The Prison Journal

Supplement to 91(3) 120 S137S

2011 SAGE Publications
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0032885511415236
http://tpj.sagepub.com
415236TPJ91310.1177/00328855114152
36Welsh and FarringtonThe Prison Journal
2011 SAGE Publications
Reprints and permission:
sagepub.com/journalsPermissions.nav
1
Northeastern University, Boston
2
Cambridge University, Cambridge, UK
Corresponding Author:
Brandon C. Welsh, 360 Huntington Avenue, Boston, MA 02115
Email: b.welsh@neu.edu
The Benefits and Costs
of Early Prevention
Compared With
Imprisonment: Toward
Evidence-Based Policy
Brandon C. Welsh
1
and David P. Farrington
2
Abstract
Spending on prisons is at an all time high in the United States. These costs are
proving unsustainable, causing a substantial drain on state and local govern-
ment budgets and diverting scarce resources from critical sectors like educa-
tion, health, and welfare. Early prevention of delinquency and later offending
represents a viable public policy option to the present focus on punishment,
and some governments are beginning to take notice. This article reviews the
highest-quality research on the economic benefits and costs of early preven-
tion compared with imprisonment. It finds that there is a growing body of
scientific research that shows that early prevention is an effective and worth-
while investment of public resources; a number of landmark, comprehensive
economic studies demonstrate the value of early prevention and other alter-
natives compared to imprisonment; and a new crop of research studies are
forging important insights on the economics of prison and alternative strate-
gies. Implications for policy are discussed.
Keywords
early prevention, effectiveness, benet-cost analysis, evidence-based policy
Welsh and Farrington 121S
We begin this article with what may very well be the best example in this coun-
try of government spending priorities shifting from prison to prevention. In
2007, the Washington State legislature abandoned plans to build one of two
large-capacity prisons and in its place approved a sizeable spending package
(US$48 million) on evidence-based crime prevention and intervention pro-
grams (Drake, Aos, & Miller, 2009). What makes this story even more impor-
tant is that it was a political decision that was made on the basis of scientific
research. This research showed that other strategies were more effective
than incarceration and would generate substantial financial savings to the
Washington State government and taxpayers.
Many scholars, policy makers, practitioners, and even some politicians,
know by now the pioneering research undertaken by Steve Aos and his col-
leagues at the Washington State Institute for Public Policy. Fewer know about
how this research has influenced policy. In 1997, the Washington State legis-
lature commissioned the institute to assess the effectiveness and economic
efficiency of a range of crime prevention and criminal justice programs with
the aim to identify interventions that reduce crime and lower total costs to
taxpayers and crime victims (Aos, Barnoski, & Lieb, 1998, p. 1). The research-
ers referred to their methodological approach as bottom-line financial anal-
ysis, which they considered to parallel the approach used by investors who
study rates of return on financial investments. The research began with a litera-
ture review of high-quality programs. A five-step analytical model was then
used to estimate each programs economic contribution.
1
What started out as a highly rigorous yet fairly modest policy research
initiative soon turned into the most comprehensive approach to develop evidence-
based crime policy in the United States and one revered by other states and
even countries (Aldhous, 2007; Greenwood, 2006). Following the institutes first
set of reports, the legislature authorized a number of system-level randomized
controlled trials of the most effective and cost-beneficial juvenile and adult
intervention programs. The results of these trials helped to refine local prac-
tice and service delivery. By 2006, the institute had systematically reviewed and
analyzed 571 of the highest-quality evaluations of crime prevention and
criminal justice programs, estimated the benefits and costs of effective pro-
grams, and projected the degree to which alternative portfolios of these pro-
grams could affect future prison construction needs, criminal justice costs, and
crime rates in Washington (Aos, Miller, & Drake, 2006, p. 1). This most
recent work was commissioned by the legislature to address the projected
need for two new state prisons by 2020 and possibly a third by 2030. Based
on a moderate-to-aggressive portfolio of evidence-based programs
(i.e., US$63-US$171 million expenditure in the first year), it was found that a
122S The Prison Journal Supplement to 91(3)
significant amount of future prison construction costs could be avoided,
about US$2 billion saved by taxpayers, and crime rates lowered slightly
(Aos et al., 2006, p. 16).
Let us be clear, the above researchand this article for that matterdoes
not say that prisons cannot reduce crime or that states should let offenders out
wholesale. Instead, it says in no uncertain terms that there are a number of
worthwhile alternatives to prison that can help state governments lower their
spending on prisons, improve the rate of return on spending on crime preven-
tion and criminal justice programs, and lower the burden of crime on its citi-
zens. Neither Aos and his colleagues nor the two of us just discovered this.
The scientific evidence on the effectiveness and economic efficiency of crime
prevention and intervention programs compared to prison has been accumulat-
ing for years. This started with landmark comprehensive studies carried out
in the 1980s and 1990s. It has continued to this day with a new crop of high-
quality studies by early childhood researchers, criminologists, and economists
and at policy research centers like the Washington State Institute for Public
Policy and the Vera Institute of Justices cost-benefit analysis unit.
In this article, we review the evidence base for the monetary costs and ben-
efits of early prevention compared to imprisonment. This comprises three
main topical areas of research: early childhood programs, landmark compre-
hensive studies, and a new crop of studies that are forging important insights
on the economics of prison and alternative strategies. It is not possible to cover
every study. Instead, we focus on the highest-quality research studies. In the
case of evaluations, this includes experimental and rigorous quasi-experimental
designs. Wherever possible, we also draw upon the most rigorous literature
reviews (i.e., systematic and meta-analysis) that include only high-quality
projects. This ensures that our conclusions are based on the best available
evidence. We begin with a brief overview of the costs of imprisonment.
Imprisonment Costs
In their recent report entitled The High Budgetary Cost of Incarceration, the
Center for Economic and Policy Research estimates that federal, state, and local
governments spent nearly US$75 billion on corrections in 2008. This trans-
lates into about US$700 for each household in the country. A large majority
of this expenditure goes toward incarceration. State and local governments
account for a little more than 90% of these expenditures (Schmitt, Warner, &
Gupta, 2010, p. 10). These costs have gone up considerably in recent years. In
the last decade, from 1999 to 2008, spending on corrections increased by more
than one fifth (21%) or US$13 billion (in inflation-adjusted dollars). In the
Welsh and Farrington 123S
past two decades, spending on corrections has more than doubled, from about
US$37 billion (in inflation-adjusted dollars) in 1989 (Schmitt et al., 2010).
So high are these costs that in Missouri, for example, judges at the sentenc-
ing phase of trials now consider the monetary costs of different sanctions
(Davey, 2010).
Most concerning is that this level of spending on prisons across the country
is causing a substantial drain on state and local government budgets and
diverting scarce resources from critical sectors like education, health, and
welfare. According to Wexler (2010), Corrections expenditures compete with
and diminish funding for education, public health, public safety, parks and
recreation, and programs specifically designed to reduce the prison popula-
tion. Only state spending on Medicaid outpaced spending on corrections dur-
ing the past two decades (Moore, 2009). Interestingly, Ellwood and Guetzkow
(2009) found that state spending on corrections had an adverse effect on spend-
ing on welfare, but not on other budget categories such as education and health.
In the next part, we discuss the public policy method of economic analysis.
Economic Analysis
Economic analysis aims to provide a fair and reliable assessment of the value
for money of a crime prevention or criminal justice program or policy. It can
be described as a policy tool that allows choices to be made between alterna-
tive uses of resources or alternative distributions of services (Knapp, 1997,
p. 11). It is a key part of evidence-based crime policy (Mears, 2010). As Drake
et al. (2009, p. 194) note, While determining whether a program reduces
crime remains the necessary first condition for rational public policy making,
an economic analysis constitutes the necessary additional condition for iden-
tifying viable and fiscally prudent options. Many criteria are used in economic
analysis. The most common is efficiency or value for money (achieving maxi-
mum outcomes from minimum inputs).
Of the two main techniques of economic analysisbenefit-cost and cost-
effectiveness analysisonly the former allows for an assessment of both
benefits and costs. A cost-effectiveness analysis can be referred to as an incom-
plete benefit-cost analysis. This is because no attempt is made to estimate the
monetary value of program effects (benefits or disbenefits), only resources
used (costs). For example, a cost-effectiveness analysis might specify how
many crimes were prevented per US$1,000 spent on a program. Another way
to think about how benefit-cost and cost-effectiveness analyses differ is that
cost-effectiveness analysis may help one decide among competing program
models, but it cannot show that the total effect was worth the cost of the pro-
gram (Weinrott, Jones, & Howard, 1982, p. 179), unlike benefit-cost analysis.
124S The Prison Journal Supplement to 91(3)
A benefit-cost analysis is a step-by-step process that follows a standard
set of procedures. There are six main steps: (1) define the scope of the analysis,
(2) obtain estimates of program effects, (3) estimate the monetary value of
costs and benefits, (4) calculate present value and assess profitability, (5)
describe the distribution of costs and benefits (an assessment of gains and
losses of program participants, the government or taxpayer, and crime victims),
and (6) conduct sensitivity analyses by varying the different assumptions
made (Barnett, 1996).
2
Two other key features of economic analysis require brief mention. First,
an economic analysis is an extension of an outcome or impact evaluation and
is only as defensible as the evaluation on which it is based. David Weimer and
Lee Friedman (1979, p. 264) recommended that economic analyses be lim-
ited to programs that have been evaluated with an experimental or strong
quasi-experimental design. Second, many perspectives can be taken in mea-
suring program costs and benefits. Some benefit-cost analyses adopt a society-
wide perspective that includes the major parties that can receive benefits or
incur costs, such as the government or taxpayer, crime victim, and program
participant. Other analyses may take a more narrow view, focusing on one or
two of these parties. The decision about which perspective to take has impor-
tant implications for evaluating the program, particularly if it is being funded
by public money; that is, if conclusions are to be drawn about the monetary
benefits or costs of a program to the public, the benefits or costs must be
those that the public will either receive or incur. In reporting on the benefit-
cost findings of the studies reviewed here, we have used, as far as possible,
a middle-of-the-road approachfocusing on the costs and benefits for the
government or taxpayer and crime victims.
Saving Children From a Life of Crime
Early childhood programs have wide appeal across a large spectrum of con-
stituencies. These programs help societys most vulnerable members. Their
explicit aims include the betterment of childrens immediate learning and
social and emotional competencies and the improvement of childrens suc-
cess over the life course. Also, they are implemented at a time when children
are most impressionable and, hence, receptive to interventions (Duncan &
Magnuson, 2004).
In our book, Saving Children from a Life of Crime (Farrington & Welsh,
2007), we proposed a risk-focused, evidence-based strategy to make the early
prevention of delinquency and later offending a top priority of U.S. crime
policy. We called for the creation of a national council to take the lead on
Welsh and Farrington 125S
early crime prevention. It could provide technical assistance, skills, and knowl-
edge to state and local agencies in implementing prevention programs; provide
funding, continuity, coordination, and monitoring of local programs; provide
training in prevention science for local people; and maintain high standards
for evaluation research.
This was not some pie-in-the-sky idea. It is a well-developed practice that
is used with much success in Canada, Australia, and many Western European
countries. Even more important, it is founded on two separate but interrelated
research developments that focus on the importance of early intervention pol-
icy and programs to improve childrens life chances and prevent them from
embarking on a life of crime. First, after decades of rigorous research, a great
deal is known about the early causes of delinquency and later offending.
Second, there is a growing body of high-quality scientific evidence on the effec-
tiveness of early prevention programs designed to tackle these causes. Included
in this evidence base are a number of programs that have carried out benefit-
cost analyses, which show that early crime prevention pays back program costs
and produces substantial returns to society. Other meta-analyses and system-
atic reviews include the same programs (see Manning, Homel, & Smith, 2010;
Piquero, Farrington, Welsh, Tremblay, & Jennings, 2009).
One of these programs is the famous Perry Preschool Project (Schweinhart
et al., 2005). Carried out in Ypsilanti, Michigan, 123 children were allocated
(approximately at random) to experimental and control groups. The experimen-
tal children attended a daily preschool program, backed up by weekly home
visits, usually lasting 2 years when children were aged 3 and 4 years. The aim
of the plan-do-review program was to provide intellectual stimulation,
to increase thinking and reasoning abilities, and to increase later school
achievement.
The program had long-term benefits, extending up to age 40 (the latest
follow-up). Compared to the control group, program group members had sig-
nificantly fewer lifetime arrests for violent crimes (32% vs. 48%), property
crimes (36% vs. 58%), and drug crimes (14% vs. 34%), and were significantly
less likely to be arrested 5 or more times (36% vs. 55%). Improvements were
also recorded in many other important life-course outcomes. For example, sig-
nificantly higher levels of schooling (77% vs. 60% graduating from high school),
better records of employment (76% vs. 62%), and higher annual incomes were
reported by the program group compared to the controls.
These improvements translated into substantial dollar savings. A benefit-
cost analysis at age 40 found that the Perry program produced just more than
US$17 of benefit per dollar of cost, with three quarters (76%) of this being
returned to the general publicin the form of savings in crime, education,
126S The Prison Journal Supplement to 91(3)
and welfare, and increased tax revenueand 24% benefiting program partici-
pants. James Heckman and his colleagues (Heckman, Moon, Pinto, Savelyev,
& Yavitz, 2010) reanalysis of the Perry program found a slightly lower but still
very high return of US$7-US$12 of benefit per dollar of cost.
Like Perry, the Child-Parent Center program in Chicago provides disad-
vantaged children, aged 3 to 4, with a high-quality, active learning preschool
supplemented by family support. It also provides children with the educational
enrichment component into elementary school, up to age 9. Begun in 1967,
this program is the second oldest federal preschool program in the United
States (after Head Start) and the oldest extended early intervention program.
It is located in 24 centers in high-poverty neighborhoods across Chicago.
A rigorous nonrandomized controlled design was used to evaluate the pro-
gram, with a sample of more than 1,500 children. By age 24, the experimental
groupcompared to their control counterpartshad significantly lower rates
of felony arrest (17% vs. 21%), lower rates of incarceration (21% vs. 26%),
and higher rates of school completion (71% vs. 64%; Reynolds et al., 2007). A
benefit-cost analysis of the program, which included some outcomes measured
up to age 21 (crime was measured at age 18), found that, for every dollar spent
on the program, more than US$7 was saved to society. These savings took the
form of averted government expenditures on remedial education and the jus-
tice system as well as increased economic well-being of program participants
(Reynolds, Temple, & Ou, 2003). The latter benefit accounted for the largest
share (46%) of total monetary benefits; savings to crime victims and the crimi-
nal justice system accounted for the next largest share (28%) of benefits.
Another program, the Carolina Abecedarian Project, targeted children born
to low-income, multirisk families. A sample of 111 children aged 3, mostly
African American (98%), were randomly assigned either to receive full-time
preschool childcare (focusing on the development of cognitive and language
skills) or not. Families of children in both the experimental and control groups
received supportive social services as needed (Campbell, Ramey, Pungello,
Sparling, & Miller-Johnson, 2002). At age 21, 104 of the participants were
interviewed, and it was found that fewer of the experimentals compared to
the controls (but not significantly so) reported being convicted for a misde-
meanor offense (14% vs. 18%) or a felony offense (8% vs. 12%) or had been
incarcerated (14% vs. 21%). Also, significantly fewer of the experimental
participants were regular marijuana users, significantly fewer had become a
teenage parent, significantly more had attended college or university, and
they had significantly higher-status jobs. A benefit-cost analysis found that,
for every dollar spent on the program, US$2.50 was saved to society (Barnett
& Masse, 2007).
Welsh and Farrington 127S
Head Start, long considered the nations most important early childhood
program, which reaches about half of all impoverished children, has been the
subject of a number of evaluations and benefit-cost analyses. A large-scale
study of the long-term effects of Head Start, by Eliana Garces and her col-
leagues (Garces, Thomas, & Currie, 2002), found that children who attended
Head Start (at ages 3 to 5) were significantly less likely to report being
arrested or referred to court for a crime by ages 18 to 30 compared to their
siblings who did not attend the program. In one of the first benefit-cost analy-
ses, Janet Currie (2001) found that Head Starts short- and medium-term ben-
efits could offset between 40% and 60% of its costs and the addition of a small
fraction of long-term benefits would make the program a worthwhile public
investment. Jens Ludwig and Deborah Phillipss (2008) review of the benefits
and costs of Head Start, which drew upon the first report of the programs
national randomized experiment (the national impact study) and other studies,
concluded that the program was a worthwhile investment of taxpayer dollars
in both the short- and long-term. An analysis of the final report of the national
impact study by the National Forum on Early Childhood Policy and Programs
(2010) drew attention to key gains experienced by Head Start children com-
pared to their control counterparts and the need for a long-term benefit-cost
analysis of the program based on experimental data.
Benefit-cost analyses of a number of other early prevention programs tar-
geted at different stages of the life course provide further evidence of the profit-
ability of investing in this approach. Two of these programs include the
Nurse-Family Partnership (Eckenrode et al., 2010) and the Seattle Social
Development Project (Hawkins, Kosterman, Catalano, Hill, & Abbott, 2008).
Benefit-to-cost ratios of these programs range from 2.9:1 to 4.1:1, depending
on who conducted the analysis.
Landmark Comprehensive Studies
The individual benefit-cost analysis studies reviewed above provide important
insight on the economic efficiency of early childhood programs. What is per-
haps a more pressing issue facing policy makers today is an understanding of
economic efficiency of competing policy options; that is, compared to other
crime prevention and control strategies, which one provides the best economic
return for society? This section examines the first generation of comprehensive
studies that investigated this crucial policy question.
The most well known of these studies was conducted by the RAND
Corporation (Greenwood, Model, Rydell, & Chiesa, 1998). It assessed the
cost-effectiveness (i.e., serious crimes prevented per US$1 million spent, using
128S The Prison Journal Supplement to 91(3)
1993 dollars) of Californias new three-strikes law compared with four pre-
vention and intervention strategies with demonstrated efficacy in reducing
crime: a combination of home visits and day care, parent training, graduation
incentives, and monitoring and supervising delinquent youths. Each of these
alternatives was based on well-known experiments with small to moderately
large samples.
The first step in the modeling process was the estimation of program costs
and crime reduction effectiveness of the four alternative interventions. In order
to conduct fair comparisons among these interventions and with the three-
strikes law, three main penalties were assigned to the former. These were
(1) targeting: the proportion of the population targeted by the program who are
likely to become involved in criminal behavior (e.g., children of low-income,
teenage mothers for the home visit/day care program), (2) decay: the loss of
effectiveness after treatment ends, and (3) scale-up: the decrease in program
effects from expanding the program statewide (Greenwood et al., 1998, p. 16).
The study found that parent training and graduation incentives were the
most cost-effective of the five programs, whereas home visiting/day care was
the least cost-effective. The number of serious crimes prevented per US$1 million
was estimated at 258 for graduation incentives, 157 for parent training, 72 for
delinquent supervision, 60 for the three-strikes law, and 11 for home visiting/
day care.
3
It is important to comment briefly on the poor showing of the home
visiting/day care program. The small number of crimes prevented per US$1
million is attributable to two key factors (a third factor is discussed below): first,
the long-term delay in realizing an impact on crime, and, second, the high cost
of delivering the services, particularly the day care component, which was esti-
mated at US$6,000 per child per year over a 4-year period.
John Donohue and Peter Siegelman carried out an equally rigorous thought
experiment, which has some similarities to the Washington State Institutes
comparative benefit-cost model. The authors set out to investigate whether
the social resources that will be expended a decade or more from now on incar-
cerating todays youngsters could instead generate roughly comparable levels
of crime prevention if they were spent today on the most promising social
programs (1998, p. 31, emphasis in original). These included a range of early
prevention programs and the national Job Corps program, and most of the
estimates were based on well-known experiments or quasi-experiments with
small to large samples.
The first step involved estimating the crime reduction and cost associated
with a continuation of U.S. prison policy. On the basis of a 50% increase in the
prison population over a 15-year period (assumed from the level in December
1993 and trends at the time), it was estimated that this policy would cost
Welsh and Farrington 129S
US$5.6 to US$8 billion (in 1993 dollars) and result in a 5% to 15% reduction
in crime. The next steps involved estimating the percentage of the cohort of
3-year-olds who could be served by allocating the saved prison costs (the
US$5.6 to US$8 billion) to the different social programs and then estimating
the crime reduction benefits that could be achieved by selected programs
under a range of targeting conditions (i.e., the worst 6% of delinquents, males
only, and young Black males only). Two early prevention programs were
selected: the Perry Preschool Project and the Syracuse University Family
Development Research program (Lally, Mangione, & Honig, 1988). It is at this
stage that the authors apply a scaling-up penalty of 50% to each program. This
had the result of reducing Perrys crime reduction effectiveness from 40% to
20% and Syracuses effectiveness from 70% to 35%.
In the final analysis, it was found that both early prevention programs could
achieve reductions in crime that were within the range of what was expected
from a continuation of the prison policy of the day (5% to 15%) even if they
were allocated the lower-bound amount that would have been spent on prisons
(US$5.6 billion). This was considered the worse-case scenario. In the best-
case scenario, which did not include the scaling-up penalty and allocated the
upper-bound amount of prison spending (US$8 billion), the Perry and Syracuse
programs produced crime reductions of 21% and 26%, respectively.
Daniel Nagin (2001) reexamined these two studies. Although he praised
their innovative methods, one of his main concerns was that they failed to con-
sider the full potential benefits of the early prevention programs. By its very
nature, developmental crime prevention is designed to improve individual
functioning across multiple domains. Indeed, this is precisely what evalua-
tions of the included programs showed. But Greenwood et al. (1998) and
Donohue and Siegelman (1998) only considered the benefits from lower rates
of delinquent and criminal activity. Educational, employment, family, health,
and other important benefits were not taken into account, which had the effect
of greatly reducing the true economic return to society. This point is made
clear in another RAND study by Lynn Karoly and her colleagues (1998). In
the case of the Nurse-Family Partnership program, for example, which was
used in the study by Greenwood et al., an analysis of the distribution of ben-
efits accruing to government revealed that more than half (57%) of the bene-
fits were caused by a reduction in welfare costs, almost one quarter (23%) were
due to tax revenue from increased employment income, 20% were caused by
a reduction in juvenile and criminal justice costs, and less than 1% were due
to a reduction in health care services. These conclusions are based on a conser-
vative benefit-cost analysis that found that, in the home visits program, bene-
fits exceeded costs by a ratio of 4.1:1. A more recent benefit-cost analysis by
130S The Prison Journal Supplement to 91(3)
Donohue (2009) embraces this view; he concludes that there is reason to
believe that alternatives to incarceration might well be more socially attrac-
tive than our current reliance on incarceration as the predominant crime-
fighting strategy (p. 308).
New Research
A new crop of research has continued to investigate the crucial question of
which policy option provides the best economic return for society, as well as
to probe key economic issues central to this debate. For sure, the most impor-
tant body of new research in this area is the Washington State Institute for
Public Policys comparative benefit-cost model. Although work in this regard
began in the late 1990s, it still remains cutting edge. This is due to ongoing
refinements to the crime-based model, inclusion of additional outcomes (e.g.,
education, substance abuse, health), new analyses, and an expansion of its work
with the support of the Pew Charitable Trusts to aid other states to identify
evidence-based policy options to reduce crime and lower the taxpayer costs of
the criminal justice system (Aos & Drake, 2010, p. 1). As noted previously,
this research is already paying substantial dividends to the state government,
local governments, and taxpayers in Washington State.
Closely related to this research, a number of new benefit-cost analyses of
standard prison and other sentencing alternatives demonstrate that prisons are
not economically efficient (Bierie, 2009; Marsh & Fox, 2008; see also McDougall,
Cohen, Perry, & Swaray, 2006). It is only when prisons are reserved for the
most serious and violent offenders that their benefits begin to offset their
operating costs (Marsh, Fox, & Hedderman, 2009).
Another important line of research has involved the use of the methodol-
ogy of public opinion polling known as contingent valuation, which has many
advantages over conventional polling methods. The contingent valuation
approach allows for the comparison of respondents willingness to pay for
competing policy alternatives (Nagin, Piquero, Scott, & Steinberg, 2006,
p. 629). In a large-scale study of public preferences for responses to juvenile
offending in the Commonwealth of Pennsylvania, Nagin et al. found that the
public values early prevention and offender treatment significantly more than
increased incarceration. Households were willing to pay an average of US$126
in additional taxes on nurse home visitation programs to prevent delinquency
compared to US$81 on longer sentences. Households were also willing to pay
more in taxes for treatment than for longer sentences: US$98 versus US$81.
At the state level, public support for the prevention option translated into
US$601 million that hypothetically could be used to prevent delinquency com-
pared to US$387 million for longer sentences for juvenile offenders.
Welsh and Farrington 131S
In another contingent valuation study to gauge the publics preferences for
a range of alternative responses to crime, Cohen, Rust, and Steen (2006) found
that the public overwhelmingly supported the increased spending of tax dol-
lars on youth prevention programs compared to building more prisons. Public
support for spending more taxes on drug treatment for nonviolent offenders
as well as on police also ranked higher than support for building more prisons,
but not as high as for youth prevention programs. Many other studies show
public support for allocating money to early prevention rather than to building
more prisons (see Cullen, Vose, Jonson, & Unnever, 2007).
Recent studies on the costs of crime are also making important contribu-
tions to assessing the benefits and costs of early prevention compared to prison.
But it is important to note that high crime costs do not themselves suggest a
policy solution. For this we need to carry out an economic analysis, prefera-
bly a benefit-cost analysis. What these cost studies do is to provide information
on the unit costs of crime as well as the order of magnitude of the problem. Mark
Cohen and Alex Piquero (2009) estimated that the typical criminal career
over the juvenile and adult years costs society between US$2.1 and US$3.7
million (in 2007 dollars). They also estimated the dollar costs of the associated
problem behaviors of heavy drug use and dropping out of high school, bringing
the total societal cost of a high-risk youth to US$2.6 to US$4.4 million. In a
follow-on study that looked at the costs of crime across offender trajectories
for a real-life cohort in Philadelphia, Cohen, Piquero, and Jennings (2010)
found that the 200 high-rate chronic offenders cost society more than US$200
million (in 2007 dollars). Based on another real-life cohort of 500 Pittsburgh
boys, Welsh and his colleagues (2008) conservatively estimated that this group
caused a substantial burden of harm to society in the form of victimization
costs, ranging from US$89 to US$110 million (in 2000 dollars). From an early
age, the cohort was responsible for substantial crime victim losses, with these
losses mounting in the teen years.
Discussion and Conclusions
This article reviewed the highest-quality research on the costs and benefits of
early prevention compared to prison. It reaffirms the importance of economic
analysis in the development of evidence-based crime policy. It finds that a
growing body of economic analyses demonstrates that early crime prevention
can be a sound investment of public resources and an economically viable
alternative to prison. It also makes the case for more states to follow Washington
States lead, by investing in high-quality outcome and economic evaluation
research (to allow for valid comparisons among crime policy options), providing
132S The Prison Journal Supplement to 91(3)
an agency to conduct this research and make it transparent and understand-
able to legislators and their constituents (in the form of a nonpartisan research
institute), and following through with the adoption of crime prevention poli-
cies that are evidence-based. We recommend that the National Criminal Justice
Commission adopt this benefit-cost model as the foundation of its efforts to
promote sound, evidence-based policies.
There are, of course, challenges to adopting early prevention as a policy
for reducing crime. Some will argue that more attention needs to be paid to
intervening with criminally involved populations, of which there is an impres-
sive body of evidence-based programs (see Drake et al., 2009; Lipsey, 2009;
Lipsey & Cullen, 2007). We do not dispute this. Indeed, one of us (Farrington)
has long championed the research finding that it is never too early and never
too late to intervene to reduce criminal offending (Loeber & Farrington,
1998, 2011). What is sorely needed is a much stronger focus on prevention in
the early years to help ensure that there is a greater balance among prevention,
treatment, and punishment. We recommend that the commission give due
attention to the early prevention of criminal offending.
Another challenge that confronts prevention is the worry by politicians that
they may be perceived as soft on crime by supporting prevention instead of
law and order measures. Politicians think that citizens are punishment oriented,
in part, because of the shortcomings of traditional public opinion polling about
crime. Lawmakers often predicate their decisions about sentencing based
on flawed survey data. As shown above, the contingent valuation approach
provides a more textured and useful illumination of the publics views of
punishment and should be undertaken to better inform and support legisla-
tors decisions.
Yet another challenge is that the benefits from reduced crime will not be
apparent for a number of yearsat least until children reach adolescence. This
can conflict with the short time horizons of politicians; that is, programs that
show results only in the longer term can be less appealing to those who have
to face reelection every few years. One potential antidote to this situation is to
educate politicians about the many other benefits from prevention that accrue
to taxpayers in the interim. Crime prevention is often a spin-off benefit.
There is also the challenge that primary responsibility for (and spending
on) early crime prevention can fall between different government depart-
ments or be located in the wrong place altogether. Peter Greenwood (2006)
argues that Health and Human Services should be the lead federal department
for early prevention, whereas Justice is the more appropriate department for
serving the needs of high-risk and adjudicated youths. One argument in favor
of Health and Human Services is that the benefits of early prevention are not
Welsh and Farrington 133S
restricted to crime but include many other aspects of a successful or healthy
lifestyle, including education, employment, substance abuse, relationships,
and mental health.
The good news is that these challenges can be overcome. The economic
analysis results reviewed here show that benefits from early prevention begin
to accrue at an early age and cascade over multiple domains, that the public is
supportive of prevention even to the point of being willing to spend more in
taxes on it, and that legislators and other decision makers are beginning to real-
ize that they can no longer afford to rely on prisons to reduce crime. What is
needed now is the political will to cap prison spending and allocate greater
resources to early prevention. It is time to get smart on crime.
Acknowledgments
We are grateful to the editors, Hiro Yoshikawa, Holly Schindler, and Frank Cullen for
helpful comments.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research,
authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or pub-
lication of this article.
Notes
1. The first step of the model involves estimating each programs most likely practi-
cal application within the states justice or early intervention systems (Aos, Bar-
noski, & Lieb, 1998, p. 8). Step 2 looks at whether program results can be replicated
in the state. Step 3 involves an assessment of program costs, estimated on the basis
of what it would cost the Washington State government to implement a similar
program (if the program was not already operating in the state). Step 4 involves
monetizing each programs effects on crime. Savings to the criminal justice system
and crime victims are estimated. The final step involves calculating the economic
contribution of the program, expressed as a benefit-to-cost ratio. From this, pro-
grams can then be judged on their independent and comparative monetary value.
2. It is beyond the scope of this article to discuss each step, but interested readers
should consult reviews of this methodology as applied in the context of devel-
opmental and criminal justice crime prevention (see Welsh & Farrington, 2000).
In addition, for methodological features of benefit-cost analysis in general, see
Layard and Glaister (1994).
134S The Prison Journal Supplement to 91(3)
3. Crimes prevented were not calculated on a per annum basis. Instead they represent
the predicted total effect produced by each program on each California cohort (an
estimated 150,000 at-risk children born in California every year).
References
Aldhous, P. (2007, February 10). Applying science to prison overcrowding. New
Scientist. Retrieved from http://www.newscientist.com.
Aos, S., Barnoski, R., & Lieb, R. (1998). Preventive programs for young offenders:
Effective and cost-effective. Overcrowded Times, 9(2), 1, 7-11.
Aos, S., & Drake, E. K. (2010). Fight crime and save money. Olympia: Washington
State Institute for Public Policy.
Aos, S., Miller, M. G., & Drake, E. K. (2006). Evidence-based public policy options
to reduce future prison construction, criminal justice costs, and crime rates.
Olympia: Washington State Institute for Public Policy.
Barnett, W. S. (1996). Lives in the balance. Ypsilanti, MI: High/Scope Press.
Barnett, W. S., & Masse, L. N. (2007). Comparative benefit-cost analysis of the Abece-
darian program and its policy implications. Economics of Education Review, 26,
113-125.
Bierie, D. (2009). Cost matters: A randomized experiment comparing recidivism
between two styles of prisons. Journal of Experimental Criminology, 5, 371-397.
Campbell, F. A., Ramey, C. T., Pungello, E., Sparling, J., & Miller-Johnson, S. (2002).
Early childhood education: Young adult outcomes from the Abercedarian project.
Applied Developmental Science, 6, 42-57.
Cohen, M. A., & Piquero, A. R. (2009). New evidence on the monetary value of saving
a high risk youth. Journal of Quantitative Criminology, 25, 25-49.
Cohen, M. A., Piquero, A. R., & Jennings, W. G. (2010). Studying the costs of crime
across offender trajectories. Criminology and Public Policy, 9, 279-305.
Cohen, M. A., Rust, R. T., & Steen, S. (2006). Prevention, crime control or cash?
Public preferences toward criminal justice spending priorities. Justice Quarterly,
23, 317-335.
Cullen, F. T., Vose, B. A., Jonson, C. N. L., & Unnever, J. D. (2007). Public support for
early intervention: Is child saving a habit of the heart? Victims and Offenders,
2, 109-124.
Currie, J. (2001). Early childhood education programs. Journal of Economic Perspectives,
15, 213-238.
Davey, M. (2010, September 19). Touching off debate, Missouri tells judges cost of
sentences. The New York Times, pp. A1, A4.
Donohue, J. J. (2009). Assessing the relative benefits of incarceration: Overall
changes and the benefits on the margin. In S. Raphael & M. A. Stoll (Eds.), Do
prisons make us safer? (pp. 269-341). New York, NY: Russell Sage.
Welsh and Farrington 135S
Donohue, J. J., & Siegelman, P. (1998). Allocating resources among prisons and
social programs in the battle against crime. Journal of Legal Studies, 27, 1-43.
Drake, E. K., Aos, S., & Miller, M. G. (2009). Evidence-based public policy options to
reduce crime and criminal justice costs: Implications in Washington State. Victims
and Offenders, 4, 170-196.
Duncan, G. J., & Magnuson, K. (2004). Individual and parent-based intervention
strategies for promoting human capital and positive behavior. In P. L. Chase-
Lansdale, K. Kiernan, & R. J. Friedman (Eds.), Human development across lives
and generations (pp. 93-135). Cambridge, UK: Cambridge University Press.
Eckenrode, J., Campa, M., Luckey, D. W., Henderson, C. R., Cole, R., Kitzman,
. . . Olds L., David. (2010). Long-term effects of prenatal and infancy nurse home
visitation on the life course of youths: 19-year follow-up a randomized trial.
Archives of Pediatrics and Adolescent Medicine, 164, 9-15.
Ellwood, J. W., & Guetzkow, J. (2009). Footing the bill: Causes and budgetary conse-
quences of state spending on corrections. In S. Raphael & M. A. Stoll (Eds.), Do
prisons make us safer? (pp. 207-238). New York, NY: Russell Sage.
Farrington, D. P., & Welsh, B. C. (2007). Saving children from a life of crime: Early
risk factors and effective interventions. New York, NY: Oxford University Press.
Garces, E., Thomas, D., & Currie, J. (2002). Longer-term effects of Head Start. American
Economic Review, 92, 999-1012.
Greenwood, P. W. (2006). Changing lives. Chicago: University of Chicago Press.
Greenwood, P. W., Model, K. E., Rydell, C. P., & Chiesa, J. (1998). Diverting children
from a life of crime (2nd ed.). Santa Monica, CA: RAND.
Hawkins, J. D., Kosterman, R., Catalano, R. F., Hill, K. G., & Abbott, R. D. (2008).
Effects of social development intervention in childhood 15 years later. Archives of
Pediatrics and Adolescent Medicine, 162, 1133-1141.
Heckman, J. J., Moon, S. H., Pinto, R., Savelyev, P. A., & Yavitz, A. (2010). The rate
of return to the HighScope Perry Preschool program. Journal of Public Economics,
94, 114-128.
Karoly, L. A., Greenwood, P. W., Everingham, S. S., Houb, J., Kilburn, M. R., Rydell,
C. P., . . . Chiesa, James. (1998). Investing in our children. Santa Monica, CA:
RAND.
Knapp, M. (1997). Economic evaluations and interventions for children and adoles-
cents with mental health problems. Journal of Child Psychology and Psychiatry,
38, 3-25.
Lally, J. R., Mangione, P. L., & Honig, A. S. (1988). The Syracuse University Family
Development Research program: Long-range impact of an early intervention with
low-income children and their families. In D. R. Powell (Ed.), Parent education as
early childhood intervention (pp. 79-104). Norwood, NJ: Ablex.
136S The Prison Journal Supplement to 91(3)
Layard, R., & Glaister, S. (Eds.). (1994). Cost-benefit analysis (2nd ed.). Cambridge,
UK: Cambridge University Press.
Lipsey, M. W. (2009). The primary factors that characterize effective interventions with
juvenile offenders: A meta-analytic overview. Victims and Offenders, 4, 124-147.
Lipsey, M. W., & Cullen, F. T. (2007). The effectiveness of correctional rehabilitation:
A review of systematic reviews. Annual Review of Law and Social Science, 3,
297-320.
Loeber, R., & Farrington, D. P. (1998). Never too early, never too late: Risk factors
and successful interventions for serious and violent juvenile offenders. Studies on
Crime and Crime Prevention, 7, 7-30.
Loeber, R., & Farrington, D. P. (Eds.). (2011). Transitions from juvenile delinquency
to adult crime. New York, NY: Oxford University Press.
Ludwig, J., & Phillips, D. A. (2008). Long-term effects of Head Start on low-income
children. Annals of the New York Academy of Sciences, 1136, 257-268.
Manning, M., Homel, R., & Smith, C. (2010). A meta-analysis of the effects of early
developmental prevention programs in at-risk populations on non-health outcomes
in adolescence. Children and Youth Services Review, 32, 506-519.
Marsh, K., & Fox, C. (2008). The benefit and cost of prison in the UK. The results of a
model of lifetime re-offending. Journal of Experimental Criminology, 4, 403-423.
Marsh, K., Fox, C., & Hedderman, C. (2009). Do you get what you pay for? Assessing
the use of prison from an economic perspective. Howard Journal of Criminal Justice,
48, 144-157.
McDougall, C., Cohen, M. A., Perry, A., & Swaray, R. (2006). Costs and benefits of
sentencing. In B. C. Welsh & D. P. Farrington (Eds.), Preventing crime (pp. 117-127).
New York, NY: Springer.
Mears, D. P. (2010). American criminal justice policy. Cambridge, UK: Cambridge
University Press.
Moore, S. (2009, March 3). Prison spending outpaces all but Medicaid. The New York
Times. Retrieved from http://www.nytimes.com
Nagin, D. S. (2001). Measuring economic benefits of developmental prevention pro-
grams. In B. C. Welsh, D. P. Farrington, & L. W. Sherman (Eds.), Costs and benefits
of preventing crime (pp. 251-268). Boulder, CO: Westview.
Nagin, D. S., Piquero, A. R., Scott, E. S., & Steinberg, L. (2006). Public preferences for
rehabilitation versus incarceration of juvenile offenders: Evidence from a contin-
gent valuation survey. Criminology and Public Policy, 5, 627-652.
National Forum on Early Childhood Policy and Programs. (2010). Understanding the
Head Start impact study. Available from http://www.developingchild.harvard.edu/
Piquero, A. R., Farrington, D. P., Welsh, B. C., Tremblay, R. E., & Jennings, W. G.
(2009). Effects of early family/parent training programs on antisocial behavior
and delinquency. Journal of Experimental Criminology, 5, 83-120.
Welsh and Farrington 137S
Reynolds, A. J., Temple, J. A., & Ou, S.-R. (2003). School-based early intervention and
child well-being in the Chicago Longitudinal Study. Child Welfare, 82, 633-656.
Reynolds, A. J., Temple, J. A., Ou, S., Robertson, D. L., Mersky, J. P., Topitzes, J. W.,
. . . Niles D., Michael. (2007). Effects of a school-based, early childhood inter-
vention on adult health and well-being. Archives of Pediatrics and Adolescent
Medicine, 161,730-739.
Schmitt, J., Warner, K., & Gupta, S. (2010). The high budgetary cost of incarceration.
Washington, DC: Center for Economic and Policy Research.
Schweinhart, L. J., Montie, J., Zongping, X., Barnett, W. S., Belfield, C. R., & Nores, M.
(2005). Lifetime effects. Ypsilanti, MI: High/Scope Press.
Weimer, D. L., & Friedman, L. S. (1979). Efficiency considerations in criminal reha-
bilitation research: Costs and consequences. In L. Sechrest, S. O. White, &
E. D. Brown (Eds.), The rehabilitation of criminal offenders (pp. 251-272).
Washington, DC: National Academy of Sciences.
Weinrott, M. R., Jones, R. R., & Howard, J. R. (1982). Cost-effectiveness of teach-
ing family programs for delinquents: Results of a national evaluation. Evaluation
Review, 6, 173-201.
Welsh, B. C., & Farrington, D. P. (2000). Monetary costs and benefits of crime pre-
vention programs. In M. Tonry (Ed.), Crime and justice (Vol. 27, pp. 305-361).
Chicago: University of Chicago Press.
Welsh, B. C., Loeber, R., Stevens, B. R., Stouthamer-Loeber, M., Cohen, M. A., &
Farrington, D. P. (2008). Costs of juvenile crime in urban areas: A longitudinal
perspective. Youth Violence and Juvenile Justice, 6, 3-27.
Wexler, H. K. (2010, January 20). An opportunity to reform the criminal justice system.
Los Angeles Times. Retrieved from http://www.latimes.com.
Bios
Brandon C. Welsh is associate professor of criminology at Northeastern University
and senior research fellow at the Netherlands Institute for the Study of Crime and
Law Enforcement. He has written eight books, including The Oxford Handbook of
Crime Prevention.
David P. Farrington is professor of psychological criminology, Cambridge University.
He is cochair of the Campbell Collaboration Crime & Justice Group as well as director
of the Cambridge Study in Delinquent Development. He has published 50 books,
monographs, and government publications and more than 400 articles on criminologi-
cal and psychological topics. He received the Sellin-Glueck and Sutherland Awards
of the American Society of Criminology.