To the average citizen, the reasons for obtaining gainful employment and obeying the law seem obvious: the freedom to pursue,
and the ability to afford, the good things in lifesuch as a home, a family, and a comfortable standard of living. The
high rates of recidivism and unemployment among ex-offenders suggest that the reasons to make an honest livingand to
take the necessary steps toward doing so—are anything but obvious. Far more than a lack of education or skills,
discrimination, or other external obstacles, it is ex-offenders' impulsiveness and unfamiliarity with the world of
work and its trade-offs between sacrifice and reward that explain their poor outcomes after release from incarceration and,
for that matter, their lapses preceding it.
That is the theory behind a residential prisoner-release
program in Montgomery County, Maryland. Realizing
that neither the powerful incentives of freedom
and financial solvency nor the powerful disincentives
of re-incarceration and impoverishment have
sufficiently reshaped this troubled population's
behavior, the program has resorted to the "small
- later curfews
- access to phone cards
- more frequent visits from family
to induce program participants, some of them serious
offenders, to get and keep jobs in the surrounding
community. At the very least, the salaries they
earn go toward victim restitution, child support,
program fees, and the inmates' own savings accounts.
At best, inmates learn, in doses small enough
for them to absorb and respond to, the mainstream
value of delaying gratification and its various
offshoots: punctuality, reliability, and the
effectiveness of effort. Almost 90 percent
of program participants find employment within
three weeks of enrollment, and 54 percent still
have the same employer two months after they
have left the program.
In place of training or educational programs or counseling to produce passing scores on tests measuring inmates’ mental fitness to
rejoin society, Montgomery County's Pre-Release Center (PRC) makes inmates' actual behavior the standard by which their progress
is judged. They soon discover that their actions, constructive and otherwise, have immediate, direct, and predictable consequences.
Staying employed brings them greater measures of freedom within the residential program, to which they must return at the end of each
workday. Gradually they are able to make the mental transition from the completely controlled environment of jail or prison to the
initially shocking andenduringly challenging freedoms of society at large. In jail or prison, they are given no responsibilities; in
society, they are used to escaping them. For many of them, the PRC is their first introduction to the world of individual
accountability and the privileges that accrue from it.
Many correctional systems are not as well funded, well managed, or well situated as Montgomery County's, and
would thus be unable to replicate all of its features. Close monitoring of participants requires high staffing
levels, which are expensive. In addition, the PRC is located in a large metropolitan area with below-average
rates of unemployment, and it is in close proximity to a subway system that provides access to jobs throughout
the Washington, D.C., metropolitan area.
Expensive as the PRC is, so is standard confinement in a jail or prison. And as participants begin to adapt
to the program, they require less supervision, freeing resources for their more troubled peers. Given the
social costs of crime and dependency, a program like the PRC makes economic sense.
The study concludes with a discussion of how the principles of this and similar programs might be adopted
by parole agencies, which today focus on getting parolees to comply with the rules governing their release,
not on instilling a work ethic in those they supervise.
The Montgomery County PRC provides an alternative to incarceration and a bridge to employment and
social reintegration. It recognizes the social and psychological deficits common to the incarcerated
population and has constructed an effective and instructive system to compensate for them.
About the Author
ANNE MORRISON PIEHL is director of the Program in Criminal Justice at Rutgers University,
where she is an associate professor of economics, and a research associate at the National Bureau of
Economic Research. She received her A.B. from Harvard University and her Ph.D. from Princeton University,
both in economics.
Funding for this report was provided by the
Bodman Foundation. The author appreciates the
generosity of the staff and residents of the
Montgomery County Pre-Release Center who participated
in interviews. Opinions and errors are the responsibility
of the author, and should not be attributed
to the foundation or Montgomery County.
There is such broad recognition of the importance of prisoner reentry that substantial federal, state,
local, and nongovernmental resources have been mobilized to increase services and reform practices. The federal
government has invested $100 million under the Serious and Violent Offender Reentry Initiative (SVORI); many state
and local corrections agencies now understand their role to include easing the transition from incarceration to
freedom; and public/private partnerships have been formed to fund programs of similar purpose.
In addition to their criminal records, offenders tend to have low levels of education; minimal work experience;
drug habits, physical and mental health problems; and weak support from family, friends, and the wider
community. In small ways, progress has been made: for example, obstacles to obtaining personal
identification such as a driver's license and qualifying for Medicaid have been lowered. But former inmates' chances of overcoming
their many social and psychological disadvantages on the way to becoming self-sufficient, productive citizens remain slim.
Perhaps the best way of turning offenders away from a life of aimlessness, dependence, and crime, and instilling mainstream
capabilities and values, is by preparing them for the world of work. Employment typically provides structure and status as
well as income, the foundation of a constructive and satisfactory life.
Those who believe that offenders can be rehabilitated have always stressed the importance of improving their job skills
and job access. Yet even those who believe that the main purpose of prison is to punish frequently support work
initiatives for those who have just been released or are about to be released. As sentences became longer and
prison and jail populations grew, often the very same people who had clamored for tough-on-crime policies became
dismayed at these policies' impact on public budgets, as well as the burgeoning number of former inmates in their
communities, no better off or any less threatening for having spent months or years behind bars. Although both ends
of the political spectrum saw employment as an answer, general agreement about precisely
how to achieve this aim did not exist.
Legal advocates generally argue that the poor employment outcomes after imprisonment are due to legal restrictions
on hiring and licensing. That there are multiple restrictions on employment has been well documented.
Bushway and Sweeten (2007) cite evidence that ex-felons are excluded from 800 occupations. But even when there are no
legal restrictions, a criminal record frequently eliminates a job applicant from consideration or reduces
Economic research has established that the stigma of a criminal conviction and incarceration are not enough to
explain offenders' poor record of obtaining and retaining employment: a majority had spotty or nonexistent
employment records before they were sent away. Because inmates have poor outcomes both
before and after prison, the employment restrictions noted earlier are probably not the primary driver of low
employment rates, and thus removing or reducing legal impediments to employment is not likely to improve outcomes
substantially. Prison might even help a little: some researchers have found that earnings and labor-force
attachment are highest in the first few quarters following release from prison.
Giving credence to this possibility is the view of many correctional and social-services practitioners,
who are often able to help place such people in jobs and administer programs that require work as a
condition of participation. While they do not find it easy to place ex-offenders, these professionals
generally feel that with the appropriate organizational structure and connections to employers, jobs
can be obtained. Supporting this view is the long history of work-release programs in many corrections
systems as well as post-release programs such as Ready4Work.
Some have argued in favor of employment efforts to aid prisoner reentry on the basis of the success
of welfare-to-work programs, which increased the labor-force attachment of a population
with high unemployment and seemingly little preparation for work life. But there are reasons not to draw a close
parallel. Evaluations of supported work have long found that it has a more constructive effect on women and
older participants than on the young men who are the primary demographic group leaving the nation’s prisons.
Randomized evaluations of work programs for former inmates have generally found that they do not notably increase workforce
participation, even when they do seem to contribute to lower rates of recidivism. For example,
a recent evaluation of the Center for Employment Opportunity (CEO) in New York City found that one year after
enrollment, there was a statistically insignificant 5.5 percentage-point increase in the percent employed and
not incarcerated and a statistically significant 9 percentage-point reduction in those not employed and
A reduction in recidivism is socially beneficial, to be sure. Reduced crime means fewer victims and lower
expenses for the criminal-justice system. And benefit-cost analysis shows that these gains exceed the costs
of the CEO program. However, the lack of legitimate employment is likely to result either
in a return to crime sometime after the period studied or a life of economic dependency on others or the state.
Why don't employment rates increase when employment is the goal of programs enrolling ex-offenders, and why do
rime rates fall even without improvements in employment rates? Answers are not yet available, but several
substantial evaluations of different program models are under way. These include the seventeen-site
demonstration of the Ready4Work program. Further results are forthcoming from SVORI evaluations as well.
Preliminary results from the latter suggest that completing the program has modest positive effects: after
fifteen months, employment rates and wages were 5–15 percent higher.
Perhaps it is unsurprising that evaluation results are not impressive. Not only does the incarcerated population
display low levels of educational attainment and short and intermittent work histories,
but the labor market into which former inmates are released is difficult for those with few skills. Over the past
several decades, wage rates for high school graduates have risen only slightly (5 percent from 1979 to 2004, in
constant dollars) and have fallen (9 percent over the same time period) for high school dropouts.
The labor-force participation rates of men without a high school diploma fell dramatically over this same period.
 There is a substantial research literature on the growth in income inequality in the
United States, both within and across educational levels, over the past several decades. Whatever has caused
the increase in inequality over time, there is little decent paying work available to those with low levels
of education. And that grim picture is only worsening as the global economy contracts.
The majority of prison systems, consumed by the challenges of managing growing prison populations, fail to address
employment seriously. However, several corrections departments have had programs for some time that recognize
the role of employment in preparing inmates for release. One of these is the Pre-Release and Reentry Services
(PRRS) Division of the Montgomery County (Maryland) Department of Correction and Rehabilitation. The next
section of this essay takes a close look at this prerelease work program in order to identify the mechanisms
that the program uses to get offenders working while they are in residence as well as the key constraints under
which such efforts operate.
Montgomery County has long been considered a model program within corrections, and yet it has several features
that distinguish it from the programs discussed above. After describing its key features, I report on findings
obtained from interviews with staff and inmates to provide as rich a picture as possible of the complex realities
of the lives of the offenders and the competing pressures on the agency. The structure of the Montgomery County
program has much in common with those psychologically oriented programs for incarcerated populations that have
proved to be most effective: it requires offenders to perform tasks that are designed to help them establish
new, more constructive, habits. A close inspection of this program reveals methods for increasing labor-force
attachment that are not readily apparent in the rest of the literature. Unfortunately, most correctional
facilities are much farther geographically from active labor markets and thus cannot as easily adopt work
as an important element of a reentry strategy. Therefore, the concluding sections of this essay discuss how
such systems might adopt at least the principles of this and similar programs.
2. MONTGOMERY COUNTY PRERELEASE PROGRAM
Montgomery County Pre-Release and Reentry Services Division provides structured transitional services to
individuals who are within one year of release from local, state, and federal custody. It operates a residential
prerelease center in Rockville, Maryland, as well as a program that allows selected individuals to live at home
under strict monitoring. Altogether, it maintains an average daily enrollment of 160 to 200 offenders. Thus,
it is small in comparison with the inmate populations of jails and prisons but large in comparison with the enrollments of
work-release and prisoner-reentry programs. (It is often cited that approximately 700,000 inmates are released from state
and federal prisons each year. Less visible is the even larger flow of inmates out of
The inmate population is a county one, so it is composed of less serious offenders than the typical state prison population.
Half of them have been convicted of felonies, half of misdemeanors. The least serious offenders—that is, those with terms
of less than three months for traffic violations and such—are not eligible to join the program.
In cases where participants differ appreciably from the population from which they are drawn, it can be misleading to
compare the outcomes of program participants with those of nonparticipants. This is particularly true when either
relatively few offenders ask to participate or relatively few are selected. Most prerelease programs do "cream" and
serve the least risky and needy clients while leaving the more serious offenders to complete their sentences in
institutional confinement. Montgomery County's PRRS program could not be considered highly selective. The details
of inmate selection are described in the next section. As a result of the rules governing participation, it is
representative of the overall incarcerated population except that it omits the most serious violent offenders.
The division's goals are directed at benefiting the offenders as well as the wider society. The program helps
offenders by addressing their housing and treatment needs following release and developing a support system of
family and community institutions that can facilitate the transition back into the offenders' respective communities.
The program contributes to public safety and community well-being by ensuring that offenders are working—nearly 90
percent of them obtain jobs upon release from jail or prison—and by using their earnings to pay program fees and
child support and comply with restitution orders. Those jobs generate almost $400,000 annually in program fees;
by the time the average offender has been released from custody, he has saved $600.
A. Selection of Program Participants
Eligibility for participation in the work program, which includes residence in the Pre-Release Center (PRC), is
generally restricted to sentenced offenders who are within a year of release, with no history of escape attempts
and no serious criminal charges that are pending. In addition, program participants must
have the legal right to work in the United States. About 70 percent of PRRS participants come from the Montgomery
County jail and are serving sentences of less than eighteen months. Another 6 percent have served longer periods
of time with the State Division of Correction, are Montgomery County residents, and are within six months of release
when they enter the PRRS. Members of the remaining segment have been in the custody of the Federal Bureau of Prison
or Federal Probation, are within six months of release, and are returning to the Washington, D.C., metropolitan area.
For Montgomery County inmates who are eligible, participation in the PRC program is voluntary but infrequently
declined. Signing up for the PRC offers many benefits: inmates move to a facility that allows more freedom of
movement within the walls; they may be visited more often by their family; and they can earn a salary.
The pathways to the PRC from the state and federal systems are different. The federal system tries to place offenders in a
community-based setting before release and penalizes individuals who refuse placements in programs such as the PRC.
The state correctional system does not have such a procedure for placement in communities prior to release. Rather,
several case managers knowledgeable about the PRC channel into the program inmates who will be living in Montgomery County.
Since fitness for work is a prerequisite, the program will not admit noncitizens without work permits or anyone else with
serious medical difficulties. These are common strictures that one would expect to elevate success rates over those of
programs that do not cull candidates.
In 2007, 736 screening interviews were conducted, and 591 inmates entered the PRC program. That is, over 80 percent who
were judged eligible participated. Very few eligible individuals refused to participate. (Correctional populations are
remarkably changeable. A new indictment in the county or elsewhere, a reversal of a previous sentence, or some types of
medical or other conditions could remove someone from eligibility even after screening.) While there is some selection
bias, reflecting both the program's standards and some inmates' lack of interest, it appears to be small in comparison
with what exists in other well-regarded prisoner-reentry programs.
Some 90 percent of program participants are male, and most are between the ages of eighteen and forty-five. About 40
percent of those enrolled are serving time for a drug or an alcohol offense; another 20 percent for violations of
probation; and another 20 percent for property offenses. Violent offenders are not excluded from participation:
personal and sex offenses account for 20 percent of PRC participants. Many prerelease programs do exclude such
Only 7 percent of participants are scored as being at "minimum risk" of recidivism by the most common risk-assessment
tool for correctional populations.[23
] About 10 percent have no prior arrests. Some 35 percent
have not completed high school, 14 percent obtained a GED, and 38 percent have a high school diploma. About 12 percent
have some college experience. This distribution is generally comparable with that found in the average correctional
population, but with a somewhat greater number of high school graduates and somewhat fewer
B. Program Elements
The PRC provides many of the usual services of a community corrections agency. Most inmates are confined to the facility,
but there is also a home confinement program that employs electronic surveillance. Staff conduct drug and alcohol testing;
classes devoted to job search, job readiness, anger management, and GED preparation; and group cognitive behavioral
therapy sessions, among other activities. These are relatively standard offerings. The aspect of the PRC that is
not standard and thus deserves a detailed look is the work-release program behind the PRC's motto, "Freedom through Responsibility."
Inmates are expected to find their own work placements. The PRC staff believe that job-search skills are essential
but would not develop if staff arranged placements for inmates. These placements must
occur at legitimate places of employment—that is, those that report earnings to state and federal agencies and
comply with employment laws. Of course, this requirement makes it much harder for job seekers with low skill
The question of whether to provide training, job placement, or job-search support runs through all efforts to
increase labor-force attachment. In fact, much discussion of welfare reform revolved around the trade-offs
between "work first" and improving job skills. Each of these approaches has positive features, and each has
been tried with former inmates. The CEO program mentioned earlier provides jobs in order to develop work
skills. Previous experiments supported former inmates while searching for work to produce better placements.
If it turns out to be effective, the PRC's contrasting approach could be more easily implemented on a broad
Another potentially awkward requirement is that inmates inform potential employers of their crime and their
custody at the PRC. One might wonder why employers who know these things about such applicants would hire
them. But they do. Caseworkers report several reasons, which are reflected in employer interviews collected
by Antell et al. (undated). "Employers unanimously reported that PRC residents are virtually identical to
the employees they would hire from the street" and "PRC residents are prompt, sober, hard-working, and
reliable while staying at the PRC." In essence, the PRC provides extensive
monitoring of the substance use and lifestyles of its residents. In addition, some employers report
that this population shows lower turnover, even if it stays in these jobs for only a matter of months.
Low turnover, reliability, and sobriety are important employment attributes and can be relatively
rare in the low-wage labor market. It is worth noting that companies with local hiring authority
are more likely to hire PRC participants. When hiring decisions are reviewed by corporate parents,
PRC inmates are routinely rejected. Caseworkers assist in the job search by pointing participants
to companies that appear to be good targets, in view of the experiences of earlier applicants.
There is a third important restriction placed on employment: exclusions of particular kinds of jobs.
No bartending or cab-driving is allowed, for example. As a practical matter, retail and other jobs
are frequently off-limits to those with criminal records for theft. This makes it particularly hard
for women to find jobs. Collectively, these restrictions inordinately burden the low-wage labor market.
The PRC expects newly enrolled inmates to secure employment within three weeks. Until an inmate finds
work, he or she is required to search weekdays from 8 AM to 4:30 PM, which means no television or
recreation or visits during those times. Once an inmate is employed, there is an immediate relaxation
of restrictions. Upon landing a job, inmates are allowed out of their rooms/cells for an additional
hour, until 11 PM. Interviews with inmates revealed that the later bedtime also means an end to
teasing by peers, as well as shorter lines at the few phones available for personal calls.
In total, there are six levels of privileges that inmates may earn. The move from level one to level two includes
the later curfew and an expanded number of hours for receiving visitors. Progress to a new level occurs when
hurdles such as retaining employment for a certain period of time are surmounted. The levels provide inmates
with a series of incentives to hold on to their jobs and improve their work performance. As they move from
level to level, it takes longer for them to achieve new rewards, which take the form of expanded visiting
hours, better accommodations, and permission to spend time away from the facility for reasons other than
work or rehabilitative activities such as Alcoholics Anonymous meetings.
The behaviorist element to the reward structure is distinctive. Until recently, the PRC had a set of rewards
similar to the ones it has now, but promotion from one level to another was based on measurement of attitudes
rather than on actual conduct. (It also rewarded those who had been in the program longest, even if their
behavior was only good enough to keep them from being expelled.) That program was comparable with many others
in which the success of particular interventions is measured by an inmate's performance on psychological
tests that aim to assess his risk of failure following release (subsequent recidivism or other poor outcome).
The current system, adopted in 2006, is deliberately designed to establish new, more constructive, forms
of behavior. This structure is meant to support the changes sought by behavioral therapy. Randomized
evaluations have shown that therapeutic programs based on cognitive-behavioral treatment (to help
regulate emotions, motivate change, and apply mental techniques to daily life) lead to reductions in
recidivism of 5–20 percent. The evidence supporting the effectiveness of
behavior-based, rather than attitude-based, approaches has led to the former’s adoption in growing
numbers of prisons and jails.
The earnings of inmates at all levels of privileges are subject to all sorts of rules. Paychecks are
deposited with the PRC. Under terms that the PRC negotiates with agencies that have claims on an
inmate's wages, steady, even if small, payments are made toward child support, victim restitution,
court costs, and fines. The PRC also withdraws payments for program fees and room and board. Mandatory
savings are another deduction, so inmates accumulate savings balances by having a portion of their
paychecks set aside. The hope is that they will use this money to secure an apartment or transportation
when they are released and thus help support a productive civilian life. After all of these deductions,
there is little left for spending money and little opportunity to spend it. However, as will be seen
below, even small sums can provide a meaningful degree of independence.
C. Program Outcomes
The PRC has three overriding concerns: responsible provision of correctional services, payback to the community, and inmate
success following release. Table A reports the program’s finances for 2007 and its attractive economics. Of the over $2
million earned by participants, nearly $350,000 is paid to the county as "program fees." In addition, the county saves
63,000 bed days in the local jails. Valued at a typical per diem of $100 per day, these savings come to $6 million.
The program’s economic value extends beyond the county corrections department. Taxes, restitution, family support, and
payments to state and federal corrections sum to approximately $1.5 million, far more than the inmates keep for themselves.
Inmate outcomes are presented in Table B. The average PRC participant had not worked in fifty-seven months
much longer thanthe period of incarceration, which is about a year). This is the most striking evidence that
this population faces challenges in entering and remaining in legitimate labor markets. The time it takes to
find employment is only a few days more than the program's goal of twenty-one days. Inmates' median wage is
almost $10 per hour, in an area of the country with one of the highest costs of living. About 10 percent of
placements end badly—either with dismissal by the employer or removal by concerned staff. Despite inmates'
poor job skills and limited work histories, 86 percent are employed at the time of release from correctional
custody. At the time of release, the average job tenure is fifty-eight days, or approximately three months
of work. There is little information about how people fare after they leave correctional custody. The only
follow-up information is collected from calls to employers two months after the inmate is released from the
PRC. At that time, 54 percent are still with the same employer. This measure is clearly an underestimate of
the employment rate, as those who changed jobs are not included.
The PRRS is currently conducting a recidivism study, collecting arrest and conviction information from county,
state, and federal sources, but there is no information available about recidivism outcomes for this population
at this time. Although its policy focus is improving the post-release lives of participants, the self
evaluations it has conducted focus on how well the PRC manages inmate conduct (to keep the community, staff,
and other participants safe) and on the resources contributed by participants to government coffers. None of
the outcomes presented here comes from an experimental evaluation, making them difficult to compare with the
evaluations discussed earlier. But they do add depth to the developing picture of inmate employment. This
population, with low educational attainments and employment histories that are spotty at best, nonetheless
finds work relatively quickly and, within the structure of the program, generally maintains it for several
months. Fully half maintain these same jobs for several months after leaving the structured environment
that the program provides.
The next section provides a much richer picture of the program by adding inmates' perceptions of it. These
impressions were gathered by the author in a series of twenty inmate interviews as well as through observation
of the program facility over several days in May 2008. Interviewees represent a cross-section of the population:
some new, some near release; some male, some female; some young, some older; some who were in their first
stint at the PRC, and some who had been through the program several times. Caseworkers selected inmates
who were available to be interviewed during the day or early evening and who either represented one of
the categories listed above or were particularly reflective about their circumstances.
D. Inmate Perceptions
Inmates at the beginning generally chafe at the highly structured nature of the program. (This fact was
apparent among the newly arrived and was revealed by those who had been in the program for some months.)
Life in a correctional facility is generally characterized by plenty of "free" time. Despite the fact
that one's fundamental liberty is curtailed, the opportunity to hang out and talk, play cards, or exercise,
Early on, the expectations of the program's staff are experienced as burdensome. However, those who
find jobs and begin to succeed at them often begin to appreciate the purpose of the requirements, which
is to support their work performance. Some even begin to appreciate the curfew for that reason. At that
point, they become less resentful about having to work so hard. As one interviewee attests: "At the
beginning, I thought it was BS; they were always picking on me. Once I saw results, I changed my mind-set.”
The structure of the program, incorporating immediate, tangible rewards, is its essential feature. The
day that an inmate secures a job, a congratulatory sign is posted in the facility and restrictions on
him are relaxed. Those who progress up the reward structure are pleased with their new independence.
One salient accomplishment of theirs is having the right to purchase a phone card, relieving them of
the need to make collect calls, which shift the cost to family and friends. Inmates are shocked to
have accumulated savings, frequently for the first time in their lives. (They can monitor savings
balances by checking with program staff.)
Interviewees who were working commented on how much effort it required and, in the process, revealed
what a new experience this was for them. One representative comment was: "Never in my life have I
worked this hard." Typical jobs included stocking shelves at night and food service. One inmate I
met had been promoted to manager at his job, a source of great pride.
One feature of inmate reentry that has received little attention is just how removed many offenders
are from a middle-class way of life. In the interviews, several inmates commented that as children,
they had not been exposed to adults with regular employment. One aspect of the program is the ideal
of continuous, legitimate employment, as exemplified by the caseworkers and other staff.
Generally, older inmates do somewhat better than younger inmates, who complain that the expectations
placed on them are unreasonable or unfair, either intrinsically or in the way they are applied. Some
younger inmates are slow to realize that the requirements placed on them are as numerous as they
are because many were not being followed. Several interviewees, however, acknowledged that on
previous stays at the PRC, they did not comply with the program's rules. They have returned to
the program in the hope, shared by staff, that being older will improve their chances of succeeding.
Some inmates, mostly those without jobs, feel extremely frustrated by the restrictions on the
types of jobs that are permitted. The restrictions are onerous to them and are experienced as
punitive rather than as serving some important interest of the program or the county. Those with
jobs, though, seem to understand why they exist or are at least resigned to them. One man told
me that he would like to return to his landscaping job following release but that his employer
was unwilling to hire him if he had to report the position’s existence and make tax payments
on the occupant's behalf. This inmate felt that he could make more money after release than he
was now making, but he was willing to remain in his present job, where he could save at least
some money until his release from the program, because it allowed him to move out of the jail
and into a lower-security facility. This example points to one of the many tensions in
program design. In this inmate's case, it may well be more productive to see him placed in a
job that he is likely to keep than to force him into a less lucrative, more temporary, one.
But the PRRS cannot permit the former if it is "off the books," even though this restriction
seals off much of the low-skilled labor market.
In general, participants who have secured a job and been promoted several levels up the
privileges structure recognize that they have entered a positive feedback loop. Inmates
expressed pride in being able to contribute child support or to the rent on their fiancée's
apartment. The appreciation they receive gives them the confidence to stick with the program
and continue to move forward in their work. But those at the bottom of the ladder often have
unrealistic expectations about how well they are likely to be rewarded in the labor market.
One of the main achievements of the program may be to prepare inmates for the reality of
the situation that they will encounter when they leave custody.
One inmate who was employed and moving his way up the levels reflected on his situation: "I
know I need more time so that everything is established." But he quickly admitted, "I won't
say that if I could leave tomorrow I wouldn't, but...." This struggle between the desire for
immediate gratification and recognition that delaying it is likely to produce better outcomes
in the end is typical of PRRS enrollees.
3. COMPARISON OF PRC WITH PRISONER-REENTRY BEST PRACTICES
The Montgomery County PRC has several features that make its work-release program feasible. It is
located about a half-mile from a Metro subway stop, allowing its residents access to jobs throughout
the Washington, D.C., metropolitan area. Many correctional facilities are located far from business
districts or even public transportation, making a work requirement impossible to implement. A related
feature is that the majority of inmates plan to live in the vicinity after release, and indeed were
selected in part to increase the chances that they would stay in their jobs after release. (At state
prisons, inmates are frequently far from home, making any work placement necessarily temporary.)
Equally important is the sentencing structure governing the terms of confinement. Work release and
the gradual relaxation of restrictions are easiest to implement when corrections officials: (a) have
complete control over the conditions of confinement (so that they can be quickly relaxed or reinstated);
(b) can reasonably anticipate the date of release so that they can know the extent of demand for their
beds in any given period; and (c) operate under loose restrictions on who can participate, so that a
wide cross-section of the correctional population may be considered. A final
quality that distinguishes the PRC from other reentry efforts is that it has sufficient resources and
is well managed. Many correctional facilities lack programs that meet inmates’ needs; when such
offerings do exist, inmates usually face long waiting lists.
While the PRC has several advantages over other efforts to improve prisoner reentry, it also has
much in common with the best practices in the field. A recent report of the Pew Center on the States
recommends thirteen strategies for supervision of offenders in the community by parole and probation
- Define success as recidivism reduction and measure performance.
- Tailor conditions of supervision.
- Focus resources on higher-risk offenders.
- Front-load supervision resources.
- Implement earned discharge.
- Supervise offenders in their communities.
- Engage partners to expand intervention capacity.
- Assess criminal risk and need factors.
- Balance surveillance and treatment in case plans.
- Involve offenders in the supervision process.
- Engage informal social controls.
- Use incentives and rewards.
- Respond to violations with swift and certain sanctions
The PRC utilizes all these strategies, except the reliance on recidivism as an outcome measure. This is a
failure of the program that the agency is working to rectify, but it is far from
establishing a routine process to assess recidivism.
Item 10 requires some explication, as it is unclear what "involve offenders in the supervision process"
really means. Under this item, the Pew report recommends that the goal of supervision be to change
patterns of behavior by requiring inmates to fulfill a series of commitments. Ultimately, they learn
in this way to be accountable for their decisions. The Montgomery model incorporates this idea. Other
corrections systems follow the Pew recommendations by treating education, for example, rather than
employment as the desired form of behavior.
The last item addresses the response to violations. Because inmates at the PRC are serving correctional
sentences, revocation of their privileges is immediate. If an inmate does not report to his job after
signing out of the facility, local police will pursue him as an escapee and transport the fugitive
directly to the jail upon capture. Thus, the sanctions are swift, sure, and salient.
The accounts of how reentry programs generally fall far short of the standard set in the Pew report
are innumerable. I note just a few for the interested reader. Travis (2005) surveys the landscape of
failures at facilitating prisoner reentry and the social costs of this failure; Piehl (2002) explains
how the legal environment can often hinder efforts to prepare inmates for post-release life, which is
frequently characterized by little structure and a lack of people or institutions demanding
accountability; and Piehl et al. (2003) discuss how the complex criminal histories of prisoners
intersect with various programmatic restrictions, leading most reentry programs to serve small
subsets of the population leaving prison.
The PRC model, then, reflects practices that are recommended for correctional authorities. It is
not representative of correctional programs, which generally fall far short of this standard.
Unfortunately, most correctional systems do not have the assets of the Montgomery program, without
which it cannot be replicated. For example, the close monitoring of participants requires high
staffing levels, which are expensive. Moreover, the program is located in a large metropolitan
area with below-average unemployment, and the center is in close proximity to a subway system that
provides access to potential job placements throughout the region. In order to isolate the key features
of the PRC model, the next section considers how the Montgomery program reflects, and differs from,
4. PROVIDING EFFECTIVE INCENTIVES FOR WORK
The work program at the Montgomery County PRC has little in common with many "big ideas," such as
wage subsidies and vocational training, for increasing employment among those with poor prospects in
the labor market. When efforts to subsidize work or job searches have been rigorously evaluated, their
impact on employment and earnings has been found to be minimal or nonexistent.
Perhaps this is because even with subsidies, the kind of legitimate work available to offenders is
still unattractive to them, despite crime’s low returns and the ever-increasing severity of punishment
that criminals face. Subsidies are also expensive, as they have to be paid
to everyone in a program, even those who would not need one to obtain or remain in a legitimate job.
Inmates generally have histories of impulsive actions. Therefore, it makes sense to propose incentive
structures that offer almost immediate payoffs rather than the acquisition of skills with which to build
a career. While no doubt some criminal offenders are purposeful in their criminal activity, most offenders'
histories reveal them to be responsive to short-run incentives. The achievement of pragmatic goals,
therefore, may require the use of incentives that are short-term as well.
These ideas underlie the PRC work program. They also have much in common with ideas of behavioral
economics that are gaining currency. Behavioral economics bases its policy recommendations on how
people actually respond to situations (in contrast to how people probably—or, according to some
premises, ought to—behave). Because its insights derive from how ordinary people behave, they may
need to be modified when applied to criminal offenders, who generally have greater difficulty with
long-term planning and impulse control.
Behavioral economics challenges the orthodox presumption underlying policy design that people
decide on a course of action that promises to provide the best long-term results after trading off
all the relevant alternatives. In contrast, behavioral economists find that how choices are presented
can exert great influence on the actual choices that people make; that inertia and habit are important
factors in determining which choices get made or whether a deliberate, rather than a default, choice
is made at all; and that tight structure and immediate reward may be much more salient than the
promise of a downstream financial return.
These findings are only now beginning to be taken seriously by those involved in policy design.
This intellectual framework has been implemented most famously in the area of retirement savings.
If a company automatically enrolls new employees in savings plans, from which those who do not want
to save money in this fashion must deliberately opt out, participation rates turn out to be much
higher than they are in workplaces where employees who do want to save in this fashion must take
steps to enroll. Under both systems, the employee has free choice; but how the choice is presented
influences the rate of participation. And this effect persists for years. Thaler and Sunstein's book,
Nudge, argues that small changes to context frequently have greater practical impact than
more dramatic policy alternatives such as education about the growth potential of retirement savings
or the availability of employer matching funds.
Elements of the PRC work program can be viewed through the lens of behavioral economics. It is
true that participants operate within a conventional structure of both incentives and disincentives
that promote compliance with program expectations, such as rewards for good behavior, the chance to
accumulate savings, and the threat of being returned to the more restrictive conditions of a jail.
Like ordinary people, only more so, offenders fail to act in their own economic self-interest and
also, more than ordinary people, need to be placed in a structure that is able to compensate for
those tendencies. Thus, the right to enjoy the comfort of a room rather than a cell, for example,
entails a commitment to punctuality, reliability, and effort in the workplace that yields benefits
that surely seem to an offender just as remote as the achievement of a million-dollar 401(k)
account twenty-five years hence might seem to the average white-collar worker. (Indeed, the
inmates themselves are surprised at how quickly their savings accumulate.) Yet the right framework
can coax both groups into advancing their long-term self-interest.
Policies to increase attachment to employment face several additional issues. First, there is
the very legitimate concern about fairness. Should those who have violated the criminal laws receive
subsidies and incentives that are not available to equally poor and low-skilled people who have not
offended? Assuming they should, on the grounds that society at large benefits far more from these
measures than the offenders do collectively, then the challenge is how to make them effective.
Most of those who are incarcerated have multiple large obligations to government, especially to
pay restitution and child support. Correctional programs can use these debts as leverage to induce
inmates to seek work and remain on the job. Attaching a new worker's wages to pay off seemingly
insurmountable debts drastically suppresses his motivation to do so, especially when earnings are
already quite low. But forgiving the debt also seems unfair. Some correctional and social-services
agencies try to make paying the obligations more feasible for low-income parents. Recommendations
include sliding-scale garnishment of wages, suspension of the accumulation of debt while incarcerated
and perhaps the modification or capping of large arrearages so that any further accumulation of debt
needn’t be paid back. The situation is not unlike the one faced by homeowners
under threat of foreclosure—none of the options is terribly satisfying.
Yet the value of the government's leverage is limited. If the government tries in earnest to collect
on all the arrearages in full, it will necessarily set what is, in effect, a very high marginal tax
rate for earnings. (Recall that while living at the PRC, inmates must turn over nearly 90 percent of
their earnings to various governmental agencies.) The high tax rate acts as a disincentive to keep
working, and it undermines the government’s concurrent efforts to attach former inmates to work. The
path taken by the PRC is one of getting repayment started, even if at a very low level. Once a routine
is established, it will continue at least until employment is interrupted.
5. IMPLEMENTING A PROGRAM OF MANDATORY REENTRY ACCOUNTABILITY
The PRC combines the tight supervision of a jail with proximity to labor markets, making its program both
an alternative to incarceration and a bridge to employment and social reintegration. For correctional systems
without such a fortunate combination of circumstances, the logical agency to implement an alternative system
would be parole. Parole officers are generally located in communities near ex-inmates' place of residence, and
they have a great deal of authority over those under supervision. Because offenders are still fulfilling their
sentences, parole agents also have the authority to act swiftly. Recent studies have
shown that employment rates among inmates are highest in the quarters immediately following release.
Whether this is the result of the contemporaneous pressures of parole supervision or of enthusiasm at the time
of release for a different and better way of life, or there is a grace period before unhealthy influences
resume their effects, it is the right time to push those still unemployed
into gainful work.
Unfortunately, not all states have mandatory supervision, and it is possible for offenders to move
abruptly from secure confinement to an unmonitored life in the community. Most
prisoners released from correctional facilities receive some form of post-release supervision; but in many
cases, the agencies have large caseloads and pay minimal attention to compliance with the conditions of
release. To be sure, many parole agencies have been developing “halfway back” programs to expand the set
of punishment options beyond simple revocation of parole, which returns violators to secure confinement.
Parole could be reoriented largely but not exclusively around employment outcomes. The initial stages
of parole supervision would also need to be rigorous enough to make the relaxation of supervision a
meaningful reward. Doing that would be consistent with the Pew’s fourth and fifth recommendations: to
shift resources to the onset of supervision; and to allow inmates to earn their way off it.
Of course, caseloads would need to be modest enough to permit adequate monitoring of behavior and the tailoring
of conditions. One virtue of any PRC-like program is that those who meet expectations quickly graduate to a status
that requires less oversight. Thus resources can be concentrated on those who require them most.
What about those who cannot succeed under such an onerous structure? The fact remains that even those who do
not find a job will have been exposed to a more structured way of life as well as the realities of searching for
work at a legitimate employer. It is not reasonable to expect every participant to turn into a law-abiding,
productive citizen; for some, a lengthening of the time before the next incarceration may be all that is realistic
One potential drawback to a shift to emphasizing real-life accomplishments is that it might weaken the
current vigilance toward drug use and criminal activity (although the literature does not suggest as much).
The other potential drawback is expense. Holding inmates accountable for their behavior requires more, and
more attentive, staffing than simple "warehousing" (whether carried out in secure facilities or by parole
officers with caseloads of 100 or more). A parole program that fully commits itself to improving labor-force
outcomes could conceivably need to double its parole budget. How could it be paid for? The proposed model
has not been subjected to a formal cost-benefit analysis. However, the CEO program, which focuses on offenders'
accomplishments, produces social benefits that exceed their social costs, due primarily to the low rates of
recidivism that its graduates manifest.
The best way to fund such a program is to couple it with sentencing reform. If a demanding supervision program
with enforced work were considered part of the term of incarceration, the money saved by shrinking prison
populations could pay for such an initiative. One example of this is the $6 million in prison-bed days saved
by the PRC itself.
Another piece of evidence comes from recent research in Washington State that shows that allowing inmates to
earn early release from prison by demonstrating good behavior is cost-effective. That is, the recidivism rate
for those released early because of good behavior was no higher than it was for those who were previously
prohibited by law from earning early release. And the costs (due to the reduction in prison-bed days) are much
lower under the reformed sentencing. With the same rate of recidivism and lower costs, the cost-benefit
calculation is clear. The taxpayer savings from earned release were nearly $11,000 per offender, the result
of an average of sixty-three fewer days in prison.
The findings from the Washington evaluation indicate that moving inmates from prison to intensive parole for
two to three months could shift as much as $10,000 to $15,000 per inmate from prison budgets to parole budgets.
Because parole costs are generally several thousand dollars per offender, each inmate under such a program
could generate enough savings to triple or quadruple spending on parole. And since every participant who
spends fewer days in a prison or jail bed would, in effect, be funding the cost of supervising his or her
parole, such a model could operate at any scale.
Even if there were no net programmatic savings, the social savings would more than justify the shift. Of
course, it could turn out that not all of the money realized from shorter prison terms would have to be
reallocated to parole to produce equally positive outcomes. In that case, the social benefits realized
would not have to count toward the economic benefit of such a shift. In either case, supervision would
have to become more intensive, so that a much more tailored system of accountability could be developed.
In sum, it is possible to redesign sentencing to provide sufficient financial resources and legal
authority to parole agencies to meet high employment expectations and conduct close oversight of inmates.
Such a program is a superior alternative to abrupt, unencumbered release into the community or a lightly
supervised period of mandatory reporting. The proposal here has not been put to a rigorous test. But it
is based on ideas and practices that have strong support in the literature. Since its wide implementation
does not face true barriers of cost, it has the potential to significantly reduce the prisoner-reentry
problem in America today.
Antell, Kira, Chris Partidge, and Doug Call (undated). "Recommendations for the Work Release Component of the
Montgomery County Pre-Release and Reentry Service Program at the Pre-Release Center." Memo.
Blank, Rebecca M., Sheldon H. Danziger, and Robert F. Schoeni, eds. (2006). Working and Poor: How Economic and
Policy Changes Are Affecting Low-Wage Workers. New York: Russell Sage Foundation.
Blank, Rebecca M., and Heidi Shierholz (2006). "Exploring Gender Differences in Employment and Wage Trends among
Less-Skilled Workers." In Working and Poor, ed. Rebecca M. Blank, Sheldon H. Danziger, and Robert F. Schoeni,
chapter 1. New York: Russell Sage Foundation.
Bloom, Dan, Cindy Redcross, Janine Zweig, and Gilda Azurdia (2007). "Early Impacts from a Random Assignment
Evaluation of the Center for Employment Opportunities (CEO) Prisoner Reentry Program." MDRC, working paper.
Bushway, Shawn, Michael A. Stoll, and David F. Weiman, eds. (2007). Barriers to Reentry? The Labor Market
for Released Prisoners in Post-Industrial America. New York: Russell Sage Foundation.
Bushway, Shawn, and Gary Sweeten (2007). "Abolish Lifetime Bans for Ex-Felons.” Criminology and Public
Policy 6, no. 4:697–706.
Butcher, Kristin F., and Robert LaLonde (2006). "Female Offenders' Use of Social Welfare Programs Before
and After Jail and Prison: Does Prison Cause Welfare Dependency?." Federal Reserve Bank of Chicago, working
Cammett, Ann (2005). “Making Work Pay: Promoting Employment and Better Child Support Outcomes for Low-Income
and Incarcerated Parents.” New Jersey Institute for Social Justice.
Drake, E. K., and R. Barnoski (2008). "Increasing Earned Release from Prison: Impacts of 2003 Law on Recidivism
and Criminal Justice Costs." Washington State Institute for Public Policy, document 08-11-1201.
Farley, Chelsea, and Wendy S. McClanahan (2007). "Ready4Work: In Brief." Public/Private Ventures 6 (May).
Freeman, Richard B. (2007). America Works: Critical Thoughts on the Exceptional U.S. Labor Market. New York:
Russell Sage Foundation.
Harlow, Caroline Wolf (2003). "Education and Correctional Populations." Bureau of Justice Statistics,
Special Report, NCJ 195670.
Heinrich, Carolyn J., Peter R. Mueser, and Kenneth R. Troske (2007). "The Role of Temporary Help
Employment in Low-Wage Worker Advancement." National Bureau of Economic Research (NBER), working paper 13520.
Holzer, Harry J., Steven Raphael, and Michael A. Stoll (2006). "Perceived Criminality, Background Checks,
and the Racial Hiring Practices of Employers." Journal of Law and Economics 49 (October): 451–80.
Hughes, T. A., D. J. Wilson, and A. J. Beck (2001). "Trends in State Parole, 1990–2000." Bureau of Justice
Statistics, Special Report, NCJ 184735.
LaLonde, Robert, and Daniel Sullivan (2008). "Vocational Training." NBER, "Targeting Investments in Children"
conference, working paper.
Langan, Patrick A. (1991). "America's Soaring Prison Population." Science 251, no. 5001 (March): 1568–73.
Legal Action Center (2004). "After Prison: Roadblocks to Incarceration." New York: Legal Action Center.
Mauer, Marc, and Meda Chesney-Lind, eds. (2003). Invisible Punishment: The Collateral Consequences of
Mass Imprisonment. New York: New Press.
Mead, Lawrence M. (2007). "Toward a Mandatory Work Policy for Men." The Future of Children 17, no. 2:43–72.
Mincy, Ronald B., ed. (2006). Black Males Left Behind. Washington D.C.: Urban Institute Press.
Needels, Karen E. (1996). "Go Directly to Jail and Do Not Collect? A Long-Term Study of Recidivism,
Employment, and Earnings." Journal of Research in Crime and Delinquency 33, no. 4:471–96.
Newman, Katherine S. (1999). No Shame in My Game: The Working Poor in the Inner City. New York: Russell
Pager, Devah (2003). "The Mark of a Criminal Record." American Journal of Sociology 108:937–75.
Petersilia, Joan (2005). "From Cell to Society: Who Is Returning Home?." In Prisoner Reentry and Crime
in America, ed. Jeremy Travis and Christy Visher, pp. 15–49. New York: Cambridge University Press.
——— (2003). When Prisoners Come Home. New York: Oxford University Press.
Pew Center on the States (2008). "Putting Public Safety First: 13 Strategies for Successful Supervision
and Reentry." Public Safety Policy Brief no. 7.
Piehl, Anne M. (2002). "From Cell to Street: A Plan to Supervise Inmates after Release." Massachusetts
Institute for a New Commonwealth.
———, and Stefan F. LoBuglio (2005). "Does Supervision Matter?." In Prisoner Reentry and Crime in America,
ed. Jeremy Travis and Christy Visher, pp. 105–38. New York: Cambridge University Press.
Piehl, Anne M., Stefan F. LoBuglio, and Richard B. Freeman (2003). "Prospects for Prisoner Reentry."
Economic Policy Institute.
Rosenblum, Robert and Debra Whitcomb, An Exemplary Project: Montgomery County Work Release/Pre-Release
Program, Washington DC: U.S. Department of Justice, National Institute of Law Enforcement and Criminal Justice, 1978.
Sabol, William J. (2007). "Local Labor-Market Conditions and Post-Prison Employment Experiences of Offenders
Released from Ohio State Prisons." In Barriers to Reentry? The Labor Market for Released Prisoners in
Post-Industrial America, ed. Shawn Bushway, Michael A. Stoll, and David F. Weiman New York: Russell
Schochet, Peter Z., Sheena McConnell, and John Burghard (2003). "National Job Corps Study: Findings Using
Administrative Earnings Records Data," final report, Mathematica Policy Research.
Solomon, Amy L., Jenny Osborne, Stefan F. LoBuglio, Jeff Mellow, and Debbie Mukamal (2008). Life after Lockup:
Improving Reentry from Jail to the Community. Washington, D.C.: Urban Institute.
Thaler, Richard H., and Cass R. Sunstein (2008). Nudge: Improving Decisions about Health, Wealth, and Happiness.
New Haven: Yale University Press.
Travis, Jeremy (2005). But They All Come Back: Facing the Challenges of Prisoner Reentry. Washington, D.C.:
Urban Institute Press.
Tyler, John H., and Jeffrey R. Kling (2007). "Prison-Based Education and Re-Entry into the Mainstream Labor
Market." In Barriers to Reentry? The Labor Market for Released Prisoners in Post-Industrial America, ed. Shawn
Bushway, Michael Stoll, and David Weiman, pp. 227–56. New York: Russell Sage Foundation.
West, Heather C., and William J. Sabol (2008). "Prisoners in 2007." Bureau of Justice Statistics, Bulletin,