy makers who have helped build our big prison system have
been shortsighted in a number of ways, some of which have been discussed above. It is plausible that at least some policy makers have considered these issues. But there is a large problem looming on the horizon
that policy makers have probably not considered, which we will later
examine in greater detailthe impact of the end of the fossil fuel era on
our nations ability to continue to operate its massive prison system.
For those unfamiliar with the idea that the fossil fuel era is coming
to an end, or has reached its midway point or peak, there is a growing
scientic literature addressing this problem. The idea that there was a
limited amount of fossil fuel in the ground and that consumption demands could one day empty the worlds oil reserve was, for much of the
twentieth century, regarded as foolish. This was true despite the fact that
evidence of this possibility had been discovered in the 1950s.
47
This literature will be reviewed more fully later. For now, it is useful to bear in mind that the big, isolated prison system the United States
has built will be difcult to operate and will become much more costly
than it is now as fossil fuel supplies dwindle and the price of oil rises.
Policy makers will need to address the size of our prison system and
mechanisms for coping with the population we have incarcerated in an
era where fossil fuels become increasingly scarce. Either we plan for this
future now or face the consequences later. And, if not foreseen, those
consequences could be dramatic and involve mass riots in large prisons
where inmates are denied heat, and where security has been threatened
by the lack of fossil fuel needed to operate security devices.
Conc lusi on
A big prison system, such as the one operated in the United States, is
a big problem for a number of reasons. Big prisons produce big anticipation that they will reduce crime. The relationship between the expansion
of the American prison system and crime will be addressed later in this
book. A big prison system also produces nancial burdens. How much
do we spend on our big prison system? How much should we spend?
When we spend more on prisons, where does the money come from?
We will examine these issues in a later chapter.
It is estimated that about 25 percent of all prisoners in the world today
are housed in American prisons and jails. Given the large number of inmates
in U.S. prisons, we should expect that our country would have very low
rates of criminal offending. Compared to other nations, however, it does
not. This fact should make us wonder why our country has selected such
a different response to crime than most other nations of the world.
The big prison policy that has driven the American response to
crime has not, despite its proponents claims, met the challenge. And,
hidden behind the big prison system, lay potentially big problems that
will emerge as the predicted world oil shortage inches its way closer
and closer.
In this chapter, the idea that our big prison system is an ineffective
crime control model has been introduced. This chapter also introduced
the idea that Americas big prison model has benets outside its purported crime control function. One such function is economic, and it
seems that the continued growth of our system of imprisonment is also
48
B i g P ri s on s, B i g D ream s
being driven by private sector interests in the prison system that have
emerged over the past two decades.
In order to understand each of these issues more fully, additional
analysis of each of these points is in order. The next step is to examine the growth of the American prison system, its relationship to crime
control, and the growing interest of the private sector in the continued
expansion of the system of imprisonment.
Chap te r 3
The United States has the worlds largest prison system. At mid-year 2005, federal and state prisons in the United States
housed more than 1.4 million inmates (Harrison and Beck, 2006). Given
the long-term and recent trends in imprisonment in the United States,
we can estimate that the U.S. prison population will surpass the 2 million inmate mark before 2010. Only a decade ago (1994), the U.S prison
system had just managed to squeeze in its one-millionth inmate. In 1986,
the system incarcerated 500,000 inmates, which was twice the number
of inmates incarcerated in 1976. To make a long (three decade) story
short, in the previous twenty-seven years, the population in U.S. prisons
has doubled three times. This rate of growth in imprisonment is unprecedented in the modern history (from 1925 on) of the U.S. prison system.
Below, these trends in imprisonment are explored in greater detail.
Imprisonment Trends in the United States Since 1925
To get a better feel for these data, prison population growth for the
United States is depicted in table 3.1 and gure 3.1. Take a moment to
examine the table and graph, taking particular note of the trends you
observe in the graph. What you will see may surprise you.
The story of the rapidly expanding U.S. prison population began in
1973. Shortly after 19731976 to be exactthere were 262,833 prison
inmates incarcerated in American prisons, which was twice the number
of inmates incarcerated in 1930. The fact that the number of inmates
in American prisons had doubled between 1930 and 1976 is, in itself,
rather unremarkable because by 1976 the U.S. population had increased
by about 74 percent over its 1930 level. Still, while the population of
49
Table 3.1
Rates of Imprisonment Per 100,000 Population, United States,
19252003
Year
Number
Imprisoned
Imprisonment
Rate
Imprisonment
Index
1925
1926
1927
1928
1929
91,669
97,991
109,983
116,390
120,496
79
83
91
96
98
82
86
95
100
102
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
129,453
137,082
137,997
136,810
138,316
144,180
145,038
152,741
160,285
179,818
104
110
110
109
109
113
113
118
123
137
108
115
115
114
114
118
118
123
128
143
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
173,706
165,439
150,384
137,220
132,456
133,649
140,079
151,305
155,977
163,749
131
124
112
103
100
98
99
105
106
109
135
129
117
107
104
102
103
109
110
114
1950
1951
1952
1953
166,123
165,123
168,233
173,579
109
107
107
108
114
112
112
113
(continued)
Number
Imprisoned
Imprisonment
Rate
Imprisonment
Index
1954
182,901
112
117
1955
185,780
112
117
1956
189,565
113
118
1957
195,414
113
118
1958
205,643
117
122
1959
208,105
117
122
1960
212,953
117
122
1961
220,149
119
124
1962
218,830
117
122
1963
217,283
114
119
1964
214,336
111
116
1965
210,895
108
113
1966
199,654
102
106
1967
194,896
98
102
1968
187,914
94
98
1969
196,007
97
101
1970
196,429
96
100
1971
198,061
95
99
1972
196,092
93
97
1973
204,211
96
100
1974
218,466
102
106
1975
240,593
111
116
1976
262,833
120
125
1977
278,141
126
131
1978
294,396
132
138
1979
301,470
133
139
1980
315,974
139
145
1981
353,673
154
160
1982
395,516
171
178
1983
419,346
179
187
(continued)
Number
Imprisoned
Imprisonment
Rate
Imprisonment
Index
1984
443,398
188
196
1985
480,568
202
210
1986
522,084
217
226
1987
560,812
231
241
1988
603,732
247
257
1989
680,907
276
288
1990
739,980
297
309
1991
789,610
313
326
1992
846,277
332
346
1993
932,074
359
373
1994
1,016,691
389
405
1995
1,085,022
411
428
1996
1,137,722
427
445
1997
1,194,581
444
462
1998
1,245,402
461
480
1999
1,304,074
463
483
2000
1,321,137
469
489
2001
1,345,217
470
490
2002
1,380,516
476
496
2003
1,408,361
482
502
2004
1,433,793
486
506
53
3.1. The Trend in the Rate of Imprisonment Per 100,000 Population for the United
States, 19252000
the United States grew during this period, the rate of imprisonment was
expanding at a greater rate than did population growth.
In terms of the rate of imprisonment, the total increase from 1930
through 1976 was 15.4 percent over nearly 5 decades. During this 46-year
period the rate of imprisonment grew from a modest 104 inmates per
100,000 population (1930) to 120 inmates per 100,000 (1976). On average, that amounted to an annual growth rate of only 0.34 percent. Still, for
this 46-year period, the growth of imprisonment exceeded the growth of
the U.S. population. The margin of difference was signicant, but not tremendous. And, despite prison population growth that exceeded general
population growth, the rate of imprisonment increase was a modest 16
inmates per 100,000 American citizens over this time period.
The year 1973 marked the beginning of a period of expansive prison
growth, regardless of how this growth is measured. During a substantial portion of this period, the rate of prison growth would far exceed
54
B i g P ri s on s, B i g D ream s
55
For much of this period, however, the rate of imprisonment was maintained between the range of 104 to 118 inmates per 100,000 citizens. In
addition, during this period, the rate of imprisonment uctuated in both
an upward and downward direction. As we shall see, this is an important
characteristic of imprisonment before the mid-1970sa characteristic
that disappeared in the modern era.
In contrast to this earlier period, the rate of growth in imprisonment
after the mid-1970s was substantially larger. This growth will be examined in greater detail below. For now, it is instructive to note that the
annual rate of growth in the rate of imprisonment from 1972 through
2000 was nearly 15 percent per year compared to the less than 1 percent
annual growth before the mid-1970s. It should also be noted that the annual average growth in the imprisonment rate before the mid-1970s was
not only low, but at times it actually contracteda tendency that is never
apparent after 1973.
Growth rates can tell us a great deal about changes in the level of
imprisonment, and can help identify periods during which the growth
rate of the American prison system was unusually high. They can also be
used to identify periods of stability and prison population decline.
Growth, 19251972. To begin, let us examine the growth in the rate
of imprisonment from 1925 through 1972, focusing on three characteristics of annual imprisonment rates: growth or an increase in the imprisonment rate; contraction or a decline in the imprisonment rate; and stability or no substantial change in the rate of imprisonment. For purposes
of this investigation, an increase in imprisonment is dened as a change
in the imprisonment rate greater than 3 percent, while a decrease in imprisonment is dened as a contraction of more than 3 percent. When the
change in the annual rate of imprisonment falls between plus or minus
3 percent, the rate of change is considered to be constant. Using these
criteria, we discover that from 1925 through 1972, the annual rate of imprisonment increased twelve times, decreased nine times, and remained
constant twenty-seven times. The largest increases were recorded during
the 1930s, and probably reected dramatically transformed social and
economic conditions associated with the Great Depression.
The greatest decreases in imprisonment occurred during World War
II. From 1940 through 1945, the rate of imprisonment declined from
56
B i g P ri s on s, B i g D ream s
doubled during the forty-six years that spanned from 1930 through
1976. Due to the impact of population expansion, the increase in the
rate of imprisonment was about one-half as large (approximately 52 percent). Over the next three decades (19762000), imprisonment would
double two more times: By 1982, the number of people imprisoned was
twice the number incarcerated in 1973; in 1992, the number imprisoned doubled the 1983 total. Based on current trends, it appears that the
U.S. prison population total of nearly 850,000 in 1992 will again double
by 2008. What is also remarkable is that the time it took the prison population to double contracted signicantly. Before the 1970s, the prison
57
58
B i g P ri s on s, B i g D ream s
1964 should affect both crime and imprisonment levels in later decades.
Given that crime is most prevalent between eighteen through twentyfour, we would expect the baby boom effect to be registered on rates of
imprisonment (or numbers imprisoned) by 1965 (1947 +18 = 1965),
and continue through 1988 (1964 + 24) at the latest. In reality, the rate
of imprisonment from 1965 through the mid-1970s was falling, as was
the number of people imprisoned. In fact, the rate and number of people
imprisoned fails to respond to the baby boom effect until nearly a full
decade after this effect is expected, or at the tail end of the baby boom
generationwhich included a declining birth cohort sizereached the
crime-prone years.
In addition, consider that the rate and number of people imprisoned
not only lagged well behind the prediction we would generate using
the baby boom population, but that the rate of imprisonment did not
become lower as baby boomers aged. That is, if the baby boom affects
crime and imprisonment, we would expect to see an increase in the rate
of imprisonment by the mid-1960s, and a gradual decline in the rate of
imprisonment after 1982, when the tail end of the baby boom generation had aged beyond the crime-prone years. But, the number and rate
of people imprisoned failed to decline as would be predicted from the
baby boom effect. In fact, the imprisonment peak occurred too late, and,
moreover, failed to decline in ways that would support a connection
with the baby boom.
In sum, when looked at over the long run, the rate and number of
people imprisoned in the United States can be divided into two large
periods: 19251972, and 1973 through the present. The rst period was
marked by upward and downward swings in the use of incarceration. The
latter period was marked by a persistent high rate of growth in imprisonment not seen during the earlier time period.
P ri son Growth: A nnual Pe rc e nt C hang e s
The rather statistically unsophisticated form of trend analysis presented above represents one way that changes in the rate of imprisonment
in the United States can be tracked. To be sure, much more complicated
methods can be used to express and assess imprisonment trends. These
methods, however much they might reveal about nuances in imprisonment trends, or specic turning points in these trends, are unnecessary for
59
Table 3.2
Annual Percent Change, Rate of Imprisonment Per 100,000
Population, United States, 19262000
Year
% Change
Direction
of Change
5 % Direction
of Change*
1926
5.0
1927
9.6
1928
5.5
1929
2.1
1930
6.1
1931
5.8
1932
0.0
1933
0.9
1934
0.0
1935
3.7
1936
0.0
1937
4.4
1938
4.2
1939
11.4
1940
4.4
1941
5.3
1942
9.7
1943
8.0
1944
0.3
1945
2.0
1946
1.0
1947
6.1
1948
0.9
1949
2.8
1950
0.0
1951
1.8
1952
0.0
1953
1.9
(continued)
% Change
Direction
of Change
5 % Direction
of Change*
1954
3.7
1955
0.0
1956
0.9
1957
0.0
1958
3.5
1959
0.0
1960
0.0
1961
1.7
1962
1.6
1963
2.6
1964
2.6
1965
2.7
1966
5.6
1967
3.9
1968
4.1
1969
3.2
1970
1.0
1971
1.0
1972
2.1
1973
3.1
1974
6.3
1975
8.8
1976
8.1
1977
5.0
1978
4.8
1979
0.8
1980
4.5
1981
10.8
1982
11.0
1983
4.4
(continued)
62
B i g P ri s on s, B i g D ream s
% Change
Direction
of Change
5 % Direction
of Change*
1984
5.0
1985
7.4
1986
7.4
1987
6.5
1988
6.9
1989
11.7
1990
7.6
1991
5.4
1992
6.1
1993
8.1
1994
8.4
1995
5.7
1996
3.7
1997
4.0
1998
3.8
1999
2.6
2000
0.4
2001
0.2
2002
1.3
2003
1.3
63
several substantive differences between these two methods of determining change are evident.
First, for the period after 1972, 13 of the changes previously included as positive changes in the rate of imprisonment are now counted
as stability because they involve less than a 5 percent change in the annual rate of imprisonment. Second, 10 of the positive changes in imprisonment before 1973 are now counted as stability for the same reason.
Finally, 14 of the negative changes in imprisonment noted before 1972
are now counted as evidence of stability.
When we reassess this trend transformation, we again see that all
the periods of negative growth in imprisonment occurred before the
1970s. The number of negative changes and positive changes diminishes, and the number of relatively stable rates of imprisonment, which
were dened as between plus and minus 5 percent, increases. But, this
method also has limitations.While it more accurately classies individual
changes in imprisonment from year to year (e.g., it eliminates counting
very small changes as indicating a change in direction of the imprisonment rate), it provides a poor description of the trend in imprisonment
across blocks of time. For example, when we use a 5 percent cutoff, we
classify the change that occurred between each individual year from
1932 through 1938 (e.g., the change from 1932 to 1933; or from 1933
to 1934) as a year of no signicant change in trend. However, over the
entire span of these years (e.g., the change from 1932 to 1938), the rate
of imprisonment rose from 110 per 100,000 to 123 per 100,000, or by
11.8 percent. Thus, while the yearly changes in the rate of imprisonment are insignicant when examined on an annual basis, the changes
across a period of time as short as ve years where annual change is
stagnant can produce substantial changes in imprisonment. The annual
assessment method, in other words, leads to an error when we try to
assess the broader trend in imprisonment; and we cannot generalize
the annual changes that were signicant or insignicant into a larger
conclusion about the change in imprisonment trends across a series of
years. That is, when we aggregate the signicant or insignicant annual
changes, or when we look across a series of years rather than across each
year, the annual trends of stability, contraction, and expansion in the rate
of imprisonment are masked. They are there, but hidden by the form of
analysis. But the reverse is also true: longer-term trends are masked by
64
B i g P ri s on s, B i g D ream s
65
Table 3.3
Grouped Trends Using 5 Percent Change to
Indicate Growth (+), Stability (0), or Contraction (-)
in the Rate of Imprisonment
Direction
of Change
Number
of Years
19261931
19321938
1939
1940
19411943
19441946
1947
19481965
18
1966
19671971
19741977
Year(s)
19781980
19811982
19830
19841995
12
19962003
(less than 5 percent annual change), negative (more than 5 percent reduction in imprisonment), or positive (more than 5 percent growth in imprisonment), and better illustrates the trend in oscillations in imprisonment.
Examining table 3.3 we can isolate 15 groupings. The rst group consists
of the 7 years (19261933) during which the annual change in imprisonment across each year was 5 percent or greater; the second consists of
the 7 years (19321938) during which the annual change in the rate of
imprisonment was between + 4.95 and -4.95 percent; the third group
consists only of the year 1939, when imprisonment rose by more than 5
percent; and so on.
One of the more interesting points table 3.3 draws our attention to is
the grouping that occurs toward the end of this time spanthe fteenth
66
B i g P ri s on s, B i g D ream s
grouping, which covers the years 1996 through 2003. Even though the
rate of imprisonment rose consistently over this time period (from 427
per 100,000 in 1996 to 482 per 100,000 by 2003), the annual change
across each year was less than 5 percent. Taken together, the change in
the imprisonment rate across the entire 8-year period was 12.9 percent.
Since 1973, these years (19962003) accounted for the smallest increases
in the rate of incarceration. At the same time, the crime rate declined
over this period, and the decline was quite signicantfor example,
crimes known to the police declined by 27.5 percent. So, it appeared,
at least on the surface, that rising rates of imprisonment were related
to a decline in crime. The problem with reaching such a conclusion, as
shall be shown later, is that these short-term trends are not indicative of
the longer-term relationship between incarceration and crime, and may
present an aberration from the norm.
P ri son Growth and C ri m e
As Jenni Gainsborough and Marc Mauer point out in their study
Diminishing Returns (2000), by the late 1990s, leading policy analysts
in the United States, among them Charles Murray, James Q. Wilson, and
John DiIulio, were telling policy makers and the American public that
the reason crime had gone down in the 1990s was the signicant rise
in imprisonment. What policy pundits like DiIulio, Wilson, and Murray
left out of the story was that in 1991 the United States experienced its
highest crime rate since 1980. The 1991 crime rate for offenses known
to police was 5,898 per 100,000; the rate in 1980 was 5,950 per 100,000.
Also omitted from this discussion was the fact that the decline in crime
in the 1990s only took place following a decade and a half of rising imprisonment rates. What needs to be kept in mind is that the decline in
crime experienced in the 1990s occurred after the rate of incarceration
had risen each and every year since 1973 and crime had reached its highest point in a decade. In fact, from 1973 through 1991, the rate of imprisonment had risen by 226 percent, a signicant increase especially when
we consider that crime was on the rise throughout much of this period.
If an increase in imprisonment was responsible for lowering crime, how
did it happen that a nineteen-year period during which imprisonment
rose persistently (19731991) could yield nearly the highest level of crime
seen during his period? Moreover, given that crime had hit what might be
67
68
B i g P ri s on s, B i g D ream s
inequality that had taken place under the Clinton administration, and the
fact that crime tends to cycle, and that 1991 might perhaps mark some
kind of upper boundary in crime,
Nevertheless, by the late 1990s, crime was down, and conservatives
were giving credit to our big prison system (as well as to other crime
policies, such as community policing and zero tolerance). Prison proponents made this claim often, and in very public forums, even though they
did not have sufcient evidence that expanding imprisonment worked,
or that other factorsespecially factors they had failed to examine
might also have inuenced crime rates.
For these reasons, the trends in crime and imprisonment in the period after 1992 become of great importance. During this period, the
crime rate continued to shrink while the expansion of the prison system
slowed. Why should this happen?
In fact, by 1992, the dramatic rise in imprisonment rates in the
United States had nally begun to slow and, in some states, even reverse
itself. At the same time that the imprisonment rate leveled off, and the
annual rate of change in imprisonment was consistently less than 5 percent, for the rst time in decades crime went down consistently from
one year to the next, and not just once or twice. Beginning in the 1990s,
crime went down eleven straight years, perhaps for the rst time ever in
U.S history, and at least since reliable crime data has been available. But
could this result be attributed to a rising rate of incarceration?
To compound matters, the overall decline in crime across this latter period (19912002) was, as noted earlier, fairly large: a 30.2 percent
overall decline in crimes known to police. What no one seemed to notice was that crime was going down the most when the rate of imprisonment started to level off, not when the imprisonment rate was rising the
most from year to year.
During the early 1980s, the prison system grew very quickly, and
was lled just as quickly. Crime, however, did not fall consistently, and
in fact uctuated on its way to the highest level of crime in modern
American history. In 1973, the rate of crimes known to police was 4,154
per 100,000; in 1980, 5,950 per 100,000; in 1984, 5,038 per 100,000; and
in 1991, 5,898 per 100,000. Crime oscillated while imprisonment rose
substantially, from 96 per 100,000 in 1973, to 139 per 100,000 in 1980,
to 188 per 100,000 in 1984, to 313 per 100,000 in 1991.
69
70
B i g P ri s on s, B i g D ream s
in crime during the mid-1990s became evidence that conservative approaches to crime control worked, and no one questioned how anyone
had arrived at this conclusion. No one, for example, seemed worried that
a notable crime suppression effect could not be seen prior to this time,
even though the rate of imprisonment had expanded continually. To be
sure, as long as no one looked at the historical record, or noticed that
low rates of crime earlier this century were accompanied by low rates of
imprisonment, or noted that economic trends had been reversed in the
1990s, conservative wisdom on this issue could not be challenged.
Challenges to the conservatives conclusions do, however, exist.
Gainsborough and Mauer (2000) questioned the conclusion of conservatives and prison expansion proponents, and pointed out that during the
national decline in crime from 1991 to 1998, states with the largest increases in incarceration experienced, on average, smaller declines in crime
than other states. The above average states increased their rate of incarceration by an average of 72 % and experienced a 13 % decline in crime,
while the rate of incarceration in below average states rose by 30 % and
crime rates declined by 17 % (4). In other words, crime declined more
in states where the imprisonment rate rose at a reduced rate. As a result,
it is difcult to conclude that the rise in the incarceration rate alone was
sufcient to reduce crime. As examples, Gainsborough and Mauer offer
evidence on crime and incarceration rates from four states: Texas, New
York, California, and Massachusetts. Of these states, Texas had the largest
increase in imprisonment, 144 percent.Yet the level by which crime was
reduced in Texas was similar to the reductions in crime experienced in
New York, California, and Massachusetts, where the increase in imprisonment was substantially lower, ranging between 21 and 24 percent in
Massachusetts and New York, to 52 percent in California.
Overall, states that increased the use of incarceration the most in
the period 19841991 experienced slightly less of a rise in crime than
other states, 15 percent compared to 17 percent (4). This 2 percent difference, which is inconsequential from a statistical perspective and could
be due to random measurement error, had a substantial nancial cost:
The estimated cost for additional prison construction and housing for
this 2 percent gain was $9.5 billion (4).
It is possible to draw more extreme cross-state comparisons than
the ones offered by Gainsborough and Mauer using their data (table 1
71
Table 3.4
Changes in Incarceration and Crime Rates by State, 19911998, adapted
from Gainsborough and Mauer (2000), Showing Percent Change in (1)
Imprisonment, (2) Total UCR Crime Rate, (3) UCR Violent Crime Rate,
and (4) UCR Property Crime Rate.
(1)
(2)
(3)
(4)
Texas
144
-35
-33
-35
West Virginia
131
-4
30
-7
Wisconsin
113
-21
-10
-21
Hawaii
101
-11
-11
North Dakota
88
-4
37
-5
Iowa
79
-15
-17
Mississippi
74
South Dakota
72
-15
-15
-15
Montana
69
12
-1
12
Idaho
61
-11
-3
-12
Louisiana
59
-5
-18
-3
Pennsylvania
58
-8
-7
-8
Dist.of Col.
57
-18
-30
-14
California
52
-36
-35
-36
Vermont
52
-21
-9
-21
Minnesota
50
-10
-2
-10
Connecticut
50
-29
-32
-29
Missouri
50
-11
-27
-8
Oklahoma
50
-12
-8
-12
Nebraska
48
35
-2
Georgia
47
-16
-22
-15
Kentucky
45
-14
-35
-11
Illinois
45
-21
-22
-20
Colorado
43
-26
-32
-25
Tennessee
43
-6
-2
-7
Indiana
42
-13
-15
-13
New Mexico
42
15
-1
Wyoming
38
-13
-20
-13
21 Jurisdiction with
above average
increases in the
incarceration rate
30 States with
below average increases in the rate
of incarceration.
(continued)
73
(2)
(3)
(4)
New Hampshire
38
-30
-10
-31
Utah
38
-2
10
-2
Washington
36
-7
-18
-6
Kansas
34
-12
-21
-11
Ohio
33
-14
-35
-11
North Carolina
33
-10
-12
-9
Alabama
32
-14
-39
-10
Arkansas
31
-17
-17
-17
Florida
30
-19
-21
-19
Virginia
28
-21
-13
-21
Arizona
28
-11
-14
-11
Rhode Island
27
-30
-32
-30
New Jersey
25
-33
-31
-33
Delaware
24
-9
-11
New York
24
-43
-45
-42
Nevada
23
-16
-5
-18
Massachusetts
21
-35
-16
-39
Michigan
20
-24
-23
-24
Alaska
20
-16
-19
South Carolina
16
-7
-7
-6
Maryland
14
-14
-17
-13
Oregon
14
-2
-17
Maine
-19
-5
-20
AVERAGE
47
-15
-12
-15
Top 21
72
-12
-6.6
-13
Bottom 30
30
-17
-16.5
-16
74
B i g P ri s on s, B i g D ream s
This trend is also evident when violent and property crimes are examined separately. Most interesting is the fact that imprisonment, which
theoretically should protect society from the most dangerous criminals,
appears to have less of an effect on violent crime than property crime
rates for the 21 jurisdictions. This conclusion also holds when we compare the violent and property crime rates in the 21 high-imprisonment
states to states with lower-average increases in imprisonment separately.
These data lead to the conclusions that: (1) the relationship between
incarceration and crime is not clear, nor direct; (2) it is impossible to predict the extent to which crime will change knowing the change in incarceration; (3) the change in the level of imprisonment and crime is not
proportional; (4) substantial crime declines are associated with smallerthan-average increases in the level of incarceration. From a policy perspective, these conclusions lead to a broader, more general conclusion,
namely that the substantial investment required to increase the rate of
incarceration is not guaranteed to reduce crime.
What should be made clear at this point is that these conclusions
were not the story being told by conservative supporters of prison expansion. Indeed, rather than reecting what was known about the relationship between imprisonment and crime that could be extracted from
data from the two prior decades, or from data across states in the 1990s,
conservative policy makers latched onto short term aberrations to present a rosy picture of imprisonments crime suppression potential.
Prison Growth: The Imprisonment Index
Another means of describing the growth and relative size of the
prison system is to create a measure I call the imprisonment index (see
the last column in table 3.1). Indexes are used in many different disciplines to provide an idea of how much something has changed over time
by anchoring the measure of change to a specic level of the trend being
studied, or to a specic year in the trend. A widely known example is
the Consumer Price Index. The same idea can be used to construct an
imprisonment index.
The imprisonment index was derived by dividing the imprisonment
rate in any given year by the imprisonment rate in 1973, and multiplying the result by 100. Thus, for instance, the index of imprisonment in
1973 would be 96 / 96 * 100 = 100. This calculation standardizes the
75
imprisonment rate by its level in 1973, and allows the imprisonment rate
in all other years to be compared against the 1973 rate. For example, if the
imprisonment index in another year was greater than 100, then we know
that imprisonment in that year was higher than in 1973. An imprisonment
index of 200 indicates that the level of imprisonment was twice as high as
in 1973; an index of 300, three times as high as in 1973; and so on.
The imprisonment rate in the year 1973 was selected as the standardization marker because it marks the year when the prison system in
the United States began it precipitous expansion. In effect, this standardization allows the level of imprisonment in each year to be interpreted
as a percentage of the rate of imprisonment in 1973. For the sake of
simplicity, this index is rounded to the nearest whole number.
The imprisonment index also provides another mechanism for assessing trends in imprisonment in the United States, but this time, relative to
1973. For the seventy-nine years represented in table 3.1, imprisonment
was below the 1973 level of imprisonment only six times, at the 1973
level three times, and above the 1973 level seventy times. The highest
imprisonment index before 1973 occurred in 1939 (the index measure
for 1939 is 143) in the aftermath of the Great Depression, and was not
surpassed until 1980, following the economic recession that characterized
much of the 1970s, indicating a possible connection between incarceration and economic conditions (see chapter 5 for further discussion).
Using the imprisonment index to gauge the growth of imprisonment
and restricting our view to years after 1973 reveals that the only times
in which this index doubles is from 1973 through 1985, and again from
1985 through 1995. When we examine the entire series, we see that the
imprisonment index doubled from 1925 through 1982, a fty-seven-year
period. Looking ahead from 1982, we see the next doubling point in 1993,
just eleven years later. This compression in the doubling of the imprisonment index conrms the earlier analyses that indicated the existence of
elevated rates of doubling in the use of imprisonment since 1973the
portion of the data that corresponds with Americas imprisonment binge.
The rate of doubling indicated by this index also demonstrates that most
of the growth in the U.S. prison system occurred since 1980.
Finally, this index can be employed to consider the imprisonment
differences between 1925 and 1973, and 1973 and 2003. The 2003 imprisonment index of 502 indicates that the use of imprisonment had
76
B i g P ri s on s, B i g D ream s
increased by a factor of 5 over the 1973 level. The 1925 index level indicates that the use of imprisonment that year was 82 percent of the level
of imprisonment in 1973. Thus, the 1925 and 1973 levels of incarceration were not extraordinarily different, while the difference between the
1973 and 2002 imprisonment index was substantial.
There are a number of other questions that can be addressed using
this index. Specically, is there a relationship between the imprisonment
index and the level of crime? Let us, for the moment, focus on the period since 1960, and examine the relationship between the imprisonment index and an index of crimes known to police.
An index of crimes known to police can be created in the same way
we created an imprisonment index. The rate of crimes known to police
per 100,000 population in 1973 is the denominator, and the rate of crime
known to police in any given year is the numerator. Thus, the crimes
known to police index will, like the imprisonment index, be set to 100 in
1973. Table 3.5 shows these data for a select sample of years since 1960.
Consistent with earlier ndings, table 3.5 demonstrates that the
magnitude of change in the imprisonment rate (measured relative to
the 1973 index) has no obvious suppression effect on the level of crime
(relative to the 1973 crimes known to police index). In fact, this table
illustrates a point that is in contention when debating whether the level
of imprisonment affects crime. Ironically, the level of imprisonment before 1973, when the rate of incarceration started its annual rise, appears
to indicate a crime suppression effect. That is, higher rates of punishment prior to 1973 (in this case, 1960 and 1965) are related to lower
rates of offending. This effect disappears for the period after 1973, and
only reemerges in 2000 (or perhaps somewhere in between 1995 and
2000, since effects for individual years are not measured in is table). This
table, then, raises the possibility that crime either has a natural cycle,
or that the cycling of crime is related to some factor(s) other than the
use of incarceration.
The other trend that is evident in this table is that the change in imprisonment must be relatively large to demonstrate a crime reduction effect across a relatively short time period (e.g., the difference between 1980
and 1985, which is 45 percent; the difference between 1990 and 1995,
which is 39 percent; or the difference between 1995 and 2000, which is
15 percent). Yet, this table also illustrates that increases in punishment do
77
Table 3.5
Comparison, Imprisonment Index and Crime Known to Police Index,
Select Years, 19602000 (rounded to nearest whole number).
Year
Imprisonment
Index
(100=1973)
Crime
Index
Change1
(Imp/Crime)
Change, 1973
(Imp/Crim)2
1960
122
46
22 / -54
1965
113
59
-7 / 28
13 / -41
1970
100
96
-12 / 63
0 / -4
1973
100
100
0/4
1977
131
122
31 / 22
31 / 22
1980
145
143
11 / 17
45 / 43
1985
210
126
45 / -12
110 / 26
1990
309
140
47 / 11
209 / 40
1995
428
127
39 / -9
328 / 27
2000
490
99
15 / -22
390 / -1
1. Percent change in index measured from the previous year listed in table (e.g., 1970
compared to 1965).
2. Percent change in index compared to 1973.
78
B i g P ri s on s, B i g D ream s
79
new police commissioner William Bratton. Bratton instituted a zero-tolerance policy that centered on suppressing minor crimes and felonies by
expanding arrests, searches, and the execution of warrants. A statistical
program was also used to identify crime hot spots so that these areas
could be targeted for enhanced policing. Bratton was of the opinion that
increased police action taken against low-level offenders would sweep
up a larger proportion of serious future felons, and also send a message
to those contemplating crimes that New York City would not tolerate
criminal behavior.
As Jacobson points out, those who supported this view argued that
expanded police activity increased arrests, and thus would also generate
more punishment. Therefore, as the number of arrests for targeted offenses increased, so too should indictments, convictions, and prison and
jail sentences. Jacobson, however, suggests that this traditional view of
New York Citys crime drop has a number of serious aws.
To begin, the crime rate in New York began to decline before Giuliani and his new police commissioner could affect a change in New
York Citys crime policies. Indeed, statistics indicate that New York Citys
crime rate began its decline in the last two years of Mayor Dinkinss term
in ofce. To be sure, that trend continued under Giuliani, but, because
this trend was already underway, it cannot be solely attributed to new,
Giuliani-era crime policies.
Furthermore, during the rapid crime decline in New York City, New
York States prison population grew at a much slower rate than would
be expected given increased police activity in New York City. In fact,
the states prison population grew by only 5 percent, or at a rate that was
well below those found in other states during this time period. Jacobson
argued that the signicance of New York States prison growth has to do
with a related factthat approximately two-thirds of New York State
prison inmates are derived from New York City. In addition, while New
York States prison population grew during this time period, it was not
due to the inuence of an increase in offenders sentenced to prison who
came from New York City courts. In fact, from 1993 through 2001, the
number of inmates New York City courts sent to New York State prisons
declined by 42 percent. Thus, Jacobson argues, New York Citys crime
decline could not be attributed to a rising rate of incarceration among
offenders from New York City.
80
B i g P ri s on s, B i g D ream s
It was possible, however, that because New York Citys crime policy
targeted minor offenses, that those offenders were receiving lesser punishments, such as terms in New York Citys jail system. But New York
Citys jail population declined by 25 percent from 1993 through 2001.
Thus, not only could the decline in crime not be explained by an increase in punishment, but trends for incarceration among New York City
offenders deed the national trend. Nationally, prison and jail populations were expanding, with prison populations up 45 percent, and jail
population growth near 39 percent for the time period under study. This
lead Jacobson to conclude that New York Citys crime reduction could
not have been caused by incarcerating a larger number of offenders.
Finally, consider that during this period, felon indictments as a percentage of arrests made in New York City were also slipping. This ratio
declined even though the number of felon arrests dropped. In 1993, 39
percent of felony arrests led to an indictment. By 2001, this percentage
had decreased to 25 percent. Overall, fewer arrests were leading to indictment, and fewer offenders were being sent to jail or prison.
Jacobsons analysis of New York Citys crime drop illustrates that
crime reductions occur without resorting to an escalation in the use of
incarceration. Indeed, New York Citys crime rate declined alongside a
modest increase in New York States use of imprisonmentthough the
increase could not be attributed to the fact that New York City sent
more offenders to prisonand a reduction in the size of New York Citys
jail population. When added to other evidence presented by Jacobson
throughout his book, it should be evident that the relationship between a
reduction in crime and the use of incarceration supported by deterrence
and incapacitation arguments did not materialize in New York City. Furthermore, as we have seen from evidence provided earlier, crime sometimes expands along with the rate of incarceration, both at the national
and state levels. Likewise, from cross-sectional data, it is also apparent
that crime sometimes falls when the incarceration rates contracts. These
distinct empirical results provide the kinds of evidence scientists employ
when they reject theoretical proposition as untenable.
Indeed, in support of the conclusion that crime and imprisonment
are unconnected, consider Robert Agnews (2005: 189) summary on this
point: the majority of evidence points to the conclusion that punishment does not reduce offending and may sometimes increase it.
81
Conc lusi on
Taken together, the various observations made above on imprisonment rate changes and prison growth helps us to sort out the prison
expansion patterns, classify and corroborate the existence of growth patterns, and understand the shape and direction of prison growth (and
contraction) patterns in the United States from 1925 through the present. These observations do not, however, explain prison growth rate patterns, they only help us identify periods that require explanation, and
narrow down the types of explanations that may t these data trends.
Why, for example, have prisons grown so rapidly in recent years? What
now needs to be addressed are some potential reasons for this dramatic
rate of growth.
Chapte r 4
83
84
B i g P ri s on s, B i g D ream s
which factors drive prison rates. But, while it is unclear why imprisonment rates have grown so rapidly during the last thirty years, it is more
obvious that the rate of prison growth that has characterized the last three
decades is problematic. The recent tendency of the U.S. prison population to double repeatedly since 1973, for example, provides evidence of
a very troubling trend for expansive prison growth in the United States.
The prison population doubling trend is troubling for several reasons.
First, we must recognize that when crime or punishment in a society either grows or declines rapidly, this is a sign of trouble or social
instability. This observation was offered long ago by Emile Durkheim, in
his well known books The Division of Labor in Society (1893), The Rules
of Sociological Method (1895), and Suicide (1897). To understand this view,
we must recognize that Durkheim argued that crime was normal since
it was found in all contemporary societies, and, consequentially, that a
society without crime would be abnormal. In addition, if crime is normal and serves useful social functions (which we need not detail here;
see Durkheim, 1893, 1897), the eradication of crime would ultimately
disrupt social order. Following this observation, Durkheim wrote in The
Rules of Sociological Method that:crime, for its part, must no longer be
conceived as an evil that cannot be too much suppressed. There is no
occasion for self-congratulation when the crime rate drops noticeably
below the average level, for we may be certain that this apparent social
progress is associated with some social disorder. (72).
In the Durkheimian view, crime was not abnormal. All healthy societies had crime. This leads to the conclusion that crime is not a social
disease, but rather a social regularity or fact. And, for Durkheim (1895:
73), if crime is normal and not a disease, then punishment cannot be
its cure, and its true function must be sought elsewhere.1 How do these
observations apply to recent trends in punishment in the United States?
In effect, the high rates of punishment that characterize recent decades in the United States, and the elevated level in the doubling of the
rate of punishment evident over this period, indicates that there may be
some larger problem in society that is causing the United States to rely too heavily on the powerful form of social control represented by the prison. Consider,
for example, that crime is not caused by a lack of punishment. There are
numerous examples that can be cited to support this claim. Consider,
for instance, the very low levels of punishment and crime that are seen
85
86
B i g P ri s on s, B i g D ream s
87
and public policy makers and consultants; (2) researchers who examine
public punitiveness using selectively worded survey instruments that elicit
more pro-punitive responses (for exceptions, see Cullens various work
on this topic); and (3) by economic and social circumstances that generate fear of perceived criminals among the middle class, and the desire for
status separation by members of the lower classes (e.g., I may be lower
class, but at least Im not a criminal; see endnote 1 of this chapter).
Fifth, the American public and policy makers have not considered
the long-term consequences and the future path that a reliance on a
big prison system brings. What Americans will have to cope with in the
very near future is a looming nancial crisis that will be exacerbated by,
among other issues, the coming oil crisis that will impede the functions
of big prisons, and the growing racial effects Americas big prison system
generates. At some point, the tremendous nancial cost of the lock em
up approach to crime will impede our nations ability to rely on easy
and, as I shall demonstrate in later chapters, ineffective methods for controlling crime represented by imprisonment (see also Clear, 1994; Austin
and Irwin, 2003). Americans must also come to grips with the growing
proportion of the minority population housed in prison, and understand
that relying on imprisonment will exacerbate racial tensions as the likelihood of imprisonment increasingly disrupts the life of a larger percentage of American minorities.
Sixth, Americans must accept the fact that punishment may not be a
useful crime control strategy. The failure of imprisonment to effectively
control crime may, as some argue, be the problem most often hidden
behind the tremendous growth of imprisonment in the United States. If
the growth of imprisonment cannot be shown to be an effective crime
control strategy (in terms of the percent of crime that is suppressed, and
its nancial costs), then why do we continue to rely upon this strategy? In
part, society relies on this approach for two reasons. First, despite all the
evidence to the contrary, we continue to believe that incarcerating more
people ought to lower the rate of crime. The logic of such an idea has
widespread appeal. Second, imprisonment has come to serve purposes
other than reducing crime. These other purposes, described later in this
chapter, drive imprisonment independently of any effect imprisoning
people may have on crime. For now, we need to turn our attention back
to the question of the relationship between prisons and crime control.
88
B i g P ri s on s, B i g D ream s
D o P ri sons Control C ri m e ?
The experiment our nation has undertaken with respect to using imprisonment in an effort to reduce crime is a failure. There is
little evidence that sending more and more people to prison works to
control crime to any signicant degree. In addition, the evidence that
does exist has historically specic limits. In other words, the evidence
that suggests that prisons control crime only applies to certain time
periods, and does not apply consistently to the imprisonment trend that
has been in effect in the United States since 1973. Nor does it apply
across nations, nor across states within the United States, nor does it
hold when we look at the low crime rates earlier in twentieth-century
America. If this is true, why do we continue to believe that imprisoning
more people controls crime?
In part, the widely held belief that increased rates of imprisonment
reduce crime comes from a limited number of studies. Some of these
studies have been produced by highly visible policy analysts, or by important crime control agencies (see Clear, 1996, for extended discussion).
A number of these studies use short-term evidence to draw conclusions about the relationship between imprisonment and crime rates over
the long run, or attempt to transform short-term relationships between
crime and punishment into absolute, immutable relationships. The results
from short-term studies that reach sweeping conclusions may lead us
in the wrong direction. This is true because these studies focused on a
rather limited period of time (e.g., the mid-1990s) during which imprisonment increased and crime decreased. These same studies fail to address
or take into account the fact that social and economic changes or factors other than punishment that might have caused crime to fall during
this period. In short, we need to pay serious attention to the fact that
many studies that support the expansion of incarceration as a method for
controlling crime, especially those written in the mid-1990s, are suspect
because they are methodologically faulty (designed improperly) or reach
conclusions that amount to overgeneralizations from the data that the
researcher analyzed. To be sure, the kinds of methodological deciencies and overgeneralizations evident in pro-imprisonment research from
the mid-1990s are well known to social scientists. Indeed, these errors
are the substance of classic research design guides, such as Campbell and
89
90
B i g P ri s on s, B i g D ream s
for instance, the presence of two chemicals may yield a compound under known conditions. In contrast, we could argue that human behavior
is not so precise because people arent chemicals or physical properties
whose behavior is determined by a set of given conditions. In addition,
there is a tendency to argue that humans, unlike chemicals, have the ability to make choices. Sodium doesnt decide to bond to calcium to form
salt; the outcome is determined by the nature of these chemical and the
conditions of how they are combined. To be sure, not everyone placed in
the same circumstance behaves in the same way. Indeed, this is true; but
what we want to know is not that some people may behave differently,
but rather how do most people behave in a given situation. Likewise, we
could catalog a tremendous number of ways in which individuals might
differ that could affect their behavior. These difference could range from
common descriptive state data that include age, race, and gender, to family factors that include information about parents, siblings, and family
cohesion and intactness, to achievement and status variables such as level
of education and income, to name but a few possibilities.The point is that
we might say that an individuals behavior cannot be predicted with any
degree of accuracyand by accuracy, I mean that mathematical or statistical models predict a high degree of variation in the predicted behavior,
is correct more often than it is wrong, and does not simply produce
statistically signicant results for specic variables because of sample size.
This may occur because people have consciousness and choose their own
behaviors, or because of the vast variation across individuals, which, taken
together, make each person unique. All these explanations are based on
an assumption that if behavior can be explained, it can be best done by
looking at variations across individuals.
On the other hand, it may be that we cannot predict human behavior because of the way in which human behavior is normally studied.
In other words, it is possible that predictions of human behavior are not
precise because we make the wrong assumptions or expect the wrong
outcomes or measure the wrong variables. Further, when the expected
outcome does not occur, instead of looking for an alternative explanation, we invent reasons that explain why we did not discover what we expected, and look somewhere else to conrm what we believed when the
weight of the evidence suggests that we should come up with a new explanation. Thus, rather than declare the hypothesis invalid and discard the
91
92
B i g P ri s on s, B i g D ream s
Let us assume, for the moment, that the expanding rate of imprisonment in the United States could actually be shown to reduce crime. We
would have to ask whether we now have sufcient evidence to serve as the
basis of policy. Most policies, crime policies included, are not determined
simply on the basis of a policys effectiveness in curing the targeted problem. That is true because there is a difference between a useful policy and
a policy that can or should be implemented, or even a policy that someone wants to implement. Often, when the governments agents consider
whether a policy should be implemented, they also consider the costs of
implementing the policy, the cost effectiveness of the policy (the question of whether the costs of the policy reap sufcient rewards), and perhaps even if it would be feasible to implement the policy on some other
grounds. In addition, they are also forced to consider the desires of interest
groups that include lobbyists and those who donate to political funds.
Let us take a specic example that supposes that a crime gene were
to be discovered. What is the policy implication of this nding? Genes
do not, as most people think, determine a specic outcome; they may
inuence behavior, but they cannot be said to be the direct causes of behavior. In other words, the presence of a particular crime gene (if there
was one) might make crime a more probable outcome, but the outcome
is not denite, since, as genetic scientists have shown, genes can be inuenced by the environment (Ridley, 2000; Burdon, 1999, 2004). This may
be especially true for the connection between genes and behavior, where
the gene may come into play only in specic environmental conditions.
Despite these possibilities, how should the government act if there was
a gene-crime connection? Should we test everyone born in the United
States for the crime gene? Then what? Take these children from their
parents? Lock them in secure environments from an early age? Sterilize parents who have produced children with crime genes? Would this
be feasible economically? Would the costs of reducing crime represent a
greater savings than the costs of the programs required to reduce crime?
Would it be socially acceptable? Would enough crime be suppressed by
these actions alone to justify their existence? While a gene-crime connection might be found, it would be unlikely to explain most crimes.
Most crimes are property crimes. What kind of gene would make people
steal? Anthropological evidence, for example, indicates that the human
race spent the majority of its existence living in societies where property
93
was unknown, where property equity was encouraged, or where the accumulation of goods and wealth was prohibited or undermined by social
rules (Henry, 1965; Sahlins, 1972). Thus, it is unlikely that there is a property gene, or that acquiring property is part of human nature.
Here we have come to the crux of a difcult problem: even if increased rates of imprisonment appear to reduce crime, we can pose the
question of effectiveness on cost grounds. How much should we spend
to reduce crime by a specic amount? Would other policies reduce crime
at a lower cost? Sometimes, these questions cannot be addressed directly
because there are no real world examples of alternative solutions being
practiced. When this happens, as is often the case with criminal justice
policies, researchers insert assumptions or guesses about the cost and effectiveness of alternative crime control approaches. In this case, the basis
for determining whether or not a policy ought to be employed depends
entirely on the assumptions the researcher makes about the cost and effect of a nonexistent alternative. It should be clear that in such cases, altering the assumptions researchers employ changes the outcome and the
decision about whether a policy should be followed.
The point is that assumptions have played, and continue to play, a large
role in crime and justice policy in the United States. For instance, the big
prison model employed in the United States is, as we have discussed, based
on a crime reduction assumption. This, and many other assumptions that
underlie the big prison system, are faulty. These faulty assumptions, while
they fail to control crime, help to maintain prison growth. The issue of
crime control and imprisonment is examined further in chapter 6 and 7.
P romot i ng O ur Fa i l i ng P ri s on Ap p roac h
The idea that prisons fail as an effective crime control mechanism
is not a widely held view among politicians, news agencies, the public,
criminal justice agents, or by those who own businesses that rely on the
income generated by our massive modern prison system. Nor does it appear to be a view widely held by academics, since the media publicizes
the large voice of a minority of scholars who believe in the effectiveness
of prisons, and politicians have thrust the same group of academics into
the public limelight. Regardless of how it happened, what has happened
is that American society shares a widely held belief in the effectiveness
of prisons. Furthermore, this widely held belief has helped to promote
94
B i g P ri s on s, B i g D ream s
prison expansion over the past three decades. Below, we examine how the
assumptions about big prisons work at different levels of government.
Federal Agencies
Agents of the criminal justice system promote the idea that bigger
is better, and few are heard to exclaim that our experiment with rising
prison populations really hasnt mattered all that much. Speaking to this
point, Todd Clear (1996) discussed how federal agencies and federal appointees promoted research that supported prison expansion philosophies
and have actively suppressed anti-expansionist research ndings during
the late 1980s and early 1990s. On this point, it is worth quoting Clears
discussion of how the U.S. Department of Justice reacted to Joan Petersilia and Susan Turners reanalysis of a study comparing recidivism differences for California probationers and released inmates. The rst study
demonstrated that prisons were a more effective method for controlling
crime than probation. As Petersilia and Turners reanalysis showed, untenable assumptions helped to produce these results. The governments
response to Petersilia and Turners reanalysis is shocking:
When Petersilia and Turner . . . reanalyzed the . . . database and found
that, when compared to a matched prison sample, the probationers
. . . actually had lower arrest rates than the prisoners, and that with
time, even the incapacitation differences might wash out, the Department of Justice reaction to this analysis was quite different. Faced
with a set of ndings that refuted the punishment/control agenda,
suggesting that the incapacitative effects of imprisonment may be
washed out, over time, by its criminogenic effects, the Department
[of Justice] refused to support its [Petersilias and Turners report] dissemination. Instead, the National Institute of Justice, which funded
the study (expecting, no doubt, a different result) refused to allow
it to be published under federal dollars and attempted to stop [the]
Rand [Corporation, Petersilias employer], from publishing it with
its own money, claiming the research was awed (Clear, 1996: 56).
Clear proceeded to note that:
The extremely cautious handling of the second Petersilia study must
be compared to the lavish acceptance of an internally commissioned
95
96
B i g P ri s on s, B i g D ream s
of American citizens. Two things may play a role. One is related to the
tendency of politicians to point out that their political opponents have,
in the past, behaved differently, which makes them wishy-washy. Apparently, Americans prefer candidates who stick to their positions rather
than ones who revise their ideas as they gather new information. Second,
rather than confront the problem that prisons are an ineffective crime
control strategy head on, the politician is likely to invent an excuse such
as the following: We already have a big prison system. What are we to
do with all those prison beds if we dont send people to prison? We have
already committed extensive resources to this approach. In a few more
years, we will be able to determine whether or not this approach is working. How many years must pass, however, before we know the answer to
the question of whether prisons reduce crime? At this point, more than
thirty years have already passed since the United States embarked on its
great prison expansion experiment. Do we have to wait longer than the
thirty years we have already waited? Does more time have to lapse before
we can assess whether employing the bigger prison system model works?
And wont the prison system get even bigger in the meantime? Wont we
spend more money on a system that doesnt seem to meet its promise?
Are we to believe that we will, in a few more years, be able to walk away
from our big prison system when we couldnt do so on previous occasions when the prison system was smaller?
When I imagine the answers politicians give when faced with questions about the relationship between our expanding prison system and
crime, I often think of someone whose house is on re. When you see
that your house is on re, your natural instinct would be to try and extinguish the re. Politicians know that throwing a liquid on a re is supposed to extinguish it. So, they grab a liquid and start dousing the re.
But the liquid they are throwing is, itself, ammable. Nevertheless, they
keep at the task, throwing the ammable liquid on the re until there
is nothing left to extinguish. In some sense, the problem of the burning
house has been solved by throwing ammable liquids on itthe house
has been reduced to ashesonly the solution isnt a useful one.
From the Top: Presidential Politics
Todd Clear is not alone in making the claim that politicians drive
criminal justice policies, and that they have used their inuence and
97
98
B i g P ri s on s, B i g D ream s
99
work in the oil industry, has a subsidiary, Kellogg, Brown, and Root, that,
among other things, builds prisons. This kind of troubling connection
between public servants and corporations that prot from punishment
may help explain why the push for a bigger prison system has remained
so entrenched in American politics, despite evidence that this approach
has not delivered on its claim. Consider also that the corporations that
now prot handsomely from crime by providing services and commodities and prisons to the government spend a good deal of money lobbying
politicians to make sure that they dont change their minds about the
need to expand our prison system. In short, there are now many reasons
why prison expansion continues. Part of that reason involves the mechanism of support built around the prison as industry, an important idea
that will be examined in chapter 5.
Conservative Academic Support
The revelation that prisons are not an effective crime control strategy
is not news to most academics who study this topic. This is not to say
that there are not some academics who support prison expansion policy.
Indeed, some academics have written the most publicly acclaimed and relied-upon studies and arguments in favor of prison expansion (for discussion and critique, see Clear, 1996).While the majority of academics would
oppose prison expansion as an effective crime control policy, the position
of proprison expansionist academics are favored by policy makers and
politicians because they t a conservative crime control agenda that the
public believes will work despite its many failures. The get tough prison
expansion proponents are also most likely to make news.
Punishment and Reinforcement
From the perspective of the average person, the expansionist view
also ts with a seemingly common-sense notion about the association
between crime and punishment: that punishment alters behavior. Animal
experiments are often cited in support of the punishment affects criminal behavior view. The idea that negative reinforcement affects behavior
has, for instance, become part of popular vernacular. For example, say we
want to stop a puppy from chewing on shoes. One approach to this problem, perhaps the most likely, is to spank the puppy each time it chews
on the shoes. This works if the puppy is capable of mentally linking the
100
B i g P ri s on s, B i g D ream s
101
it turns out, is a more efcient and longer lasting mechanism for creating
behavioral change in animals than is punishment. The bigger problem,
however, involves whether studies on animals can be assumed to apply
equally well to humans. Without belaboring the point, we can say that
of the three methods for changing human behaviorpositive reinforcement, negative reinforcement, and punishmentpunishment is the least
effective on humans.
Obeying the Law
Why do people obey the law? Is obeying the law an outcome associated with the penalties law supplies for transgression? While most people
believe that punishment leads to law abiding behavior, research evidence
suggests otherwise. In fact, research indicates that most people follow
laws because they believe in their legitimacy, not because they fear punishment (Tyler, 1990). The average person, for instance, doesnt walk into
a store and case the joint in order to determine if he might be able to
successfully pull off a robbery without being caught or punished. On the
contrary, the average person doesnt even entertain this idea, not because
they are afraid they will get caught, or that they might go to prison, but
because they have been taught that this isnt the right thing to do. People
may respect what the law has to say about taking property in an illegitimate way; or they may respect the social contract implications of abiding
by legal rules; or they may have other options for obtaining money that
they and others perceive as legitimate, worthy, and honorable; or they
may believe that there is no honor in occupying a criminal status. The
point is that there are a host of reasons that cause people to obey the law.
Punishment is not among the most powerful of these reasons.
To illustrate this point, consider a simple example. Most well-behaved children arent well behaved because they fear their parents; they
are well behaved because they respect their parents and others and understand that behaving well earns them respect.To be sure, this is not true
for all Americans or for all children, and when systems that confer status
for legitimate activities break down, or when they are weak or nonexistent, or when there is no reward for conforming, the legitimacy of laws
that are part of that system become suspect and have little meaning. For
some Americans, for instance, there may be no advantage to conforming
because there is no reward. Similarly, some people may have few or no
102
B i g P ri s on s, B i g D ream s
legitimate options to obtain the money or status they are taught is valued
in American society, and so crime becomes a viable alternative (Messner
and Rosenfeld, 2001). It is also possible that some children have not been
taught to respect others, or how to employ legitimate pathways to earn
the respect of others. Whatever the source of this problem (e.g., family,
schools, the economy, job prospects), it will not be remedied by more
punishment.3 People who are restricted from achieving, who lack access
to achievement means, or who are not taught appropriate methods for
achieving, or who have been overexposed to achievement goals, may all
turn to crime to attain those thingsincluding status and feelings of
self-worththey cannot otherwise obtain. Punishment may force some
people to avoid crime, but it is unlikely to alter the behavior of most
people most of the time for the reasons reviewed above.
The failure of punishment and prisons as a mechanism for changing
behavior is not, however, widely recognized. As a society, we still believe
punishment works well. As a result, we continually increase the level of
punishment in the hope that it will work. And we further increase its use
the more crime rises. Thus, in the face of failing to control crime, we did
not given up on punishmentrather, we extended its reach.
Summing Up Failing Crime Control Strategies
Part of the crisis of imprisonment we are currently experiencing
in this country is the result of desiring rapid, visible results from crime
control policies that are wedded to punishment. Politicians, because they
need the support of the public to remain in ofce, try desperately to
produce the desired results. They give speeches about being tough on
crime, and endorse crime policies they view as tough. The problem is
that few crime policies create rapid and immediately visible results. And,
if politicians cannot make results appear rapidly, at least they can make
the results appear visible. The rising rate of imprisonment is one very
obvious mechanism for making efforts to control crime visible even if it
doesnt really control crime.
If Prisons Dont Cont rol Crime, What Do Th ey Do?
To begin this section, it is necessary to recognize that there are certainly people who we can say belong in prisonpeople who, perhaps,
cannot be reformed, who continue to violate the rights of others in the
103
most egregious fashion. The problem is that not everyone who violates a
law needs to be placed in prison to protect the other members of society. Imprisonment is a severe punishment. Outside of the death penalty,
which most advanced industrial nations except the United States have
abandoned, it is the most severe punishment that society can administer.
As a result, the act of imprisonment should be viewed as societys last
line of defense against criminals, not its rst. If we accept that premise, we
can reach some rational conclusions about who ought to be in prison.
For example, if prisons are the last line of defense against criminals, we
could justiably envision the prison as an institution with no purpose
other than the control of repeat violent offenders who show an inability to refrain from violence. We might even decide that there are a few
other types of offenders that need to be imprisoned to protect society.
It is plausible, for instance, that the prisons might be a useful last defense
against big-time drug dealers or racketeers. If we were to implement
these kinds of rational rules about who should be in prison, the current
system could be slashed substantially to one-fth of its present sizeto
about 300,000 inmates, the number of violent and high-level drug offenders currently housed in the U.S. prison systemwhich would represent a reasonably sized prison population. The last time the U.S. prison
system was that small was in the late 1970s.
If we can accept the proposition that prisons are a last line of defense, then there is something wrong with a society that overuses imprisonment. In addition, there is evidence that the use of imprisonment in
the United States doesnt protect society from the worst criminals.
Further discussion in latter chapters also examines the purpose of
imprisonment, where the issue of imprisonment as a form of class and
race control is discussed. As a prelude to those points, the next section
describes those offenders who are the least likely to be found in the U.S.
prison system: white-collar and corporate offenders.
The E xc lude d O f f e nde r s :
Corp orate and Wh i te - C ol lar O f f e nde r s
To be sure, it can be conceded that the current prison system does
protect us from some criminals and crimes, even if the act of imprisonment doesnt substantially lower the crime rate. We might even say that
the functions of imprisonment reviewed aboveprotecting the public
104
B i g P ri s on s, B i g D ream s
105
where between 70,000 to 120,000 people die from preventable occupationally related diseases, illnesses, and accidents (e.g., Leigh et al., 2000;
106
B i g P ri s on s, B i g D ream s
Reiman, 2003; Lynch and Michalowski, 2006). Note that this gure is
about four to six times the homicide rate. Yet, politicians and the news
media rarely focus on these deaths, or how they might be prevented.
Because these acts are not treated as crimes nor featured in the news,
the public fails to recognize that their jobs pose a greater threat to their
health and well-being than the anonymous criminal they fear. In addition, it should be recognized that in the majority of cases, adhering to
legal requirements would signicantly reduce workplace-related deaths.
In other words, the majority of occupationally related deaths result from
violations of workplace heath and safety laws. In addition, a signicant
number of these deaths could be prevented if stricter regulations were in
place. Leigh et al. (2000) estimate that slightly more than two-thirds of
occupationally related deaths are due to occupationally related cancers,
indicating that stricter workplace rules or better enforcement of existing rules could substantially lower the number of occupationally related
deaths each year.
2. Faulty consumer products. Each year, 3,000 to 4,000 Americans
die from the use of faulty consumer products. These deaths should not
be pushed aside and treated simply as accidents. Numerous case studies
reveal that corporations often knowingly market unsafe products (Friedrichs, 2003).
3. Dangerous products. Dangerous products, such as pesticides
(which include pesticides, herbicides, rodenticides, and fungicides), cause
several thousand deaths each year. These deaths are preventable to the extent that alternative, safer products could be employed to obtain similar
results (Lynch and Stretesky, 2001). In addition, the warning labels placed
on these products are written in complex language, and in rather small
print, making it difcult for the average consumer to read the instructions and comprehend their content (Lynch and Stretesky, 2001).
4. Unnecessary surgery. Each year, a substantial number of Ameri-
cans die from unnecessary surgery. These surgeries, because they are not
required to save lives, can be viewed as crimes (Reiman, 2003). It has been
estimated that there are approximately 16,000 deaths that result from unnecessary surgeries in the United States each year (Reiman, 2003). Many
107
108
B i g P ri s on s, B i g D ream s
and dangerous criminals are not the subject of the criminal justice systems
control process. Imprisonment is the punishment to which poor people
are subjected, and others have argued that the American prison system is a
substitute for the poorhouse (Goldfarb, 1969). In other words, one of the
primary purposes of the prison has become to control the poor through
either direct physical restraint, or through the threat of punishment, which
echoes the sentiments of Ebenezer Scrooge, who lamented, Are there no
prisons? No poorhouses? Scrooge would have no complaint to lodge if
he lived in modern-day America.
One excuse often offered for treating white-collar and street criminals differently has to do with assumptions that the level and kind of
intent criminals form when committing these two acts is unique. The
corporate offender, for example, is often excused when he harms the
public because he did not intentionally target a specic individual. In
contrast, the victim of an ordinary street crime of violence is a specic
target, and the offender is viewed as intending to harm the victim. The
difference here is semantic rather than real. The corporate offender, like
the street offender, cannot carry out his offense without a victim. The
corporate offender knows this is true. He intends to harm a victim to
enrich himself or his corporation. He just doesnt have a specic victim
in mind because he is even more callous than the street offenderfor
the corporate offender any victim will do; they are nameless and faceless. Furthermore, the corporate offender does not think of a victim, he
thinks of a mass of victims, for this is the only way he can satisfy his unrestrained lust for more. In this respect, the corporate offenders crimes are
more heinous than the act of the ordinary criminal because he does not
care if his victim is John Smith, his neighbor, or John Smiths children.
The concern is prot, and it doesnt matter who gets killed or injured to
serve this end. Viewed in this way, corporate violence appear as random
acts of violence, a subject that is often newsworthy when the offender
uses a gun, or is easily identied as a given person, because the image
of the latter promotes a myth of crime that encourages focusing on and
fearing the poor offender (Reiman, 2003).
The other excuse for corporate crime is that in law, the corporation
is treated as a person. But, we must recognize that in reality, corporations
do not act on their own. Rather, it takes the conscious activity of individuals to make corporations behave. In other words, behind every bad
109
corporate act or every criminal corporation are individuals who intentionally choose criminal behavior. They can often do so without fear of
punishment, as they hide behind the veil of the corporation.
If, as the above discussion suggests, the purpose of our big prison
system is not the control of crime, nor even the control of only the most
serious crimes in our society, what purpose does our big prison system
serve? A small part of this answer has already been offered: prisons control
the poor. But, to answer this question more completely, we must turn our
attention to the lessons offered by materialist analyses of punishment.
Conc lusi on
This chapter has reviewed some basic data about the American
prison system, detailed its growth, and examined some explanations for
why it has grown so much and so rapidly over the past thirty years. In a
later chapter, further evidence will be offered to support the contention
that the large prison system the United States has created is an ineffective
mechanism for controlling crime both because it fails to reduce crime,
and because any marginal level of crime reduction it may produce is
generated at the expense of excessive nancial costs.
This chapter also noted that, while the American prison system has
become the largest in the world, it fails to incarcerate societys most dangerous offenderscorporate and white-collar criminals. Clearly, then,
this system of punishment presents a legitimation problem of which the
poor are quite aware: they get locked up while the rich use their resources to avoid incarceration, allowing them to proceed unimpeded on
their life course, perhaps even able to achieve the lofty aspiration of occupying the White House.
Chapte r 5
There are a number of different mechanisms for understanding and explaining the extraordinary rate of prison growth experienced in the United States over the past three decades. It is common
to read or hear explanations of the following kinds:
1. Imprisonment responds to crime. In this view, imprisonment expanded in response to a growing crime problem. As the problem
of crime in the United States got worse, imprisonment grew
at an increased rate both in an effort to incapacitate and deter
criminal offenders.
2. Imprisonment responds to public demand. In a democracy, it can be
argued that trends in imprisonment should follow public demand. Public demand models argue that the expanded use of
imprisonment has been fuelled by public attitudes concerning
punitiveness toward criminals. As public punitiveness expanded,
so too did Americas system of imprisonment.
3. Political responses. Similar to the public-demand model, one version of the political model argues that public demands for increased punishment encourages politicians to put in place enhanced crime control, which, in turn, accelerated the use of
imprisonment. A second suggests that political interests may also
reect pressures and incentives that originate in the private sector. This position takes account of political campaign contributions, lobbying efforts, and other forms of industry inuence.
110
111
112
B i g P ri s on s, B i g D ream s
113
labor to edgling capitalist industries, and resocializing recalcitrant workers to accept the new labor regimes that accompanied capitalism (Foucault, 1979). Two centuries later, capitalist nations turned to the prison
as a form of punishment. Under capitalist economic arraignments, the
prison provided several economically necessary conditions: physical control of the surplus population, the potential for the production of goods
without labor costs (especially in the form of the industrial prison), and
rehabilitation and resocialization of marginalized workers through the
use hard work as part of the penal response.
Given its historical emphasis, Rusche and Kirchheimers analysis depended heavily on qualitative evidence. Other researchers in this tradition have used both qualitative and quantitative data to support Rusche
and Kirchheimers contentions. Barnes and Teeters (1945), for example,
applied Rusche and Kirchheimers thesis to an analysis of the accumulation function of industrial prisons during the early part of the twentieth
century. Specically, their quantitative data addressed the growth of the
industrial prison and the monetary value of its output. Foucaults widely
read book Discipline and Punish (1979) provides an analysis of the changing nature and goals of punishment throughout history, and employs a
rich and detailed form of historical-qualitative data. Furthermore, Foucault quotes Rusche and Kirchheimer approvingly in his analysis of the
ideological role of the prison, and in his claim that the prison replicates
the disciplinary order of capitalism.
Beginning in the 1970s, numerous researchers captured the insights
of Rusche and Kirchheimer in analyses that examined the relationship
between trends in unemployment and incarceration. More specically,
this research addressed the marginalization or labor market thesis found
in Rusche and Kirchheimers work. Simply put, this thesis stated that
under capitalism, imprisonment expands to exert greater control over
the economically marginalized segment of the working class. Others extended Rusche and Kirchheimers thesis and examined associations between alternative measures of labor market conditions and their impact
on punishment. Overwhelmingly, this research tended to support the
views of Rusche and Kirchheimer.
The researchers who examined Rusche and Kirchheimers theses were also materialists. Within criminology, these materialists were
typically called radical criminologists (Lynch, Michalowski, and Groves,
114
B i g P ri s on s, B i g D ream s
2000). Radical criminologists not only share a commitment to materialism, they also share a commitment to critiquing and exposing the
relationship between economic, racial, and gender inequality and various
crime and justice issues. As far as prison trends were concerned, radical
criminologists were also sensitive to examining the growing association
between the expansion of the prison system and the rapid growth of African-American rates of incarceration. Beyond these issues, radicals also
share a commitment to social and economic change as a mechanism
for remedying social problems such as crime or growing rates of incarceration. Thus, radical scholars point toward economic solutions that will
help minimize crime rates as well as our reliance on formal social control
mechanisms such as the prison (Lynch and Michalowski, 2006).
At the same time that support for Rusche and Kirchheimers perspective was strengthened, the nature of radical criminology was undergoing
a transformation from within. By the mid-1980s, many of those who
worked within the radical criminological perspective began a movement
to shed this label, preferring to be referred to as critical criminologists.1
The impact of this change in name signaled a redirection in the type of
research the new critical criminologists were to carry out. The majority abandoned research into the association among economic conditions,
class inequality, and crime and justice. Many were drawn into postmodern
views. A signicant and needed focus emerged on race and gender inequality. An important emphasis on working-class victimization and public policy was proposed by left-realists. These new, critical criminological
variations drew increased attention (Lynch and Michalowski, 2006).
Given the broad appeal of this new movement, few researchers were
left to continue the promising line of study suggested by radical or materialist perspectives, such as the one developed by Rusche and Kirchheimer.
In effect, the materialist roots of critical criminology were increasingly
abandoned. As a result, Rusche and Kirchheimers inuential view, which
had mounted a substantial challenge to the orthodox view that prisons
were a useful response to crime, drew less attention, though it remained a
signicant basis for critiques of orthodox explanations of the crime-punishment connection (for an opposing opinion see Sutton, 2004, 2000).
Over the past decade, the pronounced neglect of materialist views on
crime and justice has lead to what could be dened as a crisis in critical
criminological theorizing. For example, with Rusche and Kirchheimers
115
view now tucked away behind the scenes of the postmodern emphasis
common to the critical criminological enterprise, it became incumbent
on critical theorists to develop a new view of the modern U.S. prison
crisis. How would critical criminologists explain American penal practices dened by warehousing and its expanded use of imprisonment?
For the most part, critical criminologists failed to take up this challenge, and the critique of prison expansion that was developed came primarily from liberal criminologists (Austin and Irwin, 2003; Clear, 1994).
To be sure, a few researchers have continued to examine these issues
employing a materialist framework (Greenberg and West, 2001; Greenberg, 2001a), while others have turned to race-based explanations, or to
research emphasizing the association between the expansion of imprisonment and the war on drugs (Austin and Irwin, 2003; Mauer, 1999;
Welch, 1999, 2004a, 2004b). And, to be sure, these views have helped
us understand at least some of the problems behind the growing rate of
incarceration in the United States. Still, it is my assertion that the contemporary prison problem we face cannot be addressed without a strong
materialist theoretical base. Below, I clarify my position and rectify the
neglect of materialist explanations of the modern problem of rapidly
expanding rates of imprisonment in the United States.
Mate rialism and Imprisonme nt in the Unite d State s
In 1973, the rate of imprisonment in the United States began its
long-term climb. While the growth in the U.S. rate of imprisonment has
slowed in recent years, the number of inmates in U.S. prisons nevertheless reached record highs. How much did imprisonment grow during the
thirty years following 1973? And how is this growth associated with economic conditions? These trends were reviewed in an earlier chapter, but a
brief summary is useful for purposes of the discussion that follows.
In 1973, the rate of imprisonment per 100,000 citizens in the United
States stood at 96. What was interesting about 1973 was that the rate of
imprisonment that year was lower than the average rate of imprisonment
from 1929 through 1967. From 1929 through 1967, the lowest rate of
imprisonment, 98 inmates per 100,000 U.S. citizens, was recorded in 3
separate years1929, 1945, and 1967.
It is important to understand the historical trend of the rate of imprisonment in order to appreciate the current level at which we imprison
116
B i g P ri s on s, B i g D ream s
criminals. As reviewed in the previous chapter, reliable gures on imprisonment rates date to 1925. In that year, the rate of imprisonment was 79
per 100,000 population. By 1929, the rate had risen to 98 per 100,000, a
rather substantial increase of 24 percent spread out across 5 years. From
1929 through 1939, the imprisonment rate continued to climb, from 98
to 137 per 100,000 (although there was a slight decline and a steady
state in 1933 and 1934). The increase from 1929 through 1939 was
nearly 40 percent, though the increase was spread out over a decade.
Furthermore, this increase occurred during a period of adverse economic
conditions marked by high unemployment rates that followed the Great
Depression. From 1940 through 1945, however, the rate of imprisonment
declined to its 1929, pre-Depression level of 98 per 100,000a decline
of more then 28 percent in 5 years. This decline in the imprisonment rate
corresponded with enhanced economic performance of the U.S. economy linked to wartime production. During this period, unemployment
declined sharply, the result of employment opportunities in war-related
industries and the effect of the draft on the pool of available workers.
At the end of World War II, the rate of imprisonment once again began to ascend, probably in response to labor market shifts caused by the
return of war veterans into the economy and the tightening of the labor
market and economic opportunities. By 1950, the rate of imprisonment
had reached 109 per 100,000. This was only a slight (11 percent) increase
since 1945. The imprisonment rate remained relatively steady through
1956 (112 / 100,000). Over the next several years, the rate of imprisonment rose only slightly, to 119 per 100,000 in 1961. Overall, the imprisonment rate expanded only 9 percent from 1950 through 1961.
From 1962 through 1972, the imprisonment rate declined during a
period of economic expansion and increased employment opportunities.
By 1972, the imprisonment rate reached its lowest point since 1927 at 93
per 100,000. During this short period (19611972), the imprisonment
rate declined by nearly 22 percent.
By the mid-1970s, however, the United States began to experience
an economic recession and associated increases in the unemployment
and interest rates. This period of economic difculty was mirrored by
growth in the U.S. prison system. From 1973 through 1979, the rate
of imprisonment expanded from 96 per 100,000 to 133 per 100,000,
nearly 39 percent. The importance of this latter period is that it would
117
mark the last time that there was seemingly full concordance between
trends in the economic system and trends in the imprisonment rateat
least from the traditional materialist view extracted from Rusche and
Kirchheimers work. That is, the economic recovery experienced during the late 1980s and 1990s did not have the expected impact on the
incarceration rateat least not the outcome that would be predicted by
relying on the traditional interpretation of the Rusche-Kirchheimer thesis, which would result in a declining rate of imprisonment. Specically,
using the Rusche and Kirchheimer labor market model that had been
developed during the 1970s (e.g., Greenberg, 1977), we would predict
that the upward trend in employment that emerged in the recovering
economy of the late 1980s and 1990s should cause the rate of imprisonment to decline. But the rate of imprisonment did not decline; it continued to rise in the face of expanded economic and labor market opportunities. This new pattern of continually expanding rates of imprisonment
during a period of economic recovery became the crux of a dilemma for
radical criminologists, one that they failed to explain, and which the new
critical criminology failed to tackle as well (for an exception and discussion, see Michalowski and Carlson, 1999).
By 1980, the rate of imprisonment had reached 139; by 1990 it was
297 per 100,000; and by 2000 it was 478 per 100,000. In terms of percentage increase, the imprisonment rate expanded nearly 114 percent
from 1980 to 1990, and by 60 percent from 1990 through 2000. For
the entire period (19802000), the increase in the rate of incarceration
was over 230 percent, while the increase in the raw number of inmates
incarcerated was nearly 300 percent. During this period, the U.S. correctional system surpassed several milestones. The number of inmates
topped 500,000 during 1987; the 750,000 level was passed in 1992; and
the 1 million mark became history in 1995.
There is something deeply troubling in these gures. Why did the
rate of incarceration expand from 79 per 100,000 in 1925 to 478 per
100,000 by the year 2000? What would cause the rate of incarceration
to go up by over 500 percent during this period? These are interesting
questions that are beyond the scope of the limited investigation taken
up here. The concern here is with the latest period of incarceration and
the reasons for such a rapid rise in the rate of incarceration in such a
short period of time, from 1973 to the present. Let us also not forget
118
B i g P ri s on s, B i g D ream s
that another point of this work is to answer the following question: why
did the Rusche-Kirchheimer hypothesis suddenly stop making sense?
Before continuing, the failure of the traditional Rusche-Kirchheimer hypothesis in modern circumstances demands some minor additional comment.The sudden inability of this thesis to explain a process it
seemed capable of tackling for the centuries of data it had been applied
to speaks to some of the differences between social science theories and
natural science theories. On the one hand, the theories of natural sciences tend to deal with behaviors (if we can be allowed, for instance,
to call how two chemicals react with one another a behavior) that do
not change over time. As long as scientists know and control the surrounding environment (e.g., pressure, temperature, contaminants), the
chemicals they are studying will react in the same way in the year 1850
as they do in the year 2000. If they did not, then very simple things we
rely upon everyday (from our coffee pots to our computers) would not
work as expected. Human behaviors are not so predictable because they
occur in an environment that is not as controlled as the one in which
the scientists chemical reactions occur. Economic systems change; social
relations change; political systems change; neighborhoods change; family relations change, and so on. Sometimes these changes in the human
social, economic, and political environment are so great as to require
new explanations.
U p, Up, and Away: Im p ri s onm e nt i n th e 19 9 0s
As noted, the imprisonment rate became greatly inated by the
1990s.The widely accepted explanation for this persistent growth of imprisonment suggests that conservative and punitive attitudes expressed
by the public toward criminals caused politicians to adopt a get tough
on crime approach (or even the other way around). Having heard their
constituents, the exact meaning of the idea of getting tough was left
up to politicians to determine. Politicians either thought or knew from
experience that if they did not seem tough enough on crime, their
constituents would be unhappy. Politicians responded in a variety of
ways. President Bill Clinton, for instance, pledged money toward hiring new police ofcers. Legislators responded by passing enhanced sentencing laws that included widely adopted three strikes youre out
and other career-criminal sentencing alternatives, as well as enhanced
119
penalties and lesser standards for juvenile waivers to adult courts. At local levels, police chiefs, pressured by politicians to do something about
crime, turned to zero-tolerance strategies to increase arrests. In addition,
a more global war on drugs continued to attract attention, especially at
the federal level.
To a large extent, these efforts were misguided in that they relied
more heavily on political pressures than on knowledge concerning the
impact of more severe punishments on crime rates. This effort can also
be considered misguided to the extent that the US was not a society
that could be considered soft on crime. For example, by the 1990s, the
United States already had one of the highest imprisonment rates in the
world. In addition, the United States was also among the worlds leaders
in terms of average prison sentence lengths (Welch, 2004b). Despite an
already-tough approach toward crime, however, the United States was
also among those countries with high rates of criminal offendingespecially when compared to other modern industrialized nations, and
especially with respect to violent crimes. In spite of these facts, the hope
was still held that sending even more people to prison would lower the
rate of crime. To be sure, at some point, such a strategy must at least appear to work, and at some point the cycle of crime and imprisonment
would be aligned. This alignment occurred during the 1990s. But what
percentage of the population must be locked up to make imprisonment
an effective crime control strategy? And, at what price, both in terms
of human and nancial costs? I will return to address a portion of this
question later.
While there were visible and empirically based reasons to doubt that
a get tough approach would reduce crime, or that this was a major
reason other nations had low rates of crime (see Farrington, Langan, and
Tonry, 2004), U.S. policy makers pushed the nation further in this direction. The public, spurred on by bad news about rising crime rates and
political rhetoric about crime and punishment, continued to shift to the
right in terms of their opinion about the solution to crime.
In part, the political response to crime in the 1980s and 1990s reected both common sense and the desire to do something about crime.
These policies, however, did not necessarily make sense from the perspective of what is known about the relationship between crime and
punishment. Consider, for instance, that the crime rate in the United
120
B i g P ri s on s, B i g D ream s
States had risen annually from 1985 through 1991. But so too had the
rate of incarceration. And then a funny thing happened: crime began to
fall as sentence lengths increased, imprisonment rates rose, and tougher
new crime control policies were implemented. What was funny or odd
about this circumstance was that crime had not fallen during the previous decade when sentence lengths were expanding and imprisonment
rates were growing each and every year. But, all of a sudden this strategy
appeared to work, as if some magic threshold had been crossed. And
crime didnt fall just once: it fell over and over again from 1991 through
1999. Whether through deterrence or incapacitation, or for some other
unexplained reason, crime was falling. This association allowed proponents of get tough legislation and prison expansion to argue that
their approaches were working to reduce crime. Prison expansion advocates took credit, but so, too, did sponsors of other get tough approaches, such as enhanced police enforcement, zero-tolerance policies,
and three-strikes and career-criminal legislation. Since no one could
pinpoint which of these policiesif anywas the real cause of the decline in crime, and since the decline in crime was so precipitous (27
percent overall from 1991 to 1999), there was more than enough credit
to spread around.
To be sure, there was something different about the relationship between crime and punishment in the 1990sor at least there appeared
to be something different. The visible difference did not go unnoticed
by academics, in particular Raymond Michalowski and Susan M. Carlson. Employing Rusche and Kirchheimers thesis as their starting point,
Michalowski and Carlson (1999) examined the relationship between
unemployment rates and imprisonment rates (and in a separate analysis,
crime; see Carlson and Michalowski, 1997) over a long historical period
(four decades). They discovered that the relationship between unemployment and incarceration was historically contingent. During this period,
they argued, the relationship between unemployment and incarceration
changed. While incarceration initially appeared to be a response to controlling marginalized populations, as Rusche and Kirchheimer suggested,
this relationship appeared to be contingent on other economic factors
as well. In particular, Michalowski and Carlson noted that the relationship between unemployment rates and incarceration rates since 1933 was
conditioned by social structure of accumulation (SSA) effects. Employing
121
research by economists, Michalowski and Carlson suggested that economies passed through different stages of accumulation (SSAs). These stages
(expansion, consolidation, and decay), when taken together, form a long
swing. But, during each long swing of the economy, specic phase characteristics could come into play that affected the relationship between
unemployment and imprisonment. Their analysis supported their hypotheses of: (1) a weak relationship between unemployment and imprisonment during the exploration phase of the economy from 19331947;
(2) a strong positive relationship between unemployment and imprisonment during the economic decay period, 19671979; (3) a weak or
nonexistent relationship between unemployment and imprisonment for
the 19481966 period of economic consolidation, which is marked by
economic growth and industrial contraction; and (4) an inversed relationship between unemployment and imprisonment from 19801992 during
the new expansion stage of an SSA based on a completed transition from
an industrial to a service economy.
Carlson and Michalowskis ndings made it clear that applications
of Rusche and Kirchheimers thesis did not form an ironclad law such
as the ones found in the natural sciences, and that the relationship between employment and incarceration was conditioned by other economic factors. This nding, which had been suspected for some time,
was a blow to well-entrenched radical explanations of penal trends that
drew directly on Rusche and Kirchheimers labor market hypotheses,
especially those that converted the broader implications of Rusche and
Kirchheimers theory into a narrow theory of the impact of unemployment on incarceration. Carlson and Michalowski did not, however, leave
radicals hanging. They offered an alternative explanation based on social
structures of accumulation and long cycles theory. To be sure, this was a
useful explanation because it demonstrated why imprisonment did not
respond to unemployment in the same way in every era of economic
development. The argument suggests that during this four-decade period, the U.S. economy passed through various stages of economic contraction, expansion, and exploration, which affected how imprisonment
would be used to control economically marginalized groups. To be sure,
this was a noteworthy contribution to the literature, and one way of
amending Rusche and Kirchheimers argument. Below, I review some
of the economic conditions during this period, and offer an additional,
122
B i g P ri s on s, B i g D ream s
123
have been able to use technology to maintain prot margins, they have
failed to consider the cost to the working class in terms of jobs lost.
American workers are often criticized for their lack of productivity. Moreover, the lack of productivity argument is often used to justify
moving plants and facilities overseas, or serves as the basis for downsizing
and other prot-enhancing schemes that take a slice out of the already
shrinking job market in America. Table 5.1 demonstrates that declining
worker productivity claims are false. Here we see productivity indexes
calculated by the U.S. Department of Labors Bureau of Labor Statistics for
a selection of industries for which data are available from 1950 or 1960
onward. For all industries, with the exception of millwork, productivity
gains have been substantial. It should be noted, however, that each of the
industries shown in table 5.1 has experienced a productivity gain, and even if
the gain cannot be considered substantial, it still indicates that American
worker productivity has, contrary to widely publicized claims, grown.
Another important market force during this period was deregulation. In theory, deregulation increases competition within an industry
and thereby forces improvements in efciency that should lead to a lower
price structure. In the words of Barlett and Steel (1996:106), Removing
government restrictions on the private sector would let free and open
competition rule the marketplace. Getting rid of regulation would spur
the growth of new companies. Existing companies would become more
efcient or perish. Competition would create jobs, drive down prices
and benet consumers and businesses alike. In the real world, such rosy
outcomes are less apparent.
But, as this quotation implies, this happy scenario has not been evident. In terms of pricing, for instance, consider the case of electricity
deregulation in California. In 1997, the year before deregulation of energy providers in California, private providers generated 35 percent of
Californias electricity. By 2002, private providers share of electricity
generation expanded to nearly 60 percent. During this same period,
as public electricity generation was privatized, the price per kilowatt hour of
electricity grew by 27.4 percent. To be sure, this rise in price contradicted
the claim that privatization is benecial to the public to the extent that
end price or costs are reduced. Furthermore, consider that at the same
time that private electricity generation in California forced a rise in cost,
the nationwide average price per kilowatt hour (excluding California) remained
124
B i g P ri s on s, B i g D ream s
Table 5.1
U.S. Department of Labor, Bureau of Labor Statistics, Industry Specic
Indexed Output Per Worker, Select Industries with Long Term Data,
19502000
Year
Industry
1950
1960
1970
1980
1990
2000
% Gain
Flour/Grain
29.3
43.0
64.9
93.4
106.0
151.5
417
Malt Beverages
12.6
16.5
30.3
63.6
106.4
121.6
86
Glass Containers
53.8
54.9
71.3
85.2
109.2
156.6
191
Copper
(Primary)
22.2
23.8
29.2
35.7
116.9
163.2
635
Aluminum
(Prim.)
32.6
54.6
60.8
73.4
104.5
114.9
252
Metal Drums,
Pails, & Barrels
37.4
46.5
57.6
74.0
117.8
170.5
356
Petroleum
Pipelines
13.3
29.9
75.2
88.8
102.0
141.4
963
Coal Mining
42.4
74.9
59.5
123.6
219.0
417
Milk, Liquid
29.1
43.4
74.2
106.0
111.0
281
Soft Drinks
35.9
42.3
66.2
127.4
169.1
371
Sawmills/Planing
39.9
60.2
67.0
98.8
128.6
222
Millwork
81.2
105.4
93.3
95.8
93.0
15
*Information in this table was extracted from the U.S. Department of Labor, Bureau of
Labor Statistics les, Industry Labor Productivity and Costs Tables, Output per employee subseries: ftp.bls.gov/pub/special.requests/opt/dipts/dipts/iprsicdata.txt.
125
126
B i g P ri s on s, B i g D ream s
127
much more intensely affected than other areas. For example, Detroits
economic base was devastated by the contraction of manufacturing, and
jobs were lost to both intense mechanization of the automobile industry
and the movement of auto-manufacturing plants to others states and
countries. In some respects, the crisis presented by deindustrialization
appears locally at the city or town level, while it is an indicator of a larger
structural crisis (see Mills, 1959, for discussion of how individual or local
troubles reect larger social problems).
As one might expect, crime rose during this economic transformation as a growing percentage of the U.S. work force was displaced or
marginalized by the shrinking manufacturing base. Two caveats must be
noted about this general statement.
First, the relationship between economic marginalization and crime
is not easily observable to the extent that economic marginalization is
128
B i g P ri s on s, B i g D ream s
often inappropriately measured as a simple unemployment rate. Economic marginalization extends beyond unemployment, and can include
partial and seasonal employment, as well as forms of economic exploitation found in wage-based economies where class membership denes
owning and laboring options, as well as broad shifts in salary structures
associated with the expansion of the service economy. To address these
concerns, scholars devised empirical methods that measure marginalization in ways that are more consistent with the unique theoretical anchoring of radical criminology (as opposed to the theoretical anchoring
of traditional, mainstream, or orthodox criminology; for discussion of
these methods and examples of their empirical application in criminology see Lynch, 1987, 1988; Lynch, Groves, and Lizotte, 1994; Lynch,
Hogan, and Stretesky, 1999; Nalla, Lynch, and Leiber, 1997).
Second, crime did not rise during the entire period under discussion; in fact, as noted earlier in this book, crime began to decline in the
1990s. Interestingly, the decline in crime in the 1990s came at a time where
employment opportunities had expanded and rates of wealth concentration slowed
(Wolff, 2002). The relationship between crime and unemployment is not
uniform, linear, or always in the expected direction (e.g., as unemployment goes up, crime will increase; Michalowski and Carson, 1999). Again,
this may be due to the fact that unemployment and economic marginalization are two different concepts that require different measures.
We must also consider what happened to the imprisonment rate.
Following the predictions based on the Rusche-Kirchheimer hypothesis,
we would expect the imprisonment rate to rise along with economic
marginalization. The more specic hypothesis found in the research literature, however, links Rusche and Kirchheimers model to unemployment rates, which leads to the hypothesis that unemployment and imprisonment rates should uctuate together. By the mid- to late 1980s, as
the economy began to recover from the high unemployment rates of the
mid- to late 1970s, one would expect, therefore, that the imprisonment
rate would decline. But, it did not; it continued to rise. How can this
apparent discrepancy be explained? In part, the mild economic recovery
in the 1980s was fuelled by an expansion of the service industry and the
renewed opportunities for employment this sector generated. To be sure,
the erosion of the manufacturing base was somewhat offset by growth in
service sector employment, illustrated in gure 5.2.
129
Nevertheless, and contrary to expectations, despite enhanced employment opportunities, crime continued to rise through the late 1980s.
In part, this rise in crime can be traced to the lower standard of living
among the working class. That is, while expansion of the service sector
created new jobs, the jobs that were created were lower-paying positions. It could be argued that the reality of these jobs did not meet
social expectations. Furthermore, the wage decline associated with the
transformation from an industrial to a service economy indicates that a
very broad form of economic marginalization was affecting even employed workers.
During the Clinton administration, the economy began to grow at
an accelerated pace. This economic upsurge was, to a large degree, more
imaginary than real, beneting upper-income groups through the development of the high-tech computer industry and the proliferation of the
dot.com market place. During this period, class inequality increased as
130
B i g P ri s on s, B i g D ream s
131
132
B i g P ri s on s, B i g D ream s
repeat customers. To take one example, consider the food service sector.
Most people eat three times a day, a considerably shorter consumption
cycle than for automobiles (ve years) or washing machines (ten years).
An automobile is consumed or purchased once every 5 years, or, on
average every 1,825 days, meaning there is only one opportunity in this
period, on average, to make an automobile sale to each person. In the
same 5 years, a full-service food vendor (breakfast, lunch, and dinner) has
the opportunity to make 5,475 potential sales to one customer during
the same period (at 3 meals per day). Even if this person eats out only a
few times a week, and only one-half of those times at the same location,
there would be more than 150 potential sales available to a food service
provider over 5 yearsper person.
From the vantage point of the capitalist, this unique replacement
characteristic of service-sector commodities provides both advantages
and disadvantages over the ordinary commodity. The advantage of service-sector commodities, as noted, is in the need for relatively frequent
replacement in some cases (e.g., lawn care service), or a commodity with
a relatively long-term inelastic demand (e.g., food) when compared to
the ordinary commodity. The disadvantage for the service-sector commodity is restricted to certain segments of the marketplace that have
uctuating short-term demand. In response, in some service areas, the
service-sector provider may enter into exible labor contracts capable
of adjusting to uctuating demand (e.g., temporary service provision).
Nevertheless, such contracts are not always possible, and the service-sector capitalist may have to pay employees even when there is no service
to provide.
Given these constraints, what the service-sector capitalist desires is
a market for the services she sells that is inelastic in the short term. The
mantra of capitalism, however, is to accumulate, and moreso to expand
capital accumulation. Thus, the service-sector capitalist desires more than
an inelastic markethe desires an expanding marketplace. To be sure, in
recent decades, the prison apparatus has provided one such marketplace.
Capitalists have invented a host of businesses that take advantage
of the short replacement cycle and long-term inelastic nature of demand that characterizes some service-sector commodities. A number of
these businesses sell their commodities to those who run the growing
U.S. correctional system. These services include, but are not limited to:
133
134
B i g P ri s on s, B i g D ream s
135
and the estimated percentage of the population falling within the poorest one-fth in each state can be regressed on the change in the rate of
imprisonment. Using Gainsborough and Mauers estimates of the change
in imprisonment rates by state for 19911998, and controlling for the
rate of change in the total UCR crime rate, the upper-income level for
the poorest one-fth of the population by state is a signicant predictor (p = 0.033; t = -2.203) of the change in imprisonment rates. The
percentage of those falling into the poorest one-fth is not signicant.
This may be due to the fact that this percentage varies little across states
(the majority of states [N = 45] have between 41 and 49 percent of their
population in this category). The results for the negative relationship between the upper income level for the poorest one-fth of the population
and the change in imprisonment, however, lends some evidence favoring
the view described above.
The P ri son as I ndu stry,
Soc i al Cont rol as C om mod i ty
In order to understand more fully what happened to the longstanding association between the labor market and imprisonment during the
contemporary period of American capitalism, it is necessary to recast the
prison as industry, or at least, as Nils Christie described it, as part of the
crime control industry. Understanding the prison as industry also requires that we take into account the fact that the American economy was
transformed during the 1970s from a manufacturing to a service basis.
The demise of manufacturing and the rise of the service-sector
economy has important implications for the prison as industry view
and can serve as the basis for interpreting the association between contemporary trends in imprisonment and economic factors and conditions. To do so, we must also pay attention to the arguments set forth
by James OConnor in his 1973 book, The Fiscal Crisis of the State. It
was here that OConnor argued that the state had become part of the
accumulation process. In effect, the scal crisis experienced by the state
was connected to the states role in stimulating the economy at best, and
minimally by enhancing the ability of the capitalist sector to enhance its
accumulation potential.
In OConnors view, the state can fulll its role in accumulation
either indirectly or directly. The state meets its accumulation function
136
B i g P ri s on s, B i g D ream s
indirectly by guiding the economy in an effort to improve the circumstances for capital accumulation and economic expansion (e.g., the common practice of manipulating the interest rate). The state, however, can
also facilitate accumulation directly by spending current reabsorbed
wages acquired from taxation (or future reabsorbed wages through debt
spending) in ways that transform those wages into payments that expand the available pool of potential surplus value (as surplus value is
the source of prot; see Marx, 1974). The privatization of state services,
which expanded tremendously during the Reagan administration, provides a useful example of the latter strategy. The importance and impact
of the privatization of governmental responsibilities on economic trends
should not be overlooked. Through privatization, billions of tax dollars
can be redirected to the private sector when the state is forced to purchase rather than provide services. The timing of this effort should also
not be overlooked, because the Reagan-initiated privatization movement
corresponds with the prolic expansion of the U.S. prison system, as well
as the expansion of the entire private prison apparatus.
One of the key features of privatization is that it aids in the transfer
of capital from the state to the private sector. In effect, privatization allows the state to expend money it has collected through taxationwhich
it had previously used to provide governmentally sponsored, nonprot
serviceson private, for-prot service providers. Before privatization,
the services normally provided by the state were treated as social expenses, and they were viewed as necessary state expenditures that promoted social welfare. Once privatization was implemented, however,
the state becomes a middle man in the service-provision chain, selecting service providers to replace some of its functions. That is, payments
previously made to service workers hired directly by the government
(i.e., state workers) are now made to capitalists who absorb part of the
previous social expense bill as prot before paying workers to provide
contracted services for the state. Evidence of this type of service contact
can be found in most U.S. states. In Florida, for example, where this
practice is widespread, numerous juvenile detention and rehabilitation
programs are operated by private service providers rather than by the
state directly.
Under privatization, the capitalist acting as a service provider gets a
cut of state social expenses to fulll the role previously occupied directly
137
by state agents or state employees. This service provision fee, which may
be much larger than the salary paid to state employees, diminishes the
pool of money the state has available to be spent directly on services.
Nevertheless, the private sector, conservative politicians, and social commentators hold out privatization as a more efcient mechanism for service provision (see data above on privatization, pollution, and electricity
generation in California).
Indeed, one of the primary legitimation claims made about privatesector service providers is that they are more economically efcient
than governmental services, and therefore the private sector would be
able to provide services at a lower cost than government, hence reducing the overall tax liability to the public. What was overlooked in
this argument was the fact that privatization either depresses wages of
workers in affected areas and increases the exploitation of laborers (if
we employ an objective denition of exploitation), or it increases unemployment by cutting the number of workers privatized service employers hire to provided a service. Moreover, other important issues
are left out of the privatization debate. For example, where criminal
justice or other social services are concerned, one important but omitted discussion centers on the legitimacy of allowing private providers
to supply services needed to promote social welfare. In addition, no
one suggests that the costs of services could be lowered even further if
the government placed a limit on the amount of prot capitalists were
allowed to extract from service provision agreements that are, after all,
designed to benet the public good. While these types of arguments
are of interest to radicals, they must be put aside here to return to the
primary question being pursued, namely: How, exactly, does the prison
industry t into this argument?
Before we can fully appreciate the prison as industry view, we must
rst alter our understanding of the term industry. In traditional usage,
we equate industry with manufacturing or the production of tangible
commodities that can be sold in order to transfer value and stimulate
capital accumulation. In service industries, however, there are no tangible commodities that embody dead labor values. Rather, what the
service sector offers for sale is future or potential labor that provides no
direct value in the traditional sense. Service labor, however, may be put
to a variety of uses, some of which establish a climate that enhances the
138
B i g P ri s on s, B i g D ream s
139
140
B i g P ri s on s, B i g D ream s
that was directly paid to the private correctional sector for services. An
estimate can be made by reviewing what we know about existing correctional service contract providers.
In 2000, there were thirteen private correctional management rms
with active contracts in the United States, which housed more than
120,000 inmates. The largest of these contractors, Corrections Corporation of America (CCA), housed more than 55 percent of the inmates
(67,000) in 63 facilities located in 19 states and the District of Columbia in 2005 (http://www.correctionscorp.com/aboutcca.html; Sourcebook of Criminal Justice Statistics, table 6.26). If CCA were treated as
an independent correctional system, it would be the fth-largest correctional system in the United States behind California, Texas, Florida, and
New York. In 2003 CCA earned more than $1 billion in revenue, a 12.5
percent increase over 2001 revenue (see CCAs 2003 year-end nancial report, http://www.correctionscorp.com). At the same time, CCA
expanded its employment by only 1 percent. CCA is large enough to
serve as parent corporation to Transcor America Incorporated, a prison
transportation service provider. The big business behind CCA was most
noticeable when it was added to the New York Stock Exchange in 1994
(ticker symbol CXN). CCAs own calculations of potential protability
based on man-day revenue and xed and variable expenses leaves a
prot operating margin of 25.3 percent (see, CCA, 2005: 20), indicating
that crime pays private correctional service providers quite well.
As noted above, it is impossible to estimate the exact amount of
Americas $50 billion private-sector correctional budget. Considering
that CCAs piece of the correctional pie is a little more than 2 percent,
and that there are many other types of private servicesfrom food to
laundry servicesthat correctional systems purchase, a conservative estimate that 25 to 30 percent (or $12.5 to $15 billion) of prison expenditures are directed toward the private sector does not seem unreasonable.
At this juncture, it is useful to recall OConnors argument about
the connection between government expenditures and the states role
in providing conditions conducive to the expansion of opportunities
for capital accumulation for the capitalist sector. As noted, one means of
doing so is to privatize social expenditures so that they are now made
to private businesses that carry out service-related tasks previously
undertaken by the government. It should be clear from the previous
141
142
B i g P ri s on s, B i g D ream s
the increase in the imprisonment rate, the imprisonment bill rose twice
as fast, even tough the prison system was also being privatized to keep
down costs. Thus, over time, locking up more people is not simply more
costly because we now have larger prison populations. It is also more
costly because the costs of doing so rise more quickly than the rate of
incarceration. In fact, taxpayer per capita correctional costs grew at such
a steady pace that they exceeded the growth of imprisonment, measured
by the number of inmates, beginning in 1984. The gap between the two
has expanded slowly since.
The correctional tax bill has implications beyond its cost to individual taxpayers. From a materialist perspective, taxes are important for
two reasons. First, as OConnor (1973, 203) argued, Every important
change in the balance of class and political forces is registered in the tax
structure. As noted earlier, a key element of the materialist perspective
on imprisonment is that imprisonment has become part of the class war
in modern America, and, as Ronald Goldfarb (1969) once argued, prisons are Americas poorhouse.
Class warfare is not always directly visible. In earlier times, class warfare was evident in strikes (late-nineteenth, early-twentieth century) or
in protests against corporatism (1960s). Often, however, class warfare is
hidden within the normal operation of state mechanisms such as the
criminal justice system, which primarily focuses on the crimes of the
lower classes (Lynch, Michalowki, and Groves, 2000; Reiman, 2003).
Class warfare is also evident in various forms of economic inequality.
Economic inequality doesnt just happen: it is often the result of specic
policies that affect taxation, investments, the business growth cycle, interest rates, and a host of laws that have unequal impact on people from
different class positions (Chambliss and Seidman, 1982).
The expansion of economic inequality in America is one measure
of both inequality and class conict. We commonly assume, and are often told by politicians, that there is less economic inequality today than
in previous eras. Governmental data, however, indicates that this is not
accurate. According to U.S. Census Bureau data on income distribution
(the GINI index or index of income concentration), for example, income inequality between the 20 percent of households that receive the
lowest income and the 5 percent that receive the greatest share of income has increased substantiallyby more than 25 percentsince 1968.
143
144
B i g P ri s on s, B i g D ream s
who noted that the relationship between unemployment and incarceration trends are conditioned by social structures of accumulation.
This chapter elaborated on how the U.S. economy changed during the 1980s and 1990s, and how these changes could be related to
the continued expansion of the American prison system. Important
considerations involved deregulation, globalization, and worker productivity. Each of these economic changes expanded economic opportunities for the wealthy, ushering in a period of economic growth
and job gains. These gains, however, were class-linked, and at the same
time that those with wealth saw their fortunes increase, those at the
bottom of the class hierarchy experienced diminishing returns, even
though their prospects for employment increased. That is, while visible unemployment declined, the form of employment shifted from
manufacturing to service-sector employment, with its lower wages and
reduced benefits. While more people were employed, they remained
economically marginalized. And, while more people were employed,
the incarceration rate continued to rise. This relationship appeared to
contradict the Rusche-Kirchheimer model and, most especially, the
empirical tests of that model that linked incarceration trends to unemployment levels. But, in the new U.S. economy, employed people were
being marginalized by an expansion of the service sector, a contraction of manufacturing, globalization, increased worker productivity,
and deregulation. In essence, the simple materialist assumption that the
unemployment rate was an appropriate measure of economic marginalization, and not the general assumptions of the Rusche-Kirchheimer
view itself, were questionable.
This could be illustrated, for example, by examining the relationship between a different means for measuring marginalization and prison
trends, such as the percentage of the population earning incomes that
placed them in the lowest income bracket.
In addition, in the new service economy, the prison itself became
a service industry, and adapted to the new, open market by expanding
the extent of private-sector participation. The interests of private sector capitalists help drive prison expansion through several mechanisms
discussed in this chapter. This lead to the private prison sector becoming
a multibillion-dollar industry that depended on an expanding marketplaceprisons and prison populationsfor its prot. In short, not only
145
Chapte r 6
Prison Effects
Who Gets Locke d Up
Prison Effects
147
While academics and policy makers will continue to argue about the
effectiveness of imprisonment as a crime control strategy, there continues
to be a need to examine the impacts of continually raising the rate of
incarceration in the United States. We know, for example, there are many
more Americans in prison today than in 1973. Thus, in addition to asking
how an increased rate of incarceration affects crime, we also need to ask:
Who are these Americans who are now locked safely away behind bars?
And, how has our imprisonment binge affected the characteristics of the
prison population? We examine this issue in the sections that follow.
Who I s i n A m e ri ca s P ri s on s ?
How has the imprisonment binge affected the characteristics of the
population of Americans who are locked up in the prison system? If we
follow the logic of deterrence and incapacitation arguments, we would
hypothesize that the effort to control crime by locking up more criminals should result in the incarceration of more serious offenders, and that
the characteristics of the prison population would change to reect this
emphasis on serious crime. Are the people who have been locked away
in Americas prisons the most serious criminals in society?
This is a complex question to answer, in part, because it requires
acknowledging who is not locked up in prison. As noted earlier, an important criminal element omitted from Americas prison system includes
white-collar and corporate criminals. Because this group is omitted from
prison, we cannot use them to assess whether our nations imprisonment
binge changed the characteristics of inmates housed in prison. We could,
however, argue that it would be useful to examine whether there were
any long term changes in the level of white collar and corporate crime
and the rate at which these offenders were incarcerated. Unfortunately,
there is no useful, long-term database on white-collar and corporate
crime that details the characteristics of these kinds of criminals or the
number of crimes they have committed over the time period we wish to
study. Thus, we are largely stuck with a focus on the changing characteristics of inmates in American prisons.
Before we can begin that investigation, we must admit that it is difcult to precisely measure the characteristics of prison inmates over the
past thirty years because data on the characteristics of prison inmates is
not available for the entire period. Further, because the imprisonment
148
B i g P ri s on s, B i g D ream s
binge covers such a long time period, it would be difcult to track the
many changes that have occurred in the characteristics of those who
are behind bars in America. To make this task more manageable and to
reect more reliable measures of the characteristics of the prison population, the analysis presented below is restricted to more recent years. For
some parts of this analysis, the time period is restricted to 1991 through
1997; for other portions, we will begin in 1985. These years were selected due to data limitations, and because these more recent periods
mark the greatest growth in Americas prison system.
Offense Type
One method of examining the characteristics of the prison inmate
population is to look at the crimes for which inmates are committed.
One reason to examine offense types over time is to determine whether
a change in imprisonment philosophy affects the type of inmates who
end up in prison.
Previously, it was noted that imprisonment trends in the 1990s were
driven by a strong reliance on the assumptions of the social defense or
incapactitative model of imprisonment, which is based on the idea that
locking up dangerous, repeat offenders will lower the overall rate of
criminal offending by removing known criminals from society. Thus, for
this model to work as advertised, we would expect to see an increase in
the percentage of the prison population that was incarcerated for repeat
offending or had a record of serious violent or property crime.
Violent Crimes. In 1991 violent offenders, or those sent to prison
for murder, rape, robbery, kidnapping and assault, composed 46.6 percent
of state prison populations in the United States. Over the next six years,
under the leadership of policy makers and politicians who endorsed the
social defense model, the rate of imprisonment grew by 42 percent, or
by an average of 7 percent per year. This rate of growth in the prison
population was fairly substantial. During this period of substantial prison
growth, the number of inmates incarcerated for violent offenses grew
by 41 percent. In other words, the rate of growth of incarceration was
slightly greater than the rate of growth in the percentage of inmates
sentenced to prison for violent offenses. There was some variation in the
change in incarceration by type of crime.
Prison Effects
149
150
B i g P ri s on s, B i g D ream s
Prison Effects
151
incarceration appeared to lower crime, and that for the period between
1973 and 1999 this outcome occurred only half the time. This means
that there are a number of other points in time for which we could
compare crime and imprisonment data that would not seem to support
deterrence or incapacitation arguments at all, where the relationship between the number of each type of offender incarcerated would appear
to have very different effects on crime reductions (or increases) for each
offense type.
Let us also remember that to this point we have only examined the
relationship between crime and imprisonment for violent crimes. Let us
now turn our attention to other offenses to see how the deterrence and
social defense views fair.
Property Crimes. From 1991 through 1997, there was a decline in
the percentage of the prison population incarcerated for major property
crimes (burglary, larceny, motor vehicle theft, arson, and stolen property), from 24.4 percent in 1991 to 21.5 percent in 1997. The number
of these crimes reported to police also fell during this time period. In
other words, while there were more people in prison for property crimes
in 1997 compared to 1991, the percentage of the prison population represented by property offenders declined, and the number of property
crimes was also reduced. These data indicate that the emphasis on locking up more criminals did not lead to a greater representation of major
property offenders among the prison population. To be sure, numerically, there were more property offenders locked up in 1997 compared
to 1991; but the percent of offenders they represented declined. Let us
examine what happened to the relationship between crime and imprisonment for property offenses.
In 1991, there were 142,100 property offenders (burglary, larceny,
motor vehicle theft) in prison in the United States, and 179,900 by 1997.
In 1991, there were 12,961,100 property offenses reported to police; in
1997, there were 11,558,500. In sum, in 1997, there were 37,800 more
offenders locked up in state prisons for property offenses than in 1991,
while there were 1,402,600 fewer property offenses reported to police, or
roughly 37.1 fewer property crimes per added inmate. More specically,
there were 24,600 more burglars behind bars and 696,700 fewer burglaries reported, or an average of 28.3 fewer burglaries per added burglar.
152
B i g P ri s on s, B i g D ream s
For larceny-thefts, the comparable gures are 398,400 fewer crimes and
9,400 more incarcerated larceny-theft offenders, or 42.3 fewer crimes
per added larcenist. For motor vehicle thefts there were an added 3,800
offenders, 307,500 fewer motor vehicle thefts, or 80.9 fewer motor vehicle thefts per incarcerated motor vehicle theft offender. Few of these
reductions seem plausible since it would imply that the average offender
incarcerated for any one of these offenses committed an extraordinarily
large number of offenses (social defense) or dissuaded an unusually large
number of other potential offenders (deterrence). Logically, the larger
deterrent effect seen for property crimes compared to violent crime
make sense since property offenders might be easier to deter through
imprisonment. This nding would indicate that if prisons are providing
social control or deterrence, they do so for crimes of lesser seriousness
(property crimes) rather than violent crimes. To be sure, it could easily
be argued that property offenders are more likely than violent crime offenders to consider the costs and rewards of crime. This outcome is problematic, however, given the seriousness of imprisonment as a response
to crime, and our expectations concerning those crimes imprisonment
should be the most effective at controlling.
The imprisonment proponent would cheerfully report these data to
us, and note that sending more people to prison for major property offenses, even if they were not incarcerated in the same proportions in 1991
and 1997, appeared to affect crime. In response, we might ask why the
differences are so great across individual property offenses. Consider, for
example, that the number of larcenies suppressed per incarcerated larceny
offender is well below the number of auto thefts suppressed per incarcerated motor vehicle offender. We might expect the reverse relationship
since the payoff of a larceny-theft is small compared to the payoff from a
motor vehicle theft, while the price of imprisonment is high compared
to the economic gain of a larceny versus an auto theft. We might also
suggest to the prison proponent that other factors may have affected the
level of these crimes, such as a rising employment rate and greater access to less expensive home-security and antiauto theft devices. And we
would be remiss in our argument if we did not also point out that during
other periods where incarceration rose, these crimes went up, meaning
that the relationships we are examining may just be aberrations, especially since the prison proponent cannot offer specic arguments as to
Prison Effects
153
154
B i g P ri s on s, B i g D ream s
again that the crime suppression effect noted for this specic time period
is not evident for other time periods during which incarceration rose.
Data speaking to this issue will be reviewed later in this chapter. At this
point, we must still answer one question: if prisons were locking up more
inmates, and the proportional representation of serious inmates remained
the same or declined, what crimes were now more prevalent among
those for which inmates were being incarcerated? The partial answer to
this question is that there was an increase in the proportional representation of drug and public order offenders.
From 1991 through 1997, the number of inmates incarcerated in
state prisons for drug offenses rose signicantly, from 155,000 to 222,100,
or by 67,100. To place this increase in perspective, consider that for this
same time period, the number of property offenders incarcerated expanded by 30 percent while the number of incarcerated drug offenders
grew by 43 percent.
Another interesting fact about the increase in drug-related offenses
relates to the race and ethnicity of incarcerated offenders. Approximately
57 percent of drug offenders incarcerated in state prisons were black,
while 19 percent were Hispanic, and 23 percent were white. These gures, which will be reviewed later in this chapter in greater detail, point
toward the conclusion that Marc Mauer calls the race to incarcerate,
and what Austin and Irwin call the imprisonment binge, which have
focused not only on lesser offenses, such as drug offenses, but also involves a distinct racial and ethnic bias.
The social defense argument, as we now well know, is based on
the idea that expanding the use of imprisonment for serious offenses
reduced crime. Thus, we expect that when placed into practice, an expansion of imprisonment results in locking up more offenders, provides
the public with a greater degree of protection, and results in a crime reduction because serious offenders have been targeted. This expectation
should apply to drug offenders. Thus as the proportion of drug offenders
within the prison population rises, we would expect that the increase is
reected by a substantial expansion of the proportion of drug offenders
incarcerated for more serious drug crimes, especially drug trafcking.
However, when we break down the increase in the incarceration of
drug offenders we nd that from 1991 through 1997, the percentage of
the prison population incarcerated for drug trafcking declined from
Prison Effects
155
13.3 percent to 11.3 percent. Again, while there were more drug trafckers in prison in 1997 than in 1991, they made up a smaller percentage of the total prison population and a smaller percentage of convicted
drug offenders.
Violent, property, and drug trafcking offenders committed the
kinds of serious crimes we would expect our booming prison system to
incapacitate in an effort to protect citizens from crime. The fact that the
overall percentage of these three classes of offenders as a percentage of
the total prison population declined implies that our growing prison system was being used to control some other group of less serious offenders.
Who exactly were these other less serious offenders?
The dramatic expansion in drug offenders incarcerated from 1991
through 1997 was related to an increase in the percentage of the prison
population incarcerated for drug possession (from 7.6 percent to 8.8 percent). Thus, one group that our growing prison system controlled much
better were drug users, and not the group we expected to be targeted
more forcefully through an increase in imprisonment, drug trafckers. In
addition, we would be remiss if we did not point out that as the proportion of prison admissions for drug crimes have increased, so have the
proportions of nonwhites being sent to prison (Austin and Irwin, 2003,
29). It is commonly assumed by many that such a relationship reects
actual drug use and offending patterns. Later in this chapter, data that
contradicts this widely held assumption will be presented. For now, it is
sufcient to point out that the expansion of imprisonment that funneled
more drug offenders into American prisons also served as a mechanism
for incarcerating a larger proportion of the minority population.
Other offenders were also more prevalent in prisons in 1997 compared to 1991. During this period, the percentage of the prison population incarcerated for other offenses (including trafc offenses, DWI/
DUI, drunkenness, vagrancy, nonsupport, unlawful assembly, vice and
rioting, obstruction, and weapons violations), among many other public
order offenses, also increased. Though the increase was not substantial
in terms of percent growthoffenders in these categories made up
3.4 percent of the prison population in 1991 compared to 4.3 percent by
1997the expanded number of such inmates rivaled and even surpassed
the increase in the number of offenders locked up for homicides, robberies, rapes, assaults, and all property crimes for this period. Indeed, by
156
B i g P ri s on s, B i g D ream s
1997, there were 58,700 more public order offenders in Americas prisons10,000 more than the number of added murderers; 15,000 more
than the number of offenders convicted of assaults; 20,000 more than for
property crimes. A proportion of these public order offenders were there
as third-time felons, and their violations included third-strike crimes
as lowly as vagrancy. Again, we must acknowledge that some of the offenders locked up by the sweeping imprisonment movement did include
serious public order offenders who were incarcerated for weapons offenses. But we must also recognize that it was more likely that the liberty
of much more minor criminal offenders were targeted by our imprisonment binge (Austin and Irwin, 2003).
In sum, evidence concerning the percentages of violent, property,
drug, and public order offenders indicates that one of the trends in imprisonment associated with the social defense movement involved expanded incarceration of less serious offenders as a proportion of the
prison population, as well as a contraction of more serious offenders
measured proportional to the prison population. The expanding control
exerted by imprisonment drew in more serious offenders numerically; at
the same time, however, it drew in many more less serious offenders.
Repeat Offenders. As noted on several occasions, the expansion
Prison Effects
157
158
B i g P ri s on s, B i g D ream s
by research that discovers that prisons can act as schools for teaching
criminal techniques.
Nevertheless, the discussion above does not completely eliminate
the claim that locking up a larger proportion of less serious offenders reduces crime. It should be noted, however, that these data refer only to the
period in the 1990s when crime was declining at a rapid rate. As noted
in preceding chapters, data from other time periods do not support the
contention that prison expansion reduces the level of crime. Before a
conclusion can be reached, data from different time periods and regions
must also be considered.
R ac e and I m p ri s onm e nt: Th e Dru g Ef f e c t
The data presented above about offense seriousness, type of offense,
and repeat offending point toward the conclusion that Americas imprisonment binge had much less to do with controlling major crimes than
advertised. In addition, the purported goals of Americas prison expansion movement is called into question and delegitimized because of its
apparent connection to enhanced racial control. This statement requires
some explanation.
Currently, approximately two-thirds of the state prison population
in the United States is black or Hispanic. The casual observer might conclude that the large proportion of minorities represented in state prisons
is a simple function that reects the differential participation of various
racial and ethnic groups in criminal activity (Beckett and Western, 2001).
To be sure, the casual observers conclusion is supported by the images
of crime to which they are exposed by the media. But this conclusion is
far from the truth. The errors contained in this view can be examined by
inspecting the data generated by various self-report studies. These studies are useful for establishing the approximate ratio of behaviors across
different racial and ethnic groups, even though the result may not reect
the exact extent of a behavior within any of the specic groups because
of the limitations of self-report data and methods. Nevertheless, self-report studies indicate that there appears to be two to three times as much
crime self-reported by African-Americans compared to whites. This data
would seem to justify the higher concentration of African-Americans in
the U.S. prison system. This would, however, be an inappropriate conclusion reached by failing to take into account the percentage of the popu-
Prison Effects
159
160
B i g P ri s on s, B i g D ream s
Prison Effects
161
162
B i g P ri s on s, B i g D ream s
Table 6.1
Drug Use by Type of Drug and Race of Respondents, 1998
Black
Hispanic
85.2
71.7
70.8
Marijuana
35.5
30.2
23.2
Cocaine*
11.4
8.5
8.9
Alcohol
Inhalants
6.6
2.2
4.1
Hallucinogens
11.5
4.8
5.3
Psychotherapeutics
10.3
6.6
6.3
5.0
2.9
2.6
Stimulants
Tranquillizers
3.9
2.9
2.4
PCP
3.9
2.8
2.0
LSD
9.2
4.0
4.1
Heroin
1.0
1.9
0.7
163
Prison Effects
Table 6.2
Drug Use by Type of Drug and Race of Respondents, Population
Adjusted, and Per 100 Citizens, 1998
Black
Hispanic
Alcohol
61
Marijuana
26
Cocaine*
0.7
Inhalants
0.3
0.3
Hallucinogens
0.6
0.4
Psychotherapeutics
0.5
Stimulants
0.4
0.2
Tranquillizers
0.4
0.2
PCP
0.4
0.2
LSD
0.5
0.3
Heroin
0.3
0.6
164
B i g P ri s on s, B i g D ream s
Prison Effects
165
The data in tables 6.1 and 6.2 provide a useful comparison to arrest
and imprisonment data on race and drugs. In the year 2000, for example,
nearly 53 percent of federal prison inmates serving sentences for drug
crimes were white, a gure far below the reported drug-use patterns
reported in tables 6.1 and 6.2. Blacks, in contrast, made up nearly 46 percent of federal inmates incarcerated for drug crimes, a gure far in excess
of the percentage of the U.S. population of drug users that are black.
Also, consider that nearly 64 percent of those arrested for drug crimes in
the United States are white (1999 data), and we can begin to understand
the complicated processing mechanism that lands a larger proportion of
minorities in prison for drug use. That is, minorities are more likely to be
arrested for drug use compared to their actual drug use. Once arrested,
they are more likely to be charged and prosecuted. And, once charged
and prosecuted, they are more likely to be convicted and sentenced to
prison. At each stage of process, race-related difference, which individually may be small, adds up to a rather large end result (Mauer, 1999).
The race-related drug-crime-imprisonment effects noted above are
not a simple function of an increasing incarceration rate. We should recognize that numerous other laws also affect the frequency with which
blacks are incarcerated for drug crime. The war on drugs, for example, has also taken a greater toll on black Americans. In a study conducted by the Federal Judiciary Center, Meierhoefer (1992) estimated
that mandatory minimums for crack-cocaine offenses (a drug that is the
same chemically as powdered cocaine) increased black-white sentence
lengths differences. The new mandatory crack minimums pushed differences in black-white sentence lengths from 11 percent prior to passage
of these laws to 49 percent afterward. In the long run, this means that
the number of blacks in prison for crack offenses will increase more
substantially than the number of whites even if the two populations
were sentenced at the same rates because the black offenders received
longer sentences. As indicated earlier, we know that blacks are sentenced
to prison for cocaine-related offenses at a substantially higher rate than
whites, which only exacerbates the overrepresentation of blacks in our
growing prison system.
As noted earlier, Americas imprisonment binge has been greatly affected by the number of people incarcerated for drug offenses. The impact of expanded incarceration for drug offenses has a differential impact
166
B i g P ri s on s, B i g D ream s
by race. Between 1990 and 2000, for example, the increase in drug incarcerations expanded the incarcerated black drug offender population by
27 percent, while the white incarcerated drug offender population grew
by 15 percent (Harrison and Beck, 2003). While the incarcerated drug
offender population for both races was on the rise during this period,
the increase for blacks was nearly twice the rate of increase for whites.
In part, this difference is explained by sentencing patterns: 33 percent of
white defendants convicted of a drug violation in state courts were sent
to prison in 1998, compared to 51 percent of black defendants (Durose
and Langan, 2001).
The Pe r si ste nc e of R ac e Ef f e c ts
Race differences are seen across a broad spectrum of correctional
system data (Beckett and Western, 2001; Jacobs and Carmichael, 2001;
Pettit and Western, 2004; Smith, 2004; Yates and Fording, 2005). These
data, when combined with studies of the prevalence of crime across racial
groups, and other data on criminal justice processing, indicate that more
blacks end up in prison because of various racial biases, many of which
have become institutionalized and hidden in the criminal justice system
(Austin and Allen, 2000; Lynch and Patterson, 1991, 1996). As an example
of the persistence and consistency of racial bias in imprisonment, consider the data presented in table 6.3 taken from the Sourcebook of Criminal
Justice Statistics (table 6.32, 511, 2000), showing the rate of sentenced prisoners per 100,000 United States residents by race and age.
This table indicates that for each age group, with the exception of
the fty-ve and older category, the black rate of incarceration is approximately seven to nine times higher than the white rate of incarceration, while the Hispanic rate of incarceration is two to three times higher
than the white rate of incarceration. These race-related differences might
be acceptable if they were a reection of actual offending rate differences
across racial groups. There are several reasons, however, to suspect that
this is not the case, as the examples above illustrated.
To further illustrate the degree of bias represented by these data, we
can compare them to evidence from victimization data. The National
Crime Victim Survey conducted through the U.S. Department of Justice,
Bureau of Justice Statistics, collects data from crime victims concerning
their victimization experiences using a household survey.
167
Prison Effects
Table 6.3
Incarceration Rate Per 100,000 By Race and Age, for Males in State and
Federal Prisons, 2000
All
White
Black
Hispanic
1819
785
302
2,679
1,058
2024
2,045
866
7,276
2,503
2529
2,520
1,108
9,749
2,890
3034
2,355
1,219
8,690
2,740
3539
1,889
995
7,511
2,134
4044
1,316
697
4,995
2,088
4554
707
428
2,699
1,144
55 +
164
112
540
401
Total
904
449
3,457
1,220
168
B i g P ri s on s, B i g D ream s
Prison Effects
169
born has of ending up in prison, and we can begin to understand imprisonments differential effects on the life course of people of different
racial backgrounds in the United States. Again, these differences become
more disturbing when we consider that they may be the result of racial
biases in criminal justice processes.
With these facts in mind, it can be suggested that the recent growth
of the U.S. prison system is part of the modern history of race control
and segregation that has come to characterize American society (Massey
and Denton, 1993). Our prison system, in other words, not only reects
the degree of racial segregation seen in society at large, but has become
part of the mechanism that supports and produces racial segregation. The
larger the prison system has become, the larger the percentage of the
black population segregated in prisons has also become. As the percentage of the black population that is incarcerated increases, the probability
that blacks, who now are more likely to bear the stigma associated with
the ex-con label, will have fewer opportunities to advance economically also has expanded, locking them into communities that have higher
rates of ex-offenders and a lower chance of attracting the revitalizing
economic opportunities (for discussion, see Rose and Clear, 1998; Clear,
Rose, and Ryder, 2001; Clear and Rose, 1999; Clear, 2003). Prisons, in
effect, have become a part of the now vicious circle of segregation, community decay, and disinvestment that has come to characterized modern
American minority communities (Hagan, 1997).
P ri sons: A Class Ac t
Each year, the Bureau of Justice Statistics (BJS) publishes statistics
about the characteristics of prison populations. As part of this survey, BJS
published a more extensive census of prison inmates that used to include
various personal characteristics of prison inmates, such as income in the
year prior to arrest, the inmates occupation at time of arrest, and educational achievement. These data were useful for discerning the social class
of inmates. While BJS still published this prison census, it now excludes
information on income and occupation necessary for making inmate social class determinations. Currently, the only useful information that remains for making a class assessment is level of educational attainment.
In 2002, 88 percent of inmates in state prisons in the United States
had a high school education or less. Fourteen percent had an eight grade
170
B i g P ri s on s, B i g D ream s
education or less, while 29 percent had some high school, and 45 percent
had some sort of high school diploma (GED or an actual high school
diploma). Taken together, 89 percent of inmates had a high school education or less. These gures are well below the educational attainment of
the general U.S. population, where about 41 percent of the population
fall into the same educational categories.
Furthermore, consider that from 1991 through 2000, the percentage
of inmates with college degrees declined somewhat, at the same time
that an increasing number of white-collar and corporate crime scandals
were being uncovered in America.
While it is true that white-collar and corporate offenders, if convicted and sentenced to imprisonment, are very likely to serve their sentences in federal prisons and therefore are not included in these data,
educational data for the federal prison system do not show any substantial increase in the percentage of inmates with college degrees.
Available evidence suggests that Americas imprisonment binge has
concentrated its efforts on lower-class criminals. When coupled with the
observation that Americas imprisonment binge has also concentrated on
less serious crimes, and that the expansion of imprisonment has not been
employed to incarcerate a substantial segment of white-collar and corporate criminals, we, like Jeffrey Reiman (2003), can conclude that the
rich get richer while the poor get prison.
P ri son Tre nds and G e nde r
While not a major focus of this work, I would be remiss if I failed
to note the gendered effects of imprisonment. Did the growth of U.S.
prisons affect the incarceration rates for men and women in similar ways?
The answer to this question is complex and would require extensive analysis to address completely. Nevertheless, a few general observations can
be used to depict the overall trends and differences. These trends indicate
that the imprisonment binge had a disproportionate impact on women
compared to men, and on African-American women in particular,
From the initiation of the imprisonment binge in 1973 through
2004, the rate of incarceration for men (men imprisoned per 100,000
population) grew from 96 to 486, or by more than 400 percent. During
the same period, the imprisonment rate for women expanded by 966
percent, from 6 per 100,000 to 64 per 100,000. Overall, the percentage of
Prison Effects
171
the prison population composed of women also increased from 2.7 percent to 6.7 percent. Most striking, perhaps, is that the number of women
in U.S. prisons grew from 6,004 in 1973 to over 85,000 by 2004.
Although women continue to make up a small proportion of the
U.S. prison population, the evidence suggests that prison expansion
had an unequal gender impact. Thus, the imprisonment binge not only
maintained a focus on the lower class, expanded the proportion of the
prison population composed of African-Americans and Hispanics, but
also entailed a signicant increase in the representation of women in
U.S. prisons.
The racial and ethnic differences evident for male prison populations are also evident in the female prison population. Harrison and
Beck (2006) note that in 2004 black females were incarcerated in U.S.
prisons and jails at a rate of 359 per 100,000, compared to a rate of 143
for Hispanic women and 81 for white females. The extent of racial and
ethnic disparity is lower for female compared to male inmates, but is
nevertheless problematic (for further data comparing incarceration differences by gender, race, and ethnicity see Bonczar, 2003). To put the
race-gender link into further context, consider Bonczars (2003) estimate
that the number of black women ever incarcerated in the United States
(231,000 or 1.7 percent of black women) exceeds the number of white
women ever incarcerated (225,000 or 0.3 percent of white women). This
is surprising given that the number of white females far exceeds the size
of the black female population.
In terms of offense type, about one-half of male inmates are incarcerated for violent crimes, while only one-third of women are incarcerated for those offenses (Harrison and Beck, 2006). Female inmates were
much more likely to be incarcerated for lesser offenses than males. For
example, while one in ve male inmates were serving time for drug offenses in 2002, the comparable gure for females was one out of three
(Harrison and Beck, 2006).
Overall, similar factors affected the rise in female and male imprisonment rates in the United States. For both groups, the increase is linked
to a rise in drug-related incarceration, and for both there is evidence of
racial and ethnic disparity. Thus, several of the general observations made
about the imprisonment binge apply to both female and male imprisonment trends. Analyzing the exact differences between these two trends
172
B i g P ri s on s, B i g D ream s
Chap te r 7
174
B i g P ri s on s, B i g D ream s
175
Table 7.1
Crime Suppression and Enhancement Effects of Incarceration Changes, on
Crime Changes, 19732003
Year
% Change
Imprisonment
% Change
Crime*
CSE
Effect
CEE
Effect
1973
3.2
17.2
5.38
1974
7.0
10.1
1.44
1975
10.1
0.5
0.05
1976
9.2
-3.2
-0.35
1977
8.6
2.0
0.23
1978
3.1
9.3
3.00
1979
2.4
9.5
8.95
1980
4.8
0.1
0.02
1981
12.0
-3.4
-0.28
1982
11.8
-6.7
-0.57
1983
6.0
-1.9
-0.32
1984
5.7
4.6
0.81
1985
8.4
6.3
0.75
1986
8.6
2.3
0.27
1987
7.4
3.1
0.42
1988
7.7
2.4
0.31
1989
12.8
1.6
0.13
1990
8.7
2.7
0.31
1991
6.7
-2.9
-0.48
1992
7.1
-2.0
-0.28
1993
10.1
-1.1
-0.11
1994
9.1
-0.9
-0.10
1995
6.7
-2.7
-0.41
1996
4.9
-2.2
-0.45
1997
5.0
-5.4
-1.08
1998
4.3
-6.8
-1.58
1999
4.7
-0.2
-0.09
2000
2.1
2.3
1.10
2001
1.1
0.1
0.09
2002
2.6
0.4
0.15
(continued)
177
% Change
Imprisonment
% Change
Crime*
CSE
Effect
CEE
Effect
2003
2.0
-1.1
-0.55
2004
1.8
**Sum
-6.65
23.33
14
17
Mean
-0.48
1.37
* Because a lag was included, the percent change in crime for any listed year is the change
for the next year (e.g., the percent change in crime listed for 1973 is the change in crime
between 1973 and 1974, while the imprisonment change is from 1972 to 1973).
** Excludes 2004.
The CSE and CEE scores can also be summed to calculate the mass
effect of each for this time period. The mass crime suppression effect
was -6.65, while the mass of the crime enhancement effect was 23.33. In
sum, these gures indicate that the mass CEE effect outweiged the mass
CSE over this time period.
Another important observation is that there were seventeen CEEs
and fourteen CSEs for this time period. That is, over this period where
imprisonment was continually rising, crime was more likely to rise (the
number of CEEs) than fall (the number of CSEs).
So far, the data above do not support the idea that a rise in imprisonment is related to a decline in crime. It is quite possible, however, that
a CEE or CSE outcome is related to the magnitude of the change in
imprisonment rate. Deterrence or incapacitation theorists might argue
that while a CEE is slightly more likely than a CSE for this time period, the likelihood of each is related to the magnitude of the change in
imprisonment. This proposition can be easily assessed by adding imprisonment changes for each effect together, and dividing by the number
for years for which each effect is present. Following this procedure, the
average change in imprisonment was 7.11 percent for CSE years and
6.14 percent for CEE years. While the average change in imprisonment
178
B i g P ri s on s, B i g D ream s
is indeed greater in CSE years, the difference is less than 1 percent. The
difference in magnitudes in imprisonment change, however, cannot explain why crime rose or fell since, in theory, expanding imprisonment
should cause crime to fallthe magnitude should make little difference
since imprisonment is always expanding during this time period. Furthermore, the 0.97 percent difference in imprisonment across CSE and
CEE years does not appear large enough to explain the larger difference
noted in crime between that same period (the mean of which is 1.85
percent; or the difference between the average CSE effect of -0.48 and
the average CEE effect of 1.37).
Given this result, however, is it plausible that larger changes in imprisonment are more likely to lead to a CSE. This can be assessed by
dividing the data into two groups: one above the mean imprisonment
change and the other below the mean change (6.8 percent). For the 16
years in which the mean change in imprisonment was less than 6.8 percent, there were 8 CEE effects and 8 CSE effects. Thus, for below mean
changes, either outcome was equally likely. For the fteen years above
the mean imprisonment change, there were nine CEE effects and six
CSE effects. This nding also fails to support deterrence and incapacitative arguments. Taken together, these ndings indicate that higher or
lower than average changes in imprisonment do not predict the direction of the change in crime. Moreover, and contrary to expectations,
a larger than average change in imprisonment is more likely to be associated with an increase in crime. There are some submerged trends in
these general trends that bear some comment. Because the number of
submerged trends are derived from very few cases in each category, it is
difcult to rely on these trends to any substantial degree.
What is the specic effect of a change in imprisonment that was
below the mean? For the 8 CEE years, the average CEE effect was 5.4.
For the remaining 8 years where CSE effects were seen, the average
CSE was -2.9. Thus, lower than average changes in imprisonment were
no more likely to suppress or enhance crime. However, the average CEE
effect is much larger than the average CSE effect, indicating that when
imprisonment rose less than the mean, crime was not more likely to
increase substantially.
For the 15 years in which the imprisonment change was above the
mean, the average annual CEE effect was 3.74 (N = 9), while the annual
179
180
B i g P ri s on s, B i g D ream s
181
182
B i g P ri s on s, B i g D ream s
Table 7.2
Forecast of Crime and Imprisonment Rates from
1990s Trends*
1999
2000
2001
2002
2003
.
.
2010
2011
2012
.
2018
.
2020
2021
2022
.
.
2036
Imprisonment
Crime
476
502
529
558
588
4,266.8
4,109.4
3,957.8
3,811.8
3,671.1
850
896
944
2,821.5
2,717.4
2,617.1
1,295
2,088.5
1,439
1,517
1,599
1,937.2
1,865.7
1,769.9
3,339
1,044.6
* This forecast assumes a direct relationship between increasing the rate of incarceration and the crime rate predicted by the trends in crime and imprisonment from the
1990s, when the annual average change in the imprisonment rate was 5.4 percent, and the average annual decline
in crime was 3.6 percent. To simplify this table, not all
years are presented, and where years are omitted, a dot is
placed in the table for the readers convenience.
or by more than 2,900 inmates per 100,000, which would add more inmates than were incarcerated in 1999 to the system.
Based on this forecast, and assuming a constant population (or no
population growth), the actual number of people the United States would
need to imprison to achieve crime reduction would be enormous. For
183
184
B i g P ri s on s, B i g D ream s
185
186
B i g P ri s on s, B i g D ream s
state. For example, a P-Ratio of 0.10 means there are ten times
as many whites in the population as blacks, or 1 black for every
10 whites. A P-Ratio of 0.50, would indicate that there is 1
black for every two whites in the population.
(3) The RR/PR ratio can be interpreted as an index of racial representation. This ratio expresses the extent to which blacks are
overrepresented within a state prison system compared to their
representation in the general population of each state. The RR/
PR ratio is derived by dividing the R-Ratio by the P-Ratio.
The RR/PR ratio is interpreted in the same way as the R-Ratio. An RR/PR ratio of 1.00 indicates that the ratio of blacks
imprisoned within a state is equal to the ratio of blacks to whites
found in the general population. RR/PR ratios in excess of 1
indicate that the ratio of blacks to whites in a states prison system exceeds the ratio of blacks to whites in the general population. RR/PR ratios in excess of 3 likely indicates the existence
of fairly extensive institutionalized discrimination.
Imprisonment and Crime Rates, 1999
Let us rst summarize the information about crime and imprisonment rates found in table 7.4, omitting the Washington DC area as an
outlier. The highest rates of imprisonment for 1999 were found in Louisiana (776 per 100,000) and Texas (762 per 100,000). Two additional states,
Oklahoma and Mississippi, have imprisonment rates in the mid-600s. Four
more states have rates in the 500s. Seven of the eight states with the highest
rates of incarceration (the exception is Nevada) are located in the south.
The state with the lowest rate of incarceration was Minnesota, where
125 inmates per 100,000 population were imprisoned. Minnesotas rate
of incarceration was 6.2 times lower than the rate of incarceration found
in Louisiana, the state with the highest rate of imprisonment. In addition
to Minnesota, 6 other states had incarceration rates lower than 200 per
100,000. Four of these states were in the northeast.
The mean rate of imprisonment across all states was 381 per 100,000
population. Regionally, mean imprisonment rates per 100,000 vary
greatly: northeast, 274; midwest, 319; west, 361; south, 508. It is evident
from these regional data that the mean imprisonment rate for the United
States is inuenced by the high mean rates found across southern states,
187
Table 7.3
States by Region
Northeastern:
Midwestern:
Southern:
Alabama, Arkansas, Delaware, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma,
South Carolina, Tennessee, Texas,Virginia, West Virginia
Western:
where only two states, North Carolina and Kentucky, have imprisonment
rates below 400. Taken together, these data illustrate that there is signicant variation in the use of imprisonment across states and regions within
the United States.
There is also signicant variation in the level of crime per 1,000 citizens across states and regions. The lowest crime rates, 22.8 per 1,000, are
seen in New Hampshire and Nevada. The highest crime rate was found
in Florida (62.1 per 1,000 population). Regionally, crime rates were lowest in the northeast (31.1), followed by the midwest (37.4), south (42.8),
and west (44.0).
Having summarized the general ndings in these state-based data, it
is appropriate to investigate the relationship between the level of imprisonment and crime across states. Recall that the deterrence and incapacitation arguments suggest that criminals are either deterred or prevented
from committing crime by the level of incarceration. Thus, we would
expect that states with the highest levels of incarceration would have the
lowest levels of crime. This hypothesis can be tested in a simple manner by performing a test of statistical correlation between the rates of
imprisonment and crime across states. The deterrence-incapacitation hypothesis would be rejected if the correlation between the rate of imprisonment and crime were positive, because this indicates that states with
higher levels of imprisonment also have higher levels of crime.
The Pearsons R correlation between imprisonment and crime
rates for 50 U.S. states in 1999 is .455, and, moreover, this relationship
is statistically signicant (p = .001). The correlation indicates that the
324
252
Iowa
198
Vermont
Indiana
193
Rhode Island
319
305
Pennsylvania
368
400
New York
Illinois
384
New Jersey
MIDWEST
266
187
133
Maine
New Hampshire
397
Connecticut
Massachusetts
274
NORTHEAST
Imprisonment
32.2
37.7
45.1
37.4
28.2
35.8
31.1
32.8
34.0
22.8
32.6
28.8
33.9
31.1
Crime
4,800
10,132
9,995
97,802
1,193
2,157
11,632
29,665
7,316
2,019
5,590
1,469
4,630
65,661
White
1,696
7,707
26,522
111,674
36
1,175
19,847
37,488
18,572
120
3,448
58
8,059
88,803
Black
Prison Population
0.35
0.76
2.65
1.14
0.03
0.35
1.71
1.26
2.54
0.06
0.62
0.04
1.75
1.35
R-Ratio
0.02
0.09
0.19
0.07
0.005
0.05
0.11
0.23
0.18
0.007
0.07
0.005
0.11
0.16
P-Ratio
(continued)
17.7
8.5
14.0
11.7
6.0
10.9
15.5
5.5
14.1
8.5
8.8
7.9
15.8
10.0
RR/PR
Imprisonment Rates (Per 100,000), and Crime Rates (Per 1,000) in 1999, and Number and Ratio of Black and White Inmates
Table 7.4
375
503
549
Wisconsin
SOUTH
Alabama
456
339
South Dakota
Florida
417
Ohio
1343
137
North Dakota
DOC*
217
Nebraska
443
477
Missouri
493
125
Minnesota
Delaware
472
Michigan
Arkansas
321
Kansas
Imprisonment
62.1
80.7
48.4
40.4
44.1
42.8
33.0
26.4
40.0
23.9
41.1
45.8
36.0
43.2
44.4
Crime
27,445
91
1,942
4,448
7,615
165,008
7,910
1,705
21,846
611
2,237
12,917
2,559
18,482
4,608
White
35,721
9,096
3,481
5,543
14,594
273,605
7,788
99
25,938
20
1,008
10,968
1,964
24,936
3,028
Black
Prison Population
1.30
***
1.79
1.25
1.92
1.66
0.99
0.06
1.19
0.03
0.45
0.85
0.77
1.35
0.66
R-Ratio
0.19
1.87
0.24
0.20
0.36
0.28
0.06
0.007
0.13
0.007
0.04
0.13
0.03
0.17
0.06
P-Ratio
(continued)
6.9
53.5
7.5
6.3
5.3
6.8
16.4
8.2
9.1
7.0
11.3
6.5
25.6
7.9
11.0
RR/PR
532
385
776
427
626
345
662
543
408
762
447
196
361
374
Georgia
Kentucky
Louisiana
Maryland
Mississippi
North Carolina
Oklahoma
South Carolina
Tennessee
Texas
Virginia
West Virginia
WESTERN
Alaska
Imprisonment
43.6
44.0
27.2
33.7
50.3
46.9
53.2
46.8
51.8
42.7
49.2
57.5
28.8
51.5
Crime
1,895
11,2440
2,643
9,221
39,697
8,114
6,282
11,188
10,044
3,560
4,998
6,852
8,976
11,983
White
600
63,440
502
18,970
63,883
8,437
14,762
7097
20,418
10,663
17,196
22,360
5,586
24,392
Black
Prison Population
0.32
0.56
0.19
2.06
1.61
1.04
2.35
0.63
2.03
3.00
3.44
3.26
0.62
2.04
R-Ratio
0.05
0.04
0.03
0.26
0.15
0.20
0.43
0.09
0.29
0.58
0.41
0.49
0.08
0.41
P-Ratio
(continued)
6.1
6.4
6.3
7.9
10.7
5.2
5.5
7.1
7.0
5.2
8.4
6.7
7.8
5.0
RR/PR
481
383
320
385
335
509
270
293
245
251
355
California
Colorado
Hawaii
Idaho
Montana
Nevada
New Mexico
Oregon
Utah
Washington
Wyoming
495
Arizona
Imprisonment
34.5
52.6
49.8
50.0
59.6
22.8
40.7
31.5
48.4
40.6
38.1
59.0
Crime
1,190
9,376
3,709
5,839
3,892
5,048
2,058
3,165
1,063
9,562
46,957
18,686
White
82
2,962
328
1,010
545
2,407
35
65
232
3320
48,331
3,523
Black
Prison Population
0.07
0.32
0.09
0.17
0.14
0.48
0.02
0.02
0.22
0.35
1.03
0.19
R-Ratio
0.009
0.04
0.009
0.02
0.03
0.09
0.004
0.006
0.09
0.05
0.10
0.04
P-Ratio
7.7
7.9
9.8
8.7
4.7
5.3
4.2
3.4
2.4
6.9
10.3
4.8
RR/PR
192
B i g P ri s on s, B i g D ream s
193
194
B i g P ri s on s, B i g D ream s
If policymakers, scholars, and the public have been reluctant to acknowledge segregations persistence, they have likewise been blind
to its consequences for American blacks. Residential segregation is
not a neutral fact; it systematically undermines the social and economic well-being of blacks in the United States. Because of racial
segregation, a signicant share of black America is condemned to
experience a social environment where poverty and joblessness are
the norm, where a majority of children are born out of wedlock,
where most families are on welfare, where educational failure prevails, and where social and physical deterioration abound. Through
prolonged exposure to such an environment, black chances for social
and economic success are drastically reduced (1993: 2).
In short, not only is there a high level of unacknowledged segregation
in America, but because of its widespread nature and our unwillingness
to recognize its existence, a diverse array of negative consequences befall
black communities and individuals.
Why begin a discussion of race and prisons with reference to Massey
and Dentons comments concerning forms of residential or spatial segregation that are prevalent in the free world across a variety of communities found in the United States? Because these comments not only
apply to communities, but to prisons as well, which have also become
sites of racial segregation. Massey and Dentons orientation also helps us
to better understand one of the neglected roles prisons play in American
society, and how they also contribute to detrimental life conditions for
American blacks and their communities, a point support by the research
of Todd Clear and Dina Rose.
More specically, and to be very clear on this point, if Massey and
Dentons argument is accepted and applied to the U.S. prison system,
it helps explain why the racial composition of American prisons is so
heavily slanted: because our big prison system has become part of the
purposeful institutional arraignments that help fortify racial segregation. Consider, for the moment, the evidence presented in table 7.4 concerning the prison racial ratio as compared to the racial ratio of the free
population for each state (RR/PR ratio). These ratios illustrate the extensive degree of racial segregation prisons help foster. To be sure, different states incarcerate blacks at different rates, and in different proportions
195
196
B i g P ri s on s, B i g D ream s
assume that these states may also harbor conditions conducive to crime,
and that given high concentrations of African-Americans, they are likely
to be more adversely affected than whites. This explanation is useful to
the extent that it seems logical. And, while a greater percentage of the
African-American community is likely to suffer from poverty conditions compared to whites, a larger overall percentage of the poor are
white. Thus, states with small African-American populations but high
white poverty rates (e.g., West Virginia) would also be expected to have
elevated rates of incarceration.
This proposition can be assessed by regressing the crime rate and
the black-white population ratio against the rate of imprisonment. If the
relationship between racial composition across states and imprisonment
rates is mediated by the crime rate, then the relationship between racial
composition and imprisonment would disappear or become extremely
attenuated once we control for crime in a multiple regression format.
Likewise, if the relationships between racial composition and imprisonment were conditioned by poverty rates, then adding poverty rates as a
control variable should remove or lessen the relationship between racial composition and imprisonment rates. When crime rates, state racial
composition and poverty rates are regressed against cross state imprisonment rates, the following results emerge. The adjusted R square for the
equation is 49.0 percent, meaning that these three variables alone account for close to one half of the variation in rates of incarceration across
states. Further, the relationship between state racial composition (the
black-white ratio) and imprisonment is positive and statistically signicant (p < .000; b = 532.13). Likewise, the relationship between poverty
and imprisonment is positive, though not quite statistically signicant
(p = .052; b = 10.68). The relationship between the crime rate and the
imprisonment rate is positive, but statistically insignicant (p = .157;
b = 2.35). Although this is an incomplete test of the relationship between imprisonment rates and racial composition, this analysis indicates
that of the included variables, racial composition has the strongest impact
on the level of incarceration, and, further, that this relationship exists independently of the level of crime or poverty. Moreover, any relationship
between crime rates and imprisonment rates (the Pearson R for state
crime and imprisonment rates is .613, and is signicant at p = .000) is
removed once racial composition is controlled for in the analysis.
197
198
B i g P ri s on s, B i g D ream s
contributing directly to the physical segregation of the races in America. For example, the ex-con label and record bestowed upon individuals who have been incarcerated diminishes the likelihood that they
will nd future employment. This formal label, which has numerous informal social control effects, continues the punishment of the offender
unlucky enough to have been caught, prosecuted, found guilty, and sentenced, over the remainder of that individuals life. Because institutionalized police procedures focus more on events in minority communities,
the likelihood that black law violators will be caught, prosecuted, and
convicted is greater than for whites.
The experience of being in prison, and the ex-con label, dramatically alters the lives of those who have been incarcerated, and generally
not in a positive manner. The ex-con is often socially and economically
isolated from the mainstream, either driving them further into deviant
lifestyles, embedding them into criminal careers, or attaching them to
a life that vacillates quite easily between crime and conformity. Thus,
rather than deter or rehabilitate the offender, the prison is more likely
to establish a more persistent pattern of offending. For those who doubt
that this is the case, all one needs to do is consider the high rate of recidivism among the former-inmate population.
It is necessary to understand the race-linked nature of the effects of
imprisonment. Blacks and Hispanics comprise about 70 percent of the
prison population in the United States, meaning that minorities are the
majority within Americas prisons. They are not in prison because they
are societys worst criminals, as is illustrated by the substantial literature
that focuses on notable nancial and violent crimes committed by white
corporate executives (e.g., Reiman, 2003; Simon and Eitzen, 2000; Friedrichs, 2004; Burns and Lynch, 2004; Lynch and Michalowski, 2006).
While minorities comprise the majority of the prison population in the United States, consider also that compared to other groups,
a rather sizeable proportion of the black population is or has been in
prison (Mauer, 1999), or has a substantial lifetime likelihood of being imprisoned (Bonczar and Beck, 1997; Pettit and Western, 2004). Table 7.5
shows the lifetime likelihood of imprisonment by race at different ages
as estimated by Bureau of Justice Statisticians, Thomas Bonczar and Allen Beck. As table 7.5 illustrates, black males are more likely than white
males in every age group to be incarcerated. It should be noted that the
199
diminishing race ratio that results with increased age is, in part, due to
the fact that the population of blacks available for incarceration diminishes more rapidly than the population of white available for incarceration due to an increase in the number of blacks incarcerated, as well as
reduced participation in crime that is attributable to the aging process,
and differentials in race related mortality rates.
Let us examine the race ratios found in table 7.5 more closely. These
ratios indicate how much more likely a black male is than a white male
of the same age to be incarcerated. The sizeable imprisonment chance
difference for black males compared to white males at birth (6.48) indicates that blacks are substantially more likely (nearly 6.5 times more
likely) than whites to have their life course altered by a serious form of
criminal justice intervention. This nding indicates that criminologists
who study the life course need to make a more extensive effort to study
how race shapes life course (see Lynch, 1996 for discussion).
Consider also that through age 20, the race ratio remains above 6,
and that that it is not until the age of 45 that the ratio declines signicantly (to 2.63). Thus, it is not surprising that the detrimental effects of
imprisonment fall most heavily on black offenders, and, equally if not
more importantly, on black communities (Clear, 1994), and most importantly, on young black males. While the institution of imprisonment
is not the only cause of these effects (Massey and Denton, 1993; see also,
Hacker, 1995), it certainly contributes to, and intensies the kinds of
problems black Americans encounter everyday, especially for individuals
who have been incarcerated.
Finally, it is instructive to examine one last table illustrating the
relationship between race and imprisonment in the United States. This
table also depicts changes in the racial composition of American prisons
since 1980.
In 1980, the black-white imprisonment rate ratio was 6.6, meaning that given equivalent sized base populations, blacks were 6.6 times
more likely to be in prison than whites. This level of racial disparity in
incarceration is at least twice the estimated offending difference between
whites and blacks in the United States (Lynch and Schuck, 2003; Lynch,
2002, 2000). The peak race related imprisonment difference occurred in
1996, when blacks were 8.4 times as likely to be imprisoned compared
to whites. Since 1996, the rate in racial differences has slowly declined,
200
B i g P ri s on s, B i g D ream s
Table 7.5
Lifetime Chance of Imprisonment by Age and Race for Males*
Age**
White Male
Black Male
Race Ratio***
Birth
4.4
28.5
6.48
20
4.1
25.3
6.17
25
3.0
17.3
5.77
30
2.1
10.8
5.14
35
1.5
6.5
4.33
40
1.1
3.6
3.27
45
0.8
2.1
2.63
201
Table 7.6
Rate of Imprisonment Per 100,000 Population, by Race and Ethnicity,
Various Years, 19802003
White
Black
Hispanic
B/W Ratio
M/W Ratio
2003
465
3405
1231
7.3
10.0
2002
450
3535
1177
7.9
10.5
2000
449
3457
1220
7.7
10.4
1999
417
3408
1335
8.2
11.4
1997
386
3209
1273
8.3
11.6
1996
370
3098
1278
8.4
11.9
1992
372
2678
7.2
1991
352
2523
7.2
1990
339
2376
7.0
1989
317
2200
6.9
1985
246
1559
6.4
1980
168
1111
6.6
1. Complied from various Bureau of Justice Statistics annual reports on prisoners in the
United States.
2. Data on Hispanic inmates is unavailable for the entire period under investigation.
3. B/W ratio is the black-white ratio, which is derived by dividing the black rate of
imprisonment by the white rate of imprisonment.
4. M/W ratio is the minority-white ratio, derived by adding the black/white imprisonment rate ratio and the Hispanic/white imprisonment rate ratio.
202
B i g P ri s on s, B i g D ream s
Despite these results, this chapter also examined the issue of how
large the U.S. prison system would need to become to suppress crime
signicantly. Following the lead of researchers who have focused solely
on data from the 1990s, when a rising rate of incarceration appears to
lower the rate of criminal offending, imprisonment and crime rates were
forecast forward to determine levels of each if the path taken in the
1990s continued. This analysis indicated that signicant reductions (onehalf the level of crime in 1999) in crime would not occur until 2018. To
do so, the rate of imprisonment, however, had to increase an additional
176 percent above its 1999 level, which translated into an extraordinarily
large and expensive prison system.
This chapter also examined cross sectional data on imprisonment
and crime. Here, too, the data indicated that the relationship that exists
between the two is more fancy than fact. It was demonstrated that states
with high rates of incarceration had high rates of crime and not the low
rates deterrence or incapacitation theory would predict.
In examining these data, we returned to reexamine the relationship
between race and imprisonment in the United States. State-level data
indicated that imprisonment was highest in states that had large AfricanAmerican populations, and that the size of a states African-American
population was a better predictor of incarceration rates than crime rates,
or conditions that may cause crime such as poverty.
In the next chapter, one nal issue will be examined before a general
summary of this book is presented. That chapter focuses on the problems
presented by a massive prison system in an era where energy resources
are being depleted and oil is becoming more scarce. This chapter is included to illustrate an additional problem with Americas large prison
system that policy makers need to address.
Chap te r 8
204
B i g P ri s on s, B i g D ream s
societies cannot continue to employ fossil fuels without dire environmental consequences.
End-of-oil researchers are also not all of one mind. Like any other
area of research, some believe that the issue is extremely urgent, while
others argue that the sense of urgency is a gross exaggeration (e.g., see
Michael C. Lynchs position, http://www.hubbertpeak.com/Lynch, or
that taken by the George W. Bush White House). This latter group rejects
the end of oil as plausible, and at worst sees future energy needs being
met by alternative fossil-fuel technologies (e.g., coal, shale- and sandbased oil, natural gas). This view also dismisses the related problems of
pollution and global warming as irrelevant and overblown.
The middle ground is occupied by a more optimistic group (e.g.,
Vaiteeswarah, 2003), which believes that scientic ingenuity will save us
from the end of oil, produce an alternative energy system that will spur
economic growth (Lovins et al., 2005), and that energy innovations will
push the end-of-oil scenario further into the future, allowing more time
to develop an alternative energy infrastructure. This group is optimistic
about fossil-fuel alternatives such as shale and sand oils, and typically fails
to address related environmental issues such as the impact of fossil-fuel
alternatives on global warming.
Despite this controversy, even the worlds largest oil companies recognize the problems presented by the dwindling world supply of oil.
John Browne, CEO of the worlds second-largest energy company, BP,
argues that it is necessary to develop nonfossil fuel energy alternatives.
The emerging consensus among oil companies is depicted in the following excerpt from a letter to the public published by Chevron chairman
and CEO, David J. O Reilly:
Energy will be one of the dening issues of this century. One thing
is clear: the era of easy oil is over. What we do now will determine
how well we meet the energy needs of the entire world in this century and beyond. Demand is soaring like never before . . . some say
that in twenty years the world will consume forty percent more
oil than it does today. . . . We can wait until a crisis forces us to do
something. Or we can commit to working together, and start by asking the tough questions: How do we meet the energy needs of the
developing world and those of the industrialized nations? What role
205
206
B i g P ri s on s, B i g D ream s
does not accomplish the mission of reducing crime. It is possible that the
end of oil may occur later than predicted and that new, less- or non-polluting energy sources will be discovered and made widely available. But
we cannot sit back and hope that this scenario emerges, for if it does not,
American institutions, such as the prison system, will be burdened with
a tremendous challenge.
The tendency of U.S. criminal justice policy makers to expand the
criminal justice system and continue building prisons must be considered
against a potential future energy crisis and the demands for increased energy criminal justice expansion poses. Devising criminal justice alternatives may require rethinking the philosophical basis of criminal justice in
the United States, which is currently based on enlarging criminal justice
functions to enhance the deterrent and incapacitiative effects of criminal
responses. As the early chapters illustrate, these goals are not being met.
Before turning to a discussion of how the American correctional
system might be reformed to respond to changing energy conditions,
further background information is presented.
D e pe ndi ng on O i l
Colin Campbell3 (2001:1), a leading peak-oil researcher, noted, The
fundamental driver of the 20th Centurys economic prosperity has been
an abundant supply of cheap oil. Expanding on this point, Jeremy Rifkin
(2002) suggests that in the near future, energy production and consumption will become the primary determinants of twenty-rst-century economic development. These views reect widespread agreement among
geologists, physicists, large-scale nancial planners, and even oil companies that the world is running out of oil. To make matters worse, there
is no viable (by which I mean, readily available, widespread, large-scale,
easily converted, nonglobal warming, low-pollution) energy alternative
in sight. And, like everything else, Americas prison system is dependent
on a source of energy. Furthermore, because the U.S. prison system is the
largest in the world, it consumes a great deal of energy.
If peak-oil researchers are correct, and action is not taken, the end
of oil will disrupt world economies (e.g., cause widespread ination,
depressions, increased poverty, enhanced between class disparities in
wealth), and the cost of operating social institutions like prisons will
expand. Ordinary criminal justice functions will become a challenge to
207
operate in an oil-depleted world. These observations suggest that criminal justice researchers and policy planners must become aware of warnings about the potential for a coming energy crisis, and that they must
act now to reorganize criminal justice practices and procedures to be
on the safe side. Absent signicant reform, social control institutions in
America could one day come to a grinding halt, assuming the dire forecasts
of peak-oil researchers are correct.
The D e c l i ni ng Ava i labi l i ty of O i l
As noted, there are various views concerning the scope of the problem presented by peak oil. The Society for Peak Oil Study, among others,
lays out the most extreme case. Of particular concern is transforming the
worlds current fossil-fuel economy in the face of scientic evidence that
world oil production is near its peak.
The phrase peak oil refers to the point when one-half of the worlds
oil reserve will have been used up or depleted. So what? you may be
thinking. There is still one-half of the oil reserve left. It has taken more
than a century to use the rst half. When the oil peak is reached, there
should be another one hundred years of oil remaining. The problem is
that the half remaining after the peak is reached will be insufcient to
support world energy needs beyond the middle of the twenty-rst century because of increased rates of energy use.
The idea of a peak in oil production was initiated by geologist M.
King Hubbert, who in 1956 employed U.S. oil consumption, production, and reserve data to predict the U.S. oil crisis in the 1970s. Kings
predictions (known as Hubberts Peak or Hubberts Curve) presented
at the meetings of the American Petroleum Institute were widely known
to those in the oil industry. This study was greeted with much skepticism.
Many believed that the supply of oil under the earths crust was rather
limitless and not easily depleted. As a result, Hubberts predictions were
largely ignored. Oil production evidence from the United States, however,
indicated that Hubberts Peak indeed occurred, probably in 1971 (Heinberg, 2003). Despite this evidence, U.S. oil executives and policy makers
continued to downplay the signicance of a peak oil crisis, and did little
to prepare for the era where oil supplies would dwindle. This began to
change in the early 1990s, when a number of studies on peak oil were
published or presented at conferences and made available on the Internet.
208
B i g P ri s on s, B i g D ream s
209
210
B i g P ri s on s, B i g D ream s
8.1. Peak-Oil Estimates, Four Models for World Oil Production, http://www.hubbertpeak.com/summary.htm
211
administrators join together NOW and put policies in place that will
stave off this possibility.
To change the prison system is not extraordinarily difcult. What will
be difcult is to change the opinions of the public and lawmakers, and to
convince them that such changes are needed. Lawmakers and the public
need to be taught to recognize that our country must institute a correctional system based on offender reform rather than deterrence and incapacitation (Cullen and Gilbert, 1982; Clear, 1996), and that we ought to
adopt correctional models found in nations with small, effective prison systems to deal with crime reduction, escalating prison size and costs, looming energy supply problems, and global warming. There is a need, in other
words, to unite criminal justice policy and philosophy with green theories
and practices (e.g., see Frank and Lynch, 1992; Lynch and Stretesky, 2003;
South, 1998; South and Beirne, 1998; Berine and South, 2006).
To be sure, changing the prison system so that it can be smaller will
require adopting a new philosophy of punishment. The data reviewed
in previous chapters indicates that the U.S. prison system is not effective especially given its extraordinary size. What we must keep in mind
when considering prison reform is that the majority are sent back into
society, not reformed by being locked away. As noted in an early chapter,
about 70 percent of former inmates return to crime after their release
from prison. Clearly, this fact indicates that the current strategy behind
punishment is not effectively correcting their behavior. To be sure, these
inmates are being punished, but at what cost? These costs are difcult to
calculate because they involve estimating wasted energy, global warming
effects, wasted resources employed to incarcerated inmates who are not
affected by imprisonment, and the continued costs of crime, as well as
many intangible effects on the lives of individuals who are incarcerated
and their families (Clear, 1996).
Reducing the size of the prison population will also require changing laws and policing practices; for economic, energy, and environmental reasons, the police department of the future can no longer afford to
be built around large, centralized police stations and the police cruiser.
Thus, while the remainder of this chapter examines prisons and energy
resources, I recognize that reforming prisons alone is an insufcient reform, and that the entire criminal justice apparatus needs to be revised
consistent with energy and environmental conservation concerns.
212
B i g P ri s on s, B i g D ream s
I m p ri sonm e nt and th e O i l C ri s i s
The American prison system was designed solely with reference to
assumptions about crime control that failed to take energy consumption issues into consideration. Consistent with the U.S. consumer cultural orientation, the American prison system has experienced continual
expansion over the past thirty years and become a behemoth that is now
the worlds largest. Americas environmentally outdated correctional system of isolated, high-capacity prisons presents numerous challenges in an
era where the energy needed to operate and secure those facilities will
become more limited and costly.
Financially, the rising price of fossil fuels will increase a variety of
prison system costs. An endless array of everyday commodities found in
prisons are made using oil (e.g., safety glass, detergents, linoleum, plastic
products, clothing, fertilizers, pesticides, ink, cameras, clothing, etc.). The
prices and supply of these widely used commodities, as well as the supply
and price of food, will increase as oil becomes less available.8
The large spaces enclosed within prison walls will present a nancial
burden in terms of increased heating and cooling expenses as the availability of fossil fuels declines. There are two options: plan for the use
of alternative heating and cooling sources or, as some will undoubtedly
argue, turn off prison heating and cooling systems. The latter response
is unrealistic in both humanitarian and prison security terms. Unheated
inmates in prison systems in cold climates can be expected to die; hot
and agitated inmates are more likely to act aggressively, or perhaps riot.
In addition, the lack of heating and cooling would pose a burden to staff
and make recruiting correctional personnel even more difcult.
Transportation costs to large, out-of-the-way prisons that characterize the American prison will also rise. In addition, since tarred roadways
are constructed from oil and rock, repairing them will become more
costly. Road-repair funds are likely to be concentrated where they are
most neededin urban areasleaving rural roads leading to prisons in
disrepair, which will drive up vehicle repair costs, and, in extreme cases,
may become a security issue.
Taken together, the future of rising heating, cooling, transportation,
food, clothing, and security costs associated with the declining supply
of fossil fuels will make prison operation costs rise exponentially in a
213
214
B i g P ri s on s, B i g D ream s
215
216
B i g P ri s on s, B i g D ream s
217
how the American criminal justice system will meet the challenge of
an oil-decient future. In addition to offering some preliminary ideas
concerning how the criminal justice can be reorganized and redesigned
in an energy-efcient manner, this analysis has also drawn attention to
related key issues. First, there is the concern of guring out the t between various philosophies of doing justice and energy-efcient policies.
Energy efciency may require that philosophies of justice are shifted
away from the current preoccupation with incapacitation and deterrence, which have served to create a large, energy-inefcient system of
justice that also offers little evidence of its effectiveness as a crime control
strategy (e.g., Clear, 1994; Rose and Clear, 1998; Austin and Irwin, 2003;
Welch, 2004b).
It is always a bad idea to attempt to predict the futureone is
sure to be wrong on some if not many accounts. The decline of the oil
era, however, looms large on the horizon. The folly is to think that it
will not emerge, and that it is not necessary to plan the path for a new
criminal justice and correctional system that takes energy consumption
into account.
Today, the United States imports 60 percent of its oil from foreign
nations, having already reached its own oil peak in the early 1970s. The
U.S. fossil-fuel economy and social organization is, with each passing
year, becoming more dependent on the supply of oil from foreign nations. More extensive drilling within U.S. boundaries will not solve that
problem because oil is running out worldwide, while world energy demands are increasing annually. The George W. Bush White House has
denied the possibility that we may run out of oil, and has placed greater
emphasis on fossil fuel as energy sources.9 According to available science,
this is a grand mistake. At the national level, energy policy is headed in
the wrong direction. This does not mean that state and local criminal
justice planners and researchers must follow this policy blindly, or close
their eyes and hope that the inevitable energy crisis does not occur. It is
in our best interest to recognize the end of the oil era, and to plan for a
new, energy-efcient criminal justice system now.
On the basis of scientic evidence, it is clear that the worlds remaining oil reserve will be seriously threatened or depleted by mid-century
if current levels of oil use are not curtailed. Not only must we begin to
reform basis economic production, we must also begin to redesign basic
Table 8.1
Alternative Energy Examples in the Criminal Justice System
219
Chapte r 9
A Consuming Culture
The history of the American prison system is an effort to perfect the use of the penal apparatus the Quakers introduced
in Philadelphia meant for the reform of the criminal offender. America,
more so than another other nation, has relied upon the prison as a means
of responding to criminals, and has expanded this apparatus far beyond
the level found in other nations. But, throughout its history, especially in
the modern era, the U.S. prison system has not lived up to the lofty ideals
of the Quakers.
In the United States, the prison began as a means of reform, and
was later adopted to fulll other functions that included deterrence
and incapacitation. Along the way, the American prison served industry
as a center of prot-making activity under the inuence of the Auburn model and later during the industrial prison era of the early 1900s
(Barnes and Teeters, 1945). Perhaps one reason the American prison
system is so big is that it lacks a clear, consistent, and appropriate purpose. It has been used to meet the goals of several different philosophies,
and it has been used to meet some of these goals despite the fact that
existing evidence suggests that it has been unable to fulll the functions
specied by those philosophies, especially with respect to deterrence
and incapacitation. While the American prison system is certainly much
larger than the prison systems found in other nations, it nevertheless
does not appear big enough to deter or incapacitate a signicant portion of the criminal population.
The problem may be that prisons fail to address the causes of crime.
Indeed, the data on the relationship between imprisonment and crime
reviewed here suggests as much, since the two trends are rarely found to
exist together in the expected direction, at the predicted magnitude, or
in any persistent manner. We will return to this idea below.
220
A Consuming Culture
221
The failure of the American prison system means three things. First,
there is no simple connection between crime and punishment. Second,
the absence of this simple connection means that the problem of crime
cannot be solved simply by manipulating the level of imprisonment.
Third, the failures of the American system of imprisonment makes alternative explanations for its expansion appear more legitimate and, perhaps, logical (e.g., models of economy or race control). These alternatives
only become clear when the assumed relationship between crime and
imprisonment is discarded, and the economic-, class-, and race-control
dimensions of the American system of imprisonment are revealed.
The American prison system is, and may be, troubled by other issues
as well. To be sure, as discussed in the previous chapter, policy makers
must begin to take energy issues into account when determining prison
policy. The massive prison apparatus currently in use in the United States
cannot be sustained in the future if predictions of an energy strapped
future come to pass.
Consp i c uous Con sum p ti on
One must also consider that the American prison system is the sign of
a society that has taken conspicuous consumption, which Veblen (1899)
identied as a cultural characteristic of American society, to an extreme.
In simple terms, conspicuous consumption involves the purchase of expensive commodities and services in an effort to create public displays of
both wealth and social status. In Veblens analysis, examples of conspicuous
consumption often involved extravagant or unnecessary purchases of
commodities and services in ways that communicated ones social status to
others.Veblen described this as a cultural rather than a personal tendency,
meaning that it had widespread origins and effects, and that the effort to
communicate status through this mechanism was widely understood and
especially relevant in capitalist cultures. Thus, conspicuous consumption
can be seen as being motivated by cultural and economic structures. Since
Veblen viewed conspicuous consumption as a societal feature, this idea has
applications beyond the analysis of the consumption patterns of individuals. Indeed, if Veblen was correct, social institutions might also participate
in and contribute to reinforcing patterns of conspicuous consumption.
Interpreted in this way, it is possible to view the massive expansion of U.S.
prisons as an institutional example of conspicuous consumption.
222
B i g P ri s on s, B i g D ream s
A Consuming Culture
223
224
B i g P ri s on s, B i g D ream s
system. This may have to do with the fact that they would have to explain why our nation has produced such a big prison system if it fails in
its mission of crime control. To do so might require looking at some ugly
options. As other studies and data presented in earlier chapters suggest,
Americas prison system is part of the class- and race-control structure
that has been built in American society. American prisons overwhelmingly house the poor and racial minorities. This is not, as was illustrated
earlier, because they pose the greatest threat to American society in terms
of nancial loses, threats to our personal or national security, or in terms
of the level of violence these populations commit. Clearly, the various
crimes of the upper-class person with money, power, and statuswhitecollar crime, environmental crimes, corporate crimes, governmental
crimes, and crimes against foreign nationsare much more costly on
these accounts than ordinary street crimes. Yet these are precisely the
people who are not found in Americans prisons.
R ethi nk i ng Cri m e and P un i sh m e nt
In our society, punishment is widespread. But so too is crime. Historically, this was not always the case. In previous eras of U.S. history,
lower rates of crime and lower rates of imprisonment and punishment
coexisted. The changing nature of crime and punishment in the United
States may be due to changes in the nature of the social and economic
structure. It has also been demonstrated that low rates of incarceration
do not themselves produce high rates of crime. This was demonstrated
by examining cross-cultural, time series, and cross-state data. In addition,
previous chapters also demonstrated that high rates of incarceration do
not produce low rates of offending.
Faced with these facts, we are left with a problem. That problem,
stated simply, is the identication of the causes of crime, and how to
generate public policies that address those causes.
In various literaturesfrom sociology, criminology, economics, public policy, biology, psychologya wide variety of answers to the question
what causes crime? have been offered. The search for the answer to this
question has consumed the lives of many scholars, and has lled libraries
with prospective answers. To keep the current discussion manageable, I
will address only a few relevant facts and factors (several factors I omit
are discussed by Reiman, 2003).
A Consuming Culture
225
226
B i g P ri s on s, B i g D ream s
A Consuming Culture
227
equity and where achievement was not tied to the ability to accumulate
material possessions (Sahlin, 1972).
The kinds of policies that would engender the changes outlined
above have not been implemented in the United States, and certainly
have not been entertained as a crime control strategy. To be sure, these
are ambitious goals that would benet a greater number of people than
they would disadvantage, which is the mark of utilitarian social philosophy used in other areas of U.S. policy. But America is an ambitious nation. These are by no means surere solutions to the problem of crime.
We have, however, lived too long with a punitive approach to crime that
does not work. It is time to terminate Americas big prison experiment
and look elsewhere for the solutions to the problem of crime.
Conc ludi ng Comm e nts
The American prison system is the oldest and biggest in the world.
It challenges our nation on a daily basis. This system of punishment is
in need of a drastic overhaul. It is costly, and it fails to meet its mission.
Its existence fools the American public into thinking that politicians are
doing something about crime when, in effect, by failing to reform the
prison system, those politicians are doing nothing at all about crime.
Prisons dont need to be bigger, to house more inmates, to expand.
What is needed is an effective crime response. The history of the past
three decades in the United States has proven beyond a shadow of a
doubt that prison expansion is not the way to control crime. We must
admit that our national prison policy has been wrong, and that prisons
are not an effective crime control strategy. The problem of crime is
much more complex than the simple crime-punishment assumptions
America embraces.
It is easy enough to display the various forms of evidence that support this view and which have made up the contents of this book. It is
quite another, however, to have the courage and conviction to act on
these data and to reform a system that has dominated the American response to crime for more than two centuries, and which over the past
three decades, more than in any other era, has become the symbol of
Americas failing crime control efforts.
N ote s
Chapte r 2
1
U.S.
England
Australia
Canada
Netherlands
Scotland
Sweden
Switzerland
Average
7
96.6
56.3
(58)
71.8
(74)
44.5
(46)
26.1
(27)
55.5
(58)
29.8
(31)
37.0
(38)
35.0
19.3
(55)
34.3
(98)
9.0
(26)
12.7
(36)
8
(23)
8.2
(23)
19.1
(55)
20.7
8.3
(40)
27.9
(135)
1.6
(8)
8.5
(41)
5.4
(26)
4.0
(19)
9.4
(45)
76.4
40.0
(52)
72.0
(94)
24.2
(31)
12.4
(16)
29.6
(39)
29.8
(39)
35.6
(47)
40.4
14.0
(35)
36.2
(90)
5.0
(12)
5.0
(12)
17.2
(43)
5.4
(13)
11.7
(29)
115.9
76.5
(66)
82.8
(72)
106.6
(92)
17.4
(15)
81.4
(70)
36.9
(32)
49.7
(43)
250.0
179.9
(72)
178.3
(71)
120.8
(48)
100.4
(40)
191.6
(77)
94.5
(38)
96.6
(39)
18.2
(52)
10.7
(51)
40
(52)
16.2
(40)
70.9
(61)
151.5
(61)
This table was adapted from: Farrington, David P., Patrick Langan, and Michael Tonry, 2004,
Cross National Studies in Crime and Justice, Bureau of Justice Statistics, NCJ 200988, Washington, DC: U.S. Department of Justice, http://www.ojp.usdoj.gov/bjs/pub/ pdf/cnscj.pdf. Data
for England/Wales, Canada, Netherlands, Scotland, and Switzerland are from 1999. Data for U.S.,
Australia, and Sweden are from 1996. Percentage scores are rounded to nearest whole number.
229
230
Chapter 4
1. The Durkheimian position on crime and punishment has a number of interesting ramications and implications, and has inuenced both the development of theories of crime (e.g., anomie, social disorganization, social control
theory) and punishment (e.g., stability of punishment hypothesis). There is,
however, one assumption contained within Durkheim that is most difcult to
explain, and upon which much of his discussion of crime and punishment is
hinged. That assumption has to do with the average or acceptable level of
crime in a society.
Durkheim posited that crime was normal, and that the normality of crime
could be proven by it widespread existence. In this view, all healthy societies
had crime. These crimes helped mark the social boundary of acceptable and
unacceptable behavior. For Durkheim, a healthy society also had a given level
of crime, or what he referred to as the average level of crime. For Durkheim,
this average level of crime could be determined by measuring the amount of
crime normally found in a society. While Durkheim is clear on these denitional matters, he is not at all clear on the empirical aspects of this issue.
For example, how is the normal or average amount of crime in a society
measured? Clearly, to obtain an average, this measure must be taken over a series of years. But, how many years are needed to determine this average? If the
231
time period is short, and crime is rising or falling, does this imply that a society
in this condition is normal? Because crime within a given nation may not be
stable, average, or normal over any shorter period of time, should the measurement of crime be an average across similar countries? Or should the measure
be made across a diverse array of nations to discover the social average?
It would appear from the procedures employed by Durkheim that his preference was to mix measurements across nations over a short period of time,
and from this, derive the general rule or average. In part, Durkheims point
was that an acceptable or average level of crime was not a fabrication, but a
fact that could be ascertained using scientic study.
For Durkheim, measuring crime was important because it could tell us
something about the general health of a society. The societies that had too
much or too little crime were in some way unhealthy. So, too, was the
society in which the rate of crime changed rapidly, either by expansion or
contraction. Rapid changes in the rate of crime indicated that the social order had shifted, and that people were either now overly constrained (rapidly
contracting rate of crime) or too unrestrained (rapidly rising rate of crime)
by norms and values. It has generally been argued that rapidly changing social
structures can produce crime through anomie, a state of normlessness (e.g.,
see Merton, 1938, 1968; Messner and Rosenfeld, 2001), or social disorganization (e.g., Shaw and McKay, 1942).
Durkheims view of punishment was, not surprisingly, the same as his view
of crime. A society had a normal or average level of punishment, and this could
be ascertained. Likewise, the punishment of the criminal helped maintain the
moral boundaries of society. But, like crime, there could be too much (or too
little) punishment, and rapid changes in punishment also signaled rapid social
change and trouble in other segments of social organization.
Interestingly, Durkheim argued that the primary function of punishment
was to help demarcate social boundaries, that punishment was a reaction to
crime, and that punishment could not be used for other purposes, such as
deterrence, although it could be said to play a role in social defense. As a social
defense mechanism, punishment defended secular order, but not as a repressive
mechanism as it is used in the United States; rather, its defense of secular order
was ideological, or operated to reinforce agreed upon norms and values. As
an outcome of crime, punishment did not, in this view, determine crime, and
efforts to manipulate crime by increasing penalty were sure to fail. It is easy to
understand why Durkheim believed that this was truenormal societies had
a given level of crime, and either could not repress it through punishment, or,
in choosing to attempt to do so, created other social problems.
If Durkheim was correct, his view provides a broad-ranging theoretical
critique of the use of imprisonment in the United States. The effort to control crime through punishment is, rst, impossible, and, second, indicative of
a society with other signicant social problems. The growth of the punitive
response is not, in reality, a response to crime, but an inappropriate response
other social problems, such as widespread economic inequality. The United
States, for instance, has the highest level of economic inequality of Westernized nations.
2. If, indeed, the punitive public exists in large number, its existence requires
some explanation, especially if theories of punishment or explanations of
punishment are going to credit public opinion with inuencing penal policy
232
233
234
Chapte r 5
1. Several factors inuenced this movement. First, some were concerned that the
identier radical was or could be used against those who were so labeled in
an attempt to deny them tenure, promotion, and job opportunities. Second,
radical theorists were being widely inuenced by European scholarship written by a group identied as critical theorists (and later, by postmodernists and
deconstructionists). The term radical did not t within the conventions of
these views. Third, a growing number of researchers who had been identied
as radicals objected to what they saw as this views single-minded commitment to how economic and social-class issues inuenced crime and justice.
These researchers rightfully pointed out that forms of inequality inherent in
U.S. race and gender relations also had important associations with crime and
justice. In some respect, what each of these groups had pointed out was that
economic marginalization and social-class concerns were not the only forms
of inequality that ought to be associated with the term radical criminology.
In an effort to broaden the appeal of this view and assuage widespread feelings of academic marginalization experienced by a signicant cross-section of
radical criminologists, criminologists who shared a commitment to examining the impact of inequality on crime and justice renamed themselves critical criminologists.
On a personal note, I still refer to myself as a radical criminologist. While
my own research includes issues such as racial and gender issues found in the
critical approach, I continue to also emphasize class issues that have largely
disappeared from critical criminological literature. In addition, the various
critical and postmodern views do not, in my view, place the needed emphasis
on economic and social change as a remedy to crime and justice problems.
Consequently, I prefer the term radical criminologist because it implies a concentration on economic and class issues, as well as the need for solutions to
social problems to be based in economic reform.
Chapte r 6
Prison Ef f ects
1. In reality, only 0.7 cocaine users are reported as Hispanic. This gure was
rounded to simplify the example that is provided here. Rounding this gure
to ten causes the number of whites and blacks in the sample of 100 cocaine
235
Chapte r 7
1. The data used in table 7.1 were calculated using changes in the number of
crimes and the number of people in prison. This table could also be constructed using changes in the rate of index crimes per 100,000 population and
the rate of imprisonment per 100,000 population. Both analyses produce similar results. While the rate-based data results are slightly different and produce
a slightly stronger CSE effect, the rate based analysis also failed to support
the contentions of deterrence or incapacitative approaches. To illustrate this
slightly greater effect, the Pearson R correlation between the percent change
in the number of crimes and the percent change in number imprisoned was
-0.129 (p = 0.497), while the Pearson R between the change in the rate of
crime and the percent change in imprisonment was -0.297 (p = 0.147). In
neither case is the relationship statistically signicant.
Chapter 8
1. As a criminologist, I have not been trained as a geologist or as an environmental scientist. My knowledge of these issues is derived from reading literature on these topics. As in other research literatures, there are claims made on
both sides of each issue. Some literature (the peak oil literature) suggests that
the world oil supply will peak shortly, and that at current levels of consumption or
expected increases in consumption, and given the lack of a viable alternative
energy source, the world oil supply will be depleted as early as 2040. With
respect to environmental issues, there is substantial evidence of global warming, and extensive health effects on humans and animals from toxic chemicals
(for summary of my views, see Lynch and Stretesky, 2003; Lynch and Michalowski, 2006; Burns and Lynch, 2004).
2. To be sure, there are a number of alternative energy sources. Here, we have
referred primarily to solar, wind, and geothermal energy forms as fossil-fuel
alternatives. Many continue to see a future energy economy operated on fossil
fuels made available through altered technology. These alternatives, however,
have severe limitations at the present time (Rifkin, 2002). One alternative is
coal fuels. Estimates on the life of the worlds coal reserve vary widely from
67 to 300 years. This life span depends on the form in which coal energy is
usedsolid or liqueed. Liqueed coal could replace oil, but there are dangers
to this approach. Liqueed coal, for example, has a much higher concentration of carbon atoms than other fossil fuels, and will accelerate global warming. Natural gas reserve estimates are slightly higher than those for oil. Each
of these natural resources, however, is limited, and each contributes to other
environmental problems. An alternative fuel that may also be useful for conversion into electricity is biomass, which has lower, more controllable levels of
carbon dioxide emission (90 percent lower), and is a sustainable fuel. Biomass
farms could supply a signicant portion of U.S. energy needs, and it could also
promote a revitalized rural economy (Union of Concerned Scientists, http://
www.ucsusa.org/clean_energy/renewable_energy/ page.cfm?pageID=78).
236
Emerging technologies that make more efcient use of fossil fuels should
not be overlooked. These technologies differ from fossil-fuel conversion
strategies (e.g., liqueed coal). For example, plug-in electric cars with highpowered lithium-ion batteries (see http://www.calcars.org) are more efcient that the current hybrid technology used in vehicles such as the Toyota
Prius. Traditionally, the objection to plug-in vehicles is that they did not
solve global warming because the electricity required to charge the batteries
came from coal-powered electrical plants. The electricity needed to fuel a
plug-in, however, generates less CO2 than gasoline per distance traveled. On
average, the burning of 1 gallon of gasoline will yield about 19.6 pounds of
carbon dioxide (see information presented by the Carbon Dioxide Information Analysis Center, an ofce of the U.S. Department of Energy, http://
cdiac.esd.ornl.gov/pns/ faq.html). If the average person drives approximately
15,000 miles in a year, this translates into about 680 gallons of gasoline used,
or 13,328 pounds of carbon dioxide. For the same distance, a plug-in hybrid
operated without the gasoline engine would require the use of 6 pounds of
coal to generate the electricity needed to cover 24 miles, or 625 pounds of
coal (assuming the electricity is generated using coal, and not some combination of coal and wind or water power, which is becoming more common).
The combustion of 1 pound of coal produces approximately 17.16 pounds
of carbon dioxide (depending on the type of coal; Hong and Slatick, 1994),
yielding 10,725 pounds of carbon dioxide. Compared to the traditional gasoline combustion engine, the electric hybrid reduced the production of carbon dioxide by 2,603 pounds per vehicle (a reduction of 19.5 percent in
global warming gases).
One cannot discuss energy concerns without also addressing global warming. Again, there are two sides to the issue. In the United States, for example,
the political leaders who have staffed the Bush White House, with their backgrounds in the fossil-fuel industries, have done little to address global warming. As a group, they believe that global warming is not an issue, and have
refused to participate in, for example, the Kyoto Protocol on global warming.
On the other side stand thousands of scientists who study global warming, its
path, and consequences, and who have petitioned White House to change its
views and act on global warming (see, e.g., the Union of Concerned Scientists
petition, http://www.ucsusa.org).
Before continuing, it is useful to note my position on the end of oil or
the peak oil issue, and global warming, because this view obviously affects
the discussion that follows. First, peak oil and global warming are intimately
connected. Peak oil is good news for global warming to the extent that the
end of oil will help reduce global warming if it promotes the development of alternatives to fossil-fuel energy sources since the most signicant portion of global
warming results from burning fossil fuels (Goodstein, 2004). Second, however,
the end of oil will occur so far in the future that its effect on global warming
will occur much too late to impede the impact on the earths environment.
We need to act now to curb global warming. We also need to act now, as the
chairman of Chevron noted, to devise alterative energy sources and economies. These energy alternatives must, because of global warming, be nonfossil fuel in origin. That means that plans to convert the energy infrastructure to
natural gas, or increase the use of coal, or the extract coal oils and tars, or oil
from shale, are not useful energy strategies; they will exacerbate rather than
237
238
5.
6.
7.
8.
domestic crude oil production had dropped to less than half this amount:
4,761,000 barrels (U.S. Department of Energy, 2004a). At the same time that
U.S. oil production was declining, crude-oil consumption was rising, though
at lower rates in recent decades than in the 1950s through the 1970s (Duncan,
2005). Declining domestic reserves coupled with rising demand led to an
increased reliance on foreign crude oil. In 1960, the United States imported
1,815,000 barrels of foreign crude; by 2002, this total had increased more
than sixfold to 11,580,000 barrels (U.S. Department of Energy, 2004a). In
short, between 1960 and 2002, U.S. crude oil production fell over 32 percent,
while foreign crude oil imports had risen by more than 630 percent. It should
be noted that the increased reliance on foreign oil is not simply a result of
declining crude-oil production capacity in the United States, since production
capacity expanded during the period under examination (U.S. Department of
Energy, 2004b).
Recently (2005) the George W. Bush administrationwhich is closely connected to the oil industry (e.g., the White House Energy Task Force was
headed by Vice President Cheney, former CEO of Halliburton, one of the
worlds largest suppliers of products and services to the oil and gas industry)
opened the Arctic National Wildlife Refuge (ANWR) to oil exploration in
an effort to remedy the peak-oil problem. Historically, the oil found in this
region was designated as a military reserve and held for military emergencies.
Environmentally, the damage of this policy will be large. But, moreover, the
policy will not alleviate the problem it addressesa diminishing oil supply
in the United States. First, it is estimated that it will take up to ten years for
ANWR oil to come on line, which includes a point in time that is, in all
likelihood, after the world peak in oil production. By that time, dwindling oil
supplies will have caused the price of domestic oil to rise dramatically, and
is likely to mean that the search for alternative energy sources will be well
underway. Second, the amount of oil in the ANWR is estimated to be about
10 billion barrels, or about the amount of oil the United States is estimated
to consume in 1618 months once ANWR oil is available. Thus, opening the
ANWR to oil exploration will, at best, add two years of oil supply to domestic markets, while, at the same time, expanding the environmental destructive
outcomes (e.g., global warming) of burning oil. In short, these policies are
inadequate responses to both the energy problems facing the United States
and the issue of global warming.
Natural gas reserves are also limited, and currently have a peak of ten to
fteen years beyond the oil peak. Once placed into more widespread use,
natural gas availability will decline dramatically. The best alternative is the
development of renewable energy sources that would make human populations independent of fossil fuels, and which have lower environmental costs or
impacts (Rifkin, 2002).
The CIA has been tracking world oil production and reserve levels for the
past decade. For example, a paper written by Senator Richard Lugar and former CIA chief James Woolsey that appeared in the January 1999 issue of
Foreign Affairs predicted the world oil crisis and urged the creation of new U.S.
policies to deal with this situation.
As fertilizers and pesticides produced from oil become more expensive, the cost
of operating farm equipment and shipping food escalates. As prices for pesticides, herbicides, fertilizers, and equipment operations rise, and oil becomes
239
scare, human labor will be reintroduced to farms to replace the tasks performed
by herbicides and farm machinery, which will help maintain the food supply,
but will not lower costs. Thus, the simple act of feeding and clothing large
populations of prison inmates may become very expensive.
9. Indeed, the Presidential Energy Task Force has recommended energy policies
that reinforce fossil-fuel dependence (see the Energy Task Force Report, Reliable, Affordable, and Environmentally Sound Energy for Americas Future,
http://www.whitehouse.gov/energy).
Re f e re nce s
Agnew, Robert. 2005. Why Do Criminals Offend? A General Theory of Crime and
Delinquency. Belmont, CA: Roxbury.
American Iron and Steel Institute. (n.d.). The New Steel Industry. AISI FACT
SHEETS. Washington, DC: AISI.
Austin, James, John Clark, Patricia Hardyman, and D. Alan Henry. 1999. The Impact of Three Strikes and Youre Out. Punishment & Society 1,2: 131162.
Austin, James and John Irwin. 2003. Its About Time: Americas Imprisonment Binge.
Belmont, CA: Wadsworth.
Austin, Roy L. and Mark D. Allen. 2000. Racial Disparities in Arrest Rates as an
Explanation for Racial Disparities in Commitment to Pennsylvanias Prisons.
Journal of Research in Crime and Delinquency 37,2: 200220.
Bandura, Alfred. 1986. Social Foundations of Thought and Action: A Social Cognitive
Theory. Englewood Cliffs, NJ: Prentice Hall.
. 1977. Social Learning Theory. Englewood Cliffs, NY: Prentice Hall.
Barnes, Harry and Negley Teeters. 1945. New Horizons in Criminology. New York:
Prentice Hall.
Bartlett, Donald L. and James B. Steele. 1996. America:What Went Wrong? Riverside,
NJ: Andrews McMeel Publishing.
Becker, Gary S. 1968. Crime and Punishment: An Economic Approach. The Journal of Political Economy 76,2: 169217.
Beckett, Katherine. 1997. Making Crime Pay: Law and Order in Contemporary American Politics. Oxford: Oxford University Press.
Beckett, Katherine and Bruce Western. 2001. Governing Social Marginality: Welfare, Incarceration and the Transformation of State Policy. Punishment & Society
3,1: 4359.
Beirne, Piers and Nigel South (eds). 2006. Green Criminology. Hampshire, UK: Ashgate.
Black, William, Kitty Calavita, and Henry N. Pontell. 1995. The Savings and Loan
Debacle of the 1980s: White-Collar Crime or Risky Business? Law and Policy
17,1: 2355.
Blumstein, Alfred. 1998. U.S. Criminal Justice Conundrum: Rising Prison Populations and Stable Crime Rates. Crime and Delinquency 44,1: 127135.
Bonczar, Thomas P. 2003. Prevalent of Imprisonment in the U.S. Population,
19742001. Bureau of Justice Statistics, Special Report. NCJ 197976. Washington DC: U.S. Department of Justice.
Bonczar, Thomas P. and Allen. J. Beck. 1997. Lifetime Likelihood of Going to
State or Federal Prison. Bureau of Justice Statistics, NCJ-160092. Washington
DC: U.S. Department of Justice.
241
242
References
Brownstein, Henry. 2000. The Social Reality of Violence and Violent Crime. Boston:
Allyn and Bacon.
Burdon, Roy H. 2004. The Suffering Gene: Environmental Threats to Our Health.
London: Zed Books.
. 1999. Genes and the Environment. Boca Raton, FL: CRC Press.
Burns, Ronald G. and Michael J. Lynch. 2004. Environmental Crime: A Sourcebook.
New York: LFB Scholarly Publishers.
Calavita, Kitty and Henry N. Pontell. 1994. The State and White-Collar Crime:
Saving the Savings and Loans. Law and Society Review 28,2: 297324.
Calavita, Kitty, Robert Tillman, and Henry N. Pontell. 1997. The Savings and
Loan Debacle, Financial Crime, and the State. Annual Review of Sociology
23: 1938.
Campbell, Colin J. 2003. The Coming Oil Crisis. Santa Cruz, CA: R.B. Swenson.
. 2001. Peak Oil: A Turning Point for Mankind. M. King Hubbert Newsletter. 2,1. M. King Hubbert Center, Colorado School of Mines. http://hubbert.
mines.edu/news/Campbell_012.pdf.
. 1998. Running Out of Gas: This Time the Wolf Is Coming. The National
Interest 51: 4755.
Campbell, Colon J. and Jean Laherrere. 1998. The End of Cheap Oil? Scientic
American March: 7883. http://dieoff.org/page140.htm.
Campbell, Donald T. and Julian C. Stanley. 1962. Experimental and Quasi-Experimental Designs for Research. Chicago: Rand McNally.
Cantor, David and Kenneth C. Land. 2001. Unemployment and Crime Rate
Fluctuations: A Comment on Greenberg. Journal of Quantitative Criminology
17,4: 329344.
. 1985. Unemployment and Crime Rates in the Post-World War II United
States: A Theoretical and Empirical Analysis. American Sociological Review 50,3:
317332.
Carlson, Susan M. and Raymond J. Michalowski. 1997. Crime, Unemployment
and Social Structures of Accumulation: An Investigation into Historical Contingency. Justice Quarterly 14.2: 209241.
Carnoy, Martin. 1994. Faded Dreams: The Politics and Economics of Race in America.
Cambridge, UK: Cambridge University Press.
Chambliss, William J. and Robert B. Seidman. 1982. Law, Order and Power. New
York: Addison-Wesley.
Christianson, Scott. 1981. Our Black Prisons. Journal of Research in Crime and
Delinquency 27,1:364375.
Christie, Nils. 1994. Crime Control as Industry, Towards Gulags, Western Style. New
York: Routledge.
Clear, Todd. 2003. The Problem with Addition by Subtraction: The PrisonCrime Relationship in Low Income Communities. In M. Mauer and M.
Chesney-Lind (eds), Invisible Punishment: The Collateral Consequences of Mass Imprisonment. New York: The New Press.
. Science and the Punishment/Control Movement. Social Pathology 2,1:
122.
. 1994. Harm in American Penology: Offenders, Victims and Their Communities.
Albany: State University of New York Press.
References
243
Clear, Todd and Dina Rose. 1999. When Neighbors Go to Jail: Impact on Attitudes About Formal and Informal Social Control. National Institute of Justice,
Research Preview. July (FS 0002 43). Washington DC: NIJ.
. 1998. Incarceration, Social Capital and Crime: Examining the Unintended Consequences of Incarceration. Criminology 36,3: 441479.
Clear, Todd, Dina Rose, and Judith A. Ryder. 2001. Incarceration and Community: The Problem of Removing and Returning Offenders. Crime and Delinquency 47,3: 335351.
Clear, Todd, Dina Rose, and Elin Waring. 2003. Coercive Mobility and Crime: A
Preliminary Examination of Concentrated Incarceration and Social Disorganization. Justice Quarterly 20,1: 3364.
Cochran, John K., Mitchell B. Chamlin, and Mark Seth.1994. Deterrence or
Brutalization? An Assessment of Oklahomas Return to Capital Punishment.
Criminology 32:107134.
Corrections Corporation of America, 2005. Annual Report. http://www.shareholder.com/cxw/.
Cullen, Francis T. 2005. Twelve People Who Saved Rehabilitation: How the Science of Criminology Made a Difference. Criminology 43,1: 142.
Cullen, Francis T., Bonnie S. Fisher, and Brandon K. Applegate. 2000. Public
Opinion About Punishment and Corrections. In Michael Tonry (ed), Crime
and Justice: A Review of Research. Chicago: University of Chicago Press.
Cullen, Francis T. and Paul Gendreau. 2000. Assessing Correctional Rehabilitation: Policy, Practice, and Prospects. In Julie Horney (ed), Criminal Justice 2000:
Volume 3Policies, Processes, and Decisions of the Criminal Justice System. Washington DC: U.S. Department of Justice, National Institute of Justice.
. 1989. The Effectiveness of Correctional Rehabilitation: Reconsidering
the Nothing Works Debate. In Lynne Goodstein and Doris MacKenzie (eds),
American Prisons: Issues in Research and Policy. New York: Plenum.
Cullen, Francis T. and Karen E. Gilbert. 1982. Reafrming Rehabilitation. Cincinnati:
Anderson.
Cullen, Francis T., William J. Maakestad, and Gray Cavender. 1987. Corporate Crime
Under Attack: The Ford Pinto Case and Beyond. Cincinnati: Anderson.
Deffeyes, Kenneth S. 2001. Hubberts Peak: The Impending World Oil Shortage. Princeton, NJ: Princeton University Press.
DeFina, R. H. and T. M. Arvanites. 2002. The Weak Effect of Imprisonment on
Crime: 19711998. Social Science Quarterly 83,3: 635653.
Devine, Joel A., Joseph F. Shelley, and M. Dwayne Smith. 1988. Macroeconomic
and Social Control Policy Inuences on Crime Rate Changes, 19481985.
American Sociological Review 53,3: 407420.
DiIulio, John. 1994. Let Em Rot. Wall Street Journal, January 26: A 14.
Doob, Anthony N. and Cheryl Marie Webster. 2003. Sentence Severity and Crime:
Accepting the Null Hypothesis. Crime and Justice 30: 143195.
DuBoff, Richard B. 2003. U.S. Hegemony: Continuing Decline, Enduring Danger. Monthly Review 55,7. http://www.monthlyreview.org/1203duboff.htm.
Duncan, Richard C. 2005. The Peak of World Oil Production and the Road to
the Olduvai Gorge. Institute on Man and Energy. http://www.hubbertpeak.
com/Duncan/olduvai2000.htm.
244
References
Durkheim, Emile. 1897 [1977]. Suicide. New York: The Free Press.
. 1895 [1964]. The Rules of Sociological Method. New York: The Free Press.
. 1893 [1968]. The Division of Labor in Society. New York: The Free Press.
Durose, Matthew R. and Patrick A. Langan. 2001. Court Sentencing of Convicted Felons, 1998. NCJ-190637. Washington DC: U.S. Department of Justice. http://www.ojp.usdoj.gov/bjs/pub/pdf/scscf98.pdf.
Economic Policy Institute, Pulling Apart. 2006. http://www.epi.org/content.cfm/
studies_pulling_apart_2006.
Energy Information Administration (EIA), U.S. Department of Energy. 2003.
2002 Summary Statistics, California. http://www.eia.doe.gov/cneaf/electricity/st_proles/california.pdf.
Farrington, David P. 2000. Cross-National Comparative Studies in Criminology.
In H. N. Pontell and D. Shichor (eds), Contemporary Issues in Crime and Criminal
Justice: Essays in Honor of Gilbert Geis. Upper Saddle River, NJ: Prentice Hall.
Farrington, David P., Patrick Langan, and Michael Tonry. 2004. Cross National
Studies in Crime and Justice. Bureau of Justice Statistics. NCJ 200988. Washington DC: U.S. Department of Justice. http://www.ojp.usdoj.gov/bjs/pub/
pdf/cnscj.pdf.
Federal Bureau of Investigation. 2002. Uniform Crime Report, 2002. http://www.
fbi.gov/ucr/02cius.htm.
Federal Reserve. 2003. Recent Changes in U.S. Family Finances: Evidence from
the 1998 and 2001 Survey of Consumer Finances. http://www.federalreserve.
gov/pubs/bulletin/2003/0103Lead.pdf.
Feeley, Malcolm. 2004. Crime, Social Order and the Rise of Neo-Conservative
Politics. Theoretical Criminology 7,1: 111130.
Fellner, J. and M. Mauer. 1998. Losing the Vote: The Impact of Felon Dienfranchisement
Laws in the United States. New York: Human Rights Watch and the Sentencing
Project.
Flanagan, Timothy J. 1982. Risk and the Timing of Recidivism in Three Cohorts
of Prison Releasees. Criminal Justice Review 7,2: 3445.
Flanagan Timothy J., Pauline Gasdow Brennan, and Debra Cohen. 1991. Crime
Control Ideology and Policy Positions in a State Legislature (Revisited). Criminal Justice Policy Review 5,3: 183206.
Flanagan, Timothy J., Deborah Cohen, and Pauline G. Brennan. 1993. Crime
Control Ideology Among New York State Legislators. Legislative Studies Quarterly 18,3: 411422.
Flanagan, Timothy J. and Edmund F. McGarrell. 1986. Attitudes of New York Legislators Toward Crime and Criminal Justice: A Report of the State Legislator Survey
1985. Albany: Hindelang Criminal Justice Research Center, State University of
New York at Albany.
Flanagan, Timothy J., Edmund F. McGarrell, and Alan J. Lizotte. 1989. Ideology
and Crime Control Policy Positions in a State Legislature. Journal of Criminal
Justice 17:81101.
Flanagan, William G. 2003. Dirty Rotten CEOs: How Business Leaders Are Fleecing
America. New York: Citadel Press.
Foucault, Michel. 1979. Discipline and Punish: The Birth of the Prison. New York:
Vintage.
References
245
Fox, James Alan and Marianne W. Zawitz. 2006. Homicide Trends in the United
States. U.S. Department of Justice, Bureau of Justice Statistics. http://www.ojp.
usdoj.gov/bjs/pub/pdf/htius.pdf.
Frank, Nancy and Michael J. Lynch. 1992. Corporate Crime, Corporate Violence. Albany, NY: Harrow and Heston.
Frank, Robert H. and Phillip Cook. 1996. The Winner-Take-All Society:Why the Few
at the Top Get So Much More Than the Rest of Us. New York: Penguin.
Friedrichs, David. 2004. Trusted Criminals. Belmont, CA: Wadsworth.
Gainsborough, Jenni and Marc Mauer. 2000. Diminishing Returns: Crime and
Incarcerations in the 1990s. The Sentencing Project. http://www.sentencingproject.org/pdfs/9039.pdf.
Gendreau, Paul, Claire Goggin, and Francis T. Cullen. 1999. The Effect of Prison
Sentences on Recidivism. Public Works and Government Services of Canada. Cat.
No.: J4287/1999E; ISBN: 0662284062.
Goldfarb, Ronald, 1969. Prison: The National Poorhouse. The New Republic, December 13.
Goodstein, David. 2004. Out of Gas: The End of the Age of Oil. New York: W.W.
Norton.
Gordon, Diana. 1994. The Return of the Dangerous Classes: Prohibition and Policy
Politics. New York: W.W. Norton.
Grasmick, Harold G. and G. J. Bryjack. 1980 The Deterrent Effects of Perceived
Severity of Punishment. Social Forces 59,2: 471491.
Greenberg, David F. 2001a. Novus Ordo Saeclorum?: A Commentary on Downes,
and on Beckett and Western. Punishment & Society 3,1: 8193.
. 2001b. Time Series Analysis of Crime Rates. Journal of Quantitative
Criminology 17,4: 281327.
. 1990. The Cost-Benet Analysis of Imprisonment. Social Justice 17,4: 4975.
. 1989. Imprisonment and Unemployment: A Comment. Journal of Quantitative Criminology 5,2: 187191.
. 1977. The Dynamics of Oscillatory Punishment Processes. The Journal of
Criminal Law and Criminology 68,4: 643651.
Greenberg, David F. and Valerie West. 2001. State Prison Populations and their
Growth, 19711991. Criminology 39,3: 615654.
Hacker, James. 1995. Two Nations: Black and White, Separate, Hostile and Unequal.
New York: Ballantine.
Hagan, John. 1997. Crime and Disrepute. Thousand Oaks, CA: Pine Forge Press.
Harrison, Paige M. and Allen J. Beck. 2006. Prison and Jail Inmates at Mid-Year
2005. Washington DC: U.S. Department of Justice. http://www.ojp.usdoj.gov/
bjs/pub/pdf/pjim05.pdf.
. 2003. Prisoners in 2002. Washington DC: U.S. Department of Justice.
http://www.ojp.usdoj.gov/bjs/pub/pdf/p02.pdf.
Hay, Douglas, Peter Linebaugh, John G. Rule, E. P. Thompson, and Cal Winslow.
1976. Albions Fatal Tree: Crime and Society in Eighteenth Century England. New
York: Pantheon.
Hayward, Steven and Lance T. Izumi. 1996. Crime and Punishment in California: Are
We Too Tough or Not Tough Enough? San Francisco: Pacic Research Institute for
Public Policy.
246
References
Heinberg, Richard. 2003. The Partys Over: Oil, War and the Fate of Industrial Societies. Gabriola Island, BC: New World Publishers.
Henry, Jules. 1965. Culture Against Man. New York: Random House.
Hong, B. D. and E. R. Slatick. 1994. Carbon Dioxide Emission Factors for Coal.
Quarterly Coal Report. Washington DC: DOE/EIA. http://www.eia.doe.gov/
cneaf/coal/quarterly/co2_article/co2.html.
Hufngton, Arianna 2003. Pigs at the Trough: How Corporate Greed and Political Corruption are Undermining America. New York: Crown.
Irwin, John. 2005. The Warehouse Prison: Disposal of the New Dangerous Class. Los
Angeles: Roxbury.
Jacobs, David and Jason T. Carmichael. 2001. The Politics of Punishment Across
Time and Space: A Pooled Time Series Analysis of Imprisonment Rates. Social
Forces 80,1: 6181.
Jacobson, Michael. 2005. Downsizing Prisons: How to Reduce Crime and End Mass
Incarceration. New York: New York University Press.
Johnston, C. Wayne. 1999. Incapacitation and Deterrent Effects of Incarceration: A
Pennsylvania Study. Corrections Compendium 24,12: 111.
Keister, Lisa A. 2000. Wealth in America: Trends in Wealth Inequality. Cambridge, UK:
Cambridge University Press.
Kessler, Daniel and Steven D. Levitt. 1999. Using Sentence Enhancements to Distinguish Between Deterrence and Incapacitation. Journal of Law and Economics
42,1: 343363.
Kleeper, Steven and Daniel Nagin. 1989. The Deterrent Effect of Perceived Certainty and Severity of Punishment Revisited Criminology 27,4: 721746.
Kovandzic, Tomislav V.; Thomas B. Marvell, Lynne M. Vieraitis, and Carlisle E.
Moody. 2004. When Prisoners Get Out: The Impact of Prison Releases on
Homicide Rates, 19751999. Criminal Justice Policy Review 15,2: 212228.
Kristof, Nicholas D. 2006. 100-M.P.G. Cars: Its a Start. New York Times. February
5: 4, 13.
Leigh, J. Paul, Steven Markovitz, Marianne Fahs, and Phillip Landigren. 2000. The Cost
of Occupational Injuries and Illnesses. Ann Arbor: University of Michigan Press.
Lippke, Richard L. 2002. Crime Reduction and the Length of Prison Sentences.
Law & Policy 24,1: 1735.
Lovins, Amory B., E. K. Datta, O. E. Bustnes, J. G. Koomey, and N. J. Glasgow. 2005.
Winning the Oil Endgame: Innovations for Prots, Jobs and Security. Snowmass, CO:
Rocky Mountain Institute.
Lynch, James P. 1988. A Comparison of Imprisonment in the United States, Canada, England, and West Germany: A Limited Test of the Punitiveness Hypothesis. Journal of Criminal Law and Criminology 79: 180217.
. 1993. A Cross-National Comparison of the Length of Custodial Sentences for Serious Crimes. Justice Quarterly 10,4: 639659.
Lynch, Michael J. 2004. Towards a Radical Ecology of Urban Violence: Integrating Medical, Epidemiological, Environmental and Criminological Research On
Class, Race, Lead (Pb) and Crime. In Margaret Zahn, Henry Brownstein, and
Shelly Jackson (eds), Violence: From Theory to Research. Cincinnati: Anderson.
. 2002. Misleading Evidence and the Misguided Attempt to Generate Racial Proles of Criminals: Correcting Fallacies and Calculations Concerning
References
247
Race and Crime in Taylor and Whitneys Analysis of Racial Proling. The Mankind Quarterly 42,3: 313330.
. 2000. J. Philippe Rushton on Crime: An Examination and Critique of
the Explanation of Crime in Race, Evolution and Behavior. Social Pathology.
6,3: 228244.
. 1999. Beating a Dead Horse: Is There Any Basic Empirical Evidence
for a Deterrent Effect of Imprisonment? Crime, Law and Social Change 31:
347362.
. 1996. Race, Class, Gender and Criminology: Structured Choices and the
Life Course. In M. Schwartz and D. Milovanovic (eds), Gender, Race and Class
in Criminology. Hamden, CT: Garland.
. 1988. The Extraction of Surplus Value, Crime and Punishment: A Preliminary Empirical Analysis for the U.S. Contemporary Crises 12: 329344.
. 1987. Quantitative Analysis and Marxist Criminology: Old Answers to a
Dilemma in Marxist Criminology. Crime and Social Justice 29: 110127.
Lynch, Michael J., W. Byron Groves, and Alan Lizotte. 1994. The Rate of Surplus
Value and Crime: Theoretical and Empirical Examination of Marxian Economic Theory and Criminology. Crime, Law and Social Change 21,1: 1548.
Lynch, Michael J., Michael J. Hogan, and Paul Stretesky. 1999. A Further Look at
Long Cycles, Legislation and Crime. Justice Quarterly 16,2: 431450.
Lynch, Michael J. and Raymond J. Michalowski. 2006. Primer in Radical Criminology, 4th Edition. Monsey, NY: Criminal Justice Press.
Lynch, Michael J. Raymond J. Michalowski, and W. Byron Groves. 2000. The New
Primer in Radical Criminology. Monsey, NY: Criminal Justice Press.
Lynch, Michael J. and E. Britt Patterson (eds). 1996. Justice with Prejudice. Albany,
NY: Harrow and Heston.
. 1991. Race and Criminal Justice. Albany, NY: Harrow and Heston.
Lynch, Michael J. and Amie Schuck. 2003. Picasso as Criminologist: The Abstract
Art of Racial Proling. In M. Free (ed), Racial Issues in Criminal Justice: The
Case of African Americans. Westport, CT: Praeger.
Lynch, Michael J. and Paul B. Stretesky. 2003. The Meaning of Green: Contrasting
Criminological Perspectives. Theoretical Criminology 7,2: 217238.
Lynch, Michael J. and Paul B. Stretesky. 2001. Toxic Crimes: Examining Corporate Victimization of the General Public Employing Medical and Epidemiological Evidence. Critical Criminology 10,3: 153172.
Lynch, Michael J., Paul B. Stretesky and Ronald G. Burns. 2004a. Determinants
of Environmental Law Violation Fines Against Oil Reneries: Race, Ethnicity,
Income and Aggregation Effects. Society and Natural Resources 17,4: 333347.
. 2004b. Slippery Business: Race, Class and Legal Determinants of Penalties Against Petroleum Reneries. Journal of Black Studies 34,3: 421440.
Marion, Nancy and Rick Farmer. 2004. A Preliminary Examination of Presidential Anticrime Promises. Criminal Justice Review 29,1: 173195.
Marvel, Thomas B. and Carlisle E. Moody. 1998. The Impact of Out-of-State
Prison Population on State Homicide Rates: Displacement and Free-Rider Effects. Criminology 36,3: 513535.
. 1997. The Impact of Prison Growth on Homicide. Homicide Studies 1,3:
205233.
248
References
References
249
Pettit, Becky and Bruce Western. 2004. Mass Imprisonment and the Life Course
Race and Class Inequality in U.S. Incarceration. American Sociological Review
69,2: 151169.
Pinkney, Alfonso. 1994. The Myth of Black Progress. Cambridge, UK: Cambridge
University Press.
Reiman, Jeffrey. 2003. The Rich Get Richer and the Poor Get Prison. Boston: Allyn
and Bacon.
Reynolds, Morgan O. 2000. Does Punishment Work to Reduce Crime? Heartland Institute. http://www.heartland.org/Article.cfm?artId=125.
. 1997. Crime and Punishment in America: 1997 Update. Dallas: National
Center for Policy Analysis.
Ridley, Matt. 2000. Genome:The Autobiography of a Species in 23 Chapters. New York:
Harper Perennial.
Rifkin, Jeremy. 2002. The Hydrogen Economy. New York: Jeremy P. Tarcher/Putnam.
Roberts, Paul. 2004. The End of Oil: On the Edge of a Perilous New World. London:
Bloomsbury Publishing.
. 1998. Incarceration, Social Capital and Crime: Examining the Unintended Consequences of Incarceration. Criminology 36,3: 441479.
Rose, Dina and Todd Clear. 1995. Incarceration, Social Capital and Crime:
Examining the Unintended Consequences of Imprisonment. Criminology
36,3:441479.
Ross, Jeffrey Ian. 2000. Grants R-US: Inside a Federal Grant Making Agency.
American Behavioral Scientist 43,10: 17041723.
Ruddell, Rick. 2005. Social Disruption, State Priorities, and Minority Threat: A
Cross-National Study of Imprisonment. Punishment & Society 7,1: 728.
Rusche, Georg, and Otto Kirchheimer. 1939[1968]. Punishment and Social Structure.
New York: Russell and Russell.
Ryan, William. 1982. Equality. New York:Vintage.
Sahlins, Marshall. 1972. Stone Age Economics. Piscataway, NJ: Aldine Transaction.
Sapp, Stephen G. 2002. Incomplete Knowledge and Attitude-Behavior Consistency. Social Behavior and Personality 30: 3744.
Savelsberg, Jaochim J., Lara Cleveland, and Ryan D. King. 2004. Institutional Environments and Scholarly Work: American Criminology, 19511993. Social
Forces 82,4: 12751302.
Shaw, Clifford and Henry McKay. 1942. Juvenile Delinquency and Urban Areas: a
Study of Rates of Delinquents in Relation to Differential Characteristics of Local Communities in American Cities. Chicago: University of Chicago Press.
Sheldon, Randall. 2000. Controlling the Dangerous Classes. Boston: Allyn and Bacon.
Simon, David R. and D. S. Eitzen. 2000. Elite Deviance. Boston: Allyn and Bacon.
Smith, Kevin B. 2004. The Politics of Punishment: Evaluating Political Explanations of Incarceration Rates. The Journal of Politics 66,3: 925938.
. 1997. Explaining Variation in State-Level Homicide Rates: Does Crime
Policy Pay? Journal of Politics 59,2: 350367.
Smith, Paula, Claire Goggin, and Paul Gendreau. 2002. The Effect of Prison Sentences
and Intermediate Sanctions on Recidivism: General effects and Individual Differences.
Public Works and Government Services of Canada. Cat No.: JS42103/2002.
ISBN: 0662664752.
250
References
South, Nigel. 1998. A Green Field for Criminology? A Proposal for a Perspective.
Theoretical Criminology 2,2: 211233.
South, Nigel and Piers Beirne. 1998. Editors Introduction. Theoretical Criminology 2,2: 147148.
Stretesky, Paul B. and Michael J. Lynch. 2004. The Relationship Between Lead
and Crime. Journal of Health and Social Behavior 45,2: 214229.
. 2003. Environmental Hazards and School Segregation in Hillsborough,
19871999. The Sociological Quarterly 43,4: 553573.
. 2001. The Relationship Between Lead and Homicide. Archives of Pediatric and Adolescent Medicine 155,5: 579582.
. 1999a. Corporate Environmental Violence and Racism. Crime, Law and
Social Change 30,2: 163184.
. 1999b. Environmental Justice and the Prediction of Distance to Accidental Chemical Releases in Hillsborough County, Florida. Social Science Quarterly
80,4: 830846.
Sutton, John. 2004. The Political Economy of Imprisonment in Afuent Western
Democracies, 19601990. American Sociological Review 69,22: 170189.
. 2000. Imprisonment and Social Classication in Five Commonlaw Democracies, 19551985. American Journal of Sociology 106,2: 350386.
Tillman, Robert and Henry N. Pontell. 1995. Organizations and Fraud in the
Savings and Loan Industry. Social Forces 73,4: 14391463.
Tillman, Robert, Henry N. Pontell, and Kitty Calavita. 1997. Criminalizing
White-Collar Misconduct: Determinants of Prosecution in Savings and Loan
Fraud Cases. Crime, Law and Social Change 26,1: 5376.
Tyler, Tom. 1990. Why People Obey the Law. New Haven, CT:Yale University Press.
U.S. Department of Energy. 2004a. Petroleum Overview, 19492003. http://www.
eia.doe.gov/emeu/aer/txt/ptb0501.html.
U.S. Department of Energy. 2004b. Renery Capacity and Utilization, 19492003.
http://www.eia.doe.gov/emeu/aer/txt/ptb0509.html.
Veblen, Thorsten. 1899 [1967]. Theory of the Leisure Class. New York:Viking Press.
Visher, Christie A. 1987. Incapacitation and Crime Control: Does a Lock em
Up Strategy Reduce Crime? Justice Quarterly 4,4: 513543.
Vold, George B., Thomas J. Bernard, and Jeffrey B. Snipes. 2002. Theoretical Criminology. New York: Oxford University Press.
Vogel, Richard. 2003. Capitalism and Incarceration Revisited: Factors Affecting
Incarceration Rates. Monthly Review 55,4: 3855. http://www.monthlyreview.
org/0903vogel.htm.
Weinrath, Michael and John Gartrell. 2001. Specic Deterrence and Sentence
Length: The Case of Drunk Drivers. Journal of Contemporary Criminal Justice 17,2: 105122.
Weisberg, David, E. Waring, and Ellen Chayet. 1995. Specic Deterrence in a
Sample of Offenders Convicted of White-Collar Crimes. Criminology 33,4:
587605.
Weitzer, Ronald and Charles E. Kubrin. 2004. Breaking News: How Local TV
News and Real-World Conditions Affect Fear of Crime. Justice Quarterly 21,3:
497520.
Welch, Michael. 2004a. Ironies of Imprisonment. Thousand Oaks, CA: Sage.
References
251
I nde x
253
254
Index
DeFina, R. H., 11
democracy, ix, 15, 47, 110, 168,
Denton, Nancy, 193194, 197
deregulation: airlines, 125; trucking,
125
deterrence, 1112, 14, 3236, 4042,
80, 150151, 173174, 177179,
187, 192, 231, 235; Canada, 7; homicide, 89, 11; white collar criminals,
12. See also incapacitation, prisons
DiIulio, John, 5, 66
Dinkins, (Mayor) David, 79
Doob, Anthony N., 11
drug use patterns, race and ethnicity,
161166
Durkheim, Emile, 8485, 230231
Halliburton, 98
Harm in American Penology, 12
Harrison, Paige M., 162, 171
Hispanics: drug use and imprisonment,
162164; prison population, 198
homicide, 98, 105106, 107; offenders
in prison, 149150, 229
Hubbert, M. King, 207208
Hubberts Curve (Hubberts Peak), 207
human behavior, predicting, 90, 118
Index
255
256
Index
257
Index
About th e Author