Skip navigation
The Habeas Citebook Ineffective Counsel - Header

Joyce Foundation Reentry Transitional Jobs Report 2009

Download original document:
Brief thumbnail
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
The Joyce Foundation’s

Transitional Jobs
Reentry Demonstration
testing strategies to help former prisoners
f ind and keep jobs and stay out of prison



contact information
Whitney Smith
The Joyce Foundation
70 West Madison, Suite 2750
Chicago, IL 60602
Dan Bloom
16 East 34th Street
New York, NY 10016
Written by: Dan Bloom, MDRC
Design: Jasculca/Terman and Associates, Inc.

why is successful
prisoner reentry a
national imperative?
The number of people incarcerated in the
U.S. has more than quadrupled in the last
three decades. Today, more than 2 million
people are incarcerated in federal and state
prisons and local jails, and almost 700,000
people are released from state prisons each
year. Corrections costs exceed $65 billion
per year, with most of this total borne by
state and local governments.
Men and women released from prison
often face daunting obstacles as they move
back to their communities. They frequently
have difficulties finding jobs and housing,
and experience problems reconnecting
with family and other social supports. In
addition, former prisoners are concentrated
in a relatively small number of distressed
urban neighborhoods that lack resources
to assist in the reentry process. Not surprisingly, many end up returning to prison, a
disastrous result for them, their families and
communities, taxpayers, and public safety.

The most recent
national statistics show
that two-thirds of those
released from prison are
rearrested, and half are
reincarcerated within
three years of release.
In many cases, people
return to prison not
because they commit
new crimes, but rather
because they violate
the rules of parole

Prisoner reentry has attracted increasing
attention in recent years, as states seek
ways to reduce recidivism and control
surging corrections costs. While most
experts believe that stable employment
is critical to a successful transition from
prison to the community, there is little hard
evidence about which program practices
are effective at promoting successful
transitions or reducing recidivism.

table 1: prisoners in state or federal prison per 100,000 u.s. residents, 1925 to 2004

prisoners per 100,000











Source: Steven Raphael and Michael Stoll (eds.) “Do Prisons Make Us Safer? The Benefits and Costs of the Prison Boom,”
Russell Sage Foundation, New York, 2008

table 2: surveys of government f inances, 1986 – 2001:
expenditures for total state corrections in 2001 constant dollars

total state corrections

total (in 1000’s)1,2

cost per resident 3

















































Source: Bureau of Justice Statistics Special Report, June 2004; “State Prison Expenditures, 2001”; 	
U.S. Department of Justice

Note: Correctional expenditures may be underreported.
Interviews with State budget officials by the U.S. Census
Bureau for this report produced a revised estimate of
State prison costs of $29.5 billion for FY 2001, 1.1% higher
than the 2001 Survey of Government Finances.

US Census Bureau. Censuses of Governmental Finances, 	
	 1986–1996, Tables 11 and 12; and unpublished data 	
	 compatible with this series for 1997 through 2001.
2 Economic Report of the President, February 2003. 	
	 Bureau of Economic Analysis, Chain-type price indexes 	
	 for gross domestic product, 1959–2002, Table B-7.
3 US Census Bureau, Estimates of the Population of
	 the United States to July 1, 1990, Current Population 	
	 Estimates and Projections, Series P-25, No. 1064.
	 US Census Bureau, US Population Estimates by Age,
	 Sex, Race and Hispanic Origin: 1990-1995, PPL-41. 	
	 Unpublished data 1996–2001, compatible with
	 Resident Population Estimates for Age, Sex, Race and 	
	 Hispanic Origin.

The number of
people incarcerated
in the U.S. has more
than quadrupled
in the last three

what is the transitional
jobs reentry demonstration
and why is it signif icant?
A number of states have launched multi-

The TJRD project is one of the largest and

faceted prisoner reentry initiatives – often

most rigorous evaluations of employment

with a strong emphasis on helping people

programs for former prisoners since the

find jobs after they leave prison – and the

1970s. The results, available in mid-2010,

federal government has provided special

should provide solid evidence about the

funding to support these efforts, most

effectiveness of transitional jobs, which

recently through the Second Chance Act

will inform both public policy and program

of 2008. Unfortunately, however, there is

practice at the federal, state, and

very little rigorous evidence about which

local levels.

strategies are effective at helping former
prisoners find and keep jobs.

The TJRD project was developed by the
Chicago-based Joyce Foundation, whose

The Transitional Jobs Reentry

mission includes reducing poverty and

Demonstration (TJRD) seeks to help

violence in the Great Lakes region. The

fill this gap in our knowledge by testing

project is also supported by the JEHT

innovative employment programs for former

Foundation1 and the U.S. Department of

prisoners in Chicago, Detroit, Milwaukee,

Labor. The funders are supporting both

and St. Paul using a rigorous, random-

the employment programs and a careful

assignment research design. In each city,

evaluation being conducted by MDRC, along

one employment program is built around

with the Urban Institute and the University

transitional jobs (TJ)—temporary, subsi-

of Michigan’s Gerald R. Ford School of

dized jobs that provide participants with

Public Policy. The National Transitional Jobs

a source of legitimate income, support

Network is providing technical assistance

services, and work experience as they

to the project.

return to the community.
The transitional jobs programs in the study
are being evaluated against a second set
of simpler, less expensive programs called
“job search” (JS) assistance programs that
help participants look for work but do not
provide subsidized jobs.
Ultimately, the study’s goal is to determine
whether transitional jobs programs are
an effective strategy for increasing employment and reducing recidivism among men
recently released from prison.
1 The JEHT Foundation ceased operations in January 2009.

The TJRD project is
one of the largest
and most rigorous
evaluations of
employment programs
for former prisoners
since the 1970s.

why provide former
prisoners with
transitional jobs?
Stable employment appears to be critical

Transitional jobs are also being evaluated

to a successful transition into the community,

in other major U.S. cities by the U.S. Depart-

but former prisoners often have charac-

ment of Health and Human Services and

teristics that place them at the back of the

MDRC. Early results are now available from

employment queue—for example, low levels

a random assignment evaluation of the New

of education and limited work experience.

York City-based Center for Employment

African-American men are heavily overrep-

Opportunities (CEO), one of the largest and

resented in the prison population, and they

most experienced transitional jobs programs

may also face employment discrimination

for former prisoners. During the first two

upon release. Finally, state laws bar many

years of the study’s follow-up period, CEO

former prisoners from obtaining licenses

significantly decreased crime convictions,

to work in specific occupations, and studies

reincarceration, and other measures of

have found that many employers are quite

recidivism — a result rarely found in rigorous

reluctant to hire people with criminal

evaluations. CEO substantially boosted

records. Several studies have tracked

employment, though the increase faded

employment rates for former prisoners

over time, after participants left the

during the year following release, typically

transitional jobs.

finding that fewer than half are employed
at any point.

Transitional Work Corporation (TWC),

Transitional jobs are seen as a promising

another large-scale transitional jobs

employment model, both for former pris-

program that mostly serves long-term

oners and for other hard-to-employ groups.

welfare recipients. TWC significantly

Transitional jobs programs rapidly place

reduced welfare receipt and welfare

participants into temporary, subsidized jobs,

payments during an 18-month follow-up.

usually in nonprofit or government agencies,

Like CEO, it produced a very large,

provide intensive support, and then help

but relatively short-lived increase in

participants find permanent jobs. When

employment, driven mostly by the

targeted to recently released former

transitional jobs.

prisoners, transitional jobs provide a source
of legitimate income during the critical
period just after release, and also provide
program staff with an opportunity to
identify and address workplace problems
before participants move to the regular
labor market.2
2 For more information on the transitional jobs model,

Another study is testing Philadelphia’s

	 see the National Transitional Jobs Network’s website:

Several studies have
tracked employment
rates for former
prisoners during the
year following release,
typically finding
that fewer than half
are employed at
any point.

how is the tjrd
project designed?
The TJRD project was designed from the

what difference a program makes. Many

start as a rigorous evaluation to discover

evaluations track program participants and

the difference transitional jobs can make

compare their outcomes (for example, their

in the trajectories of former prisoners. In

employment rates) with those of people

each of the four sites, the research team is

who did not participate in the program.

comparing a transitional jobs program with

But if people are not assigned to the

a basic job search assistance program.

program or the comparison group through

Former prisoners who agreed to be in

a random process, one can never be sure

the study were assigned at random to one

the two groups were similar from the start.

program or the other. The project was in-

For example, it is quite possible people who

tended to serve about 400 men in each site

choose to enroll and participate in programs

– 200 in the transitional jobs program and

have different levels of motivation or support

200 in the job search assistance program.

than those who do not, and that these dif-

The random assignment process created
two groups of people – called the TJ and JS

or more than the programs themselves.

groups – that were similar at the time they

In addition to measuring how the transitional

entered the study. If differences emerge

jobs programs affect employment and

between the two groups over time – for

recidivism, the TJRD evaluation will include

example, if one group is more likely to work

three analyses. First, it will analyze how the

or less likely to return to prison – one can

programs operate and assess their costs.

be fairly certain that this is because the

Second, it will include a series of in-depth

two groups received different kinds of

interviews with about 25 study participants

employment services, not because their

to gain a more detailed understanding

characteristics differed from the start.

of their experiences after leaving prison.

Thus, by tracking the two groups over time,

And, third, it will provide an opportunity to

the TJRD evaluation will be able to assess

learn about the operation and impacts of

whether the transitional jobs programs led

transitional jobs and job search assistance

to different employment and recidivism

programs in a range of environments.

outcomes than the job search assistance

There are important differences across

programs, and whether one strategy or

the four cities, for example, in labor market

the other was more effective for particular

conditions, population characteristics, and

subgroups of former prisoners.

criminal justice practices.

A random assignment design can provide
unusually reliable information about

ferences will affect their outcomes as much

Frank’s Story
“Frank” was born to a young mother and into a household of substance abusers and
distributors. He began stealing goods and selling marijuana when he started high
school. By the time he was 17, he was dealing cocaine.
He was first incarcerated in his early twenties, and then spent much of his adulthood
cycling between prison and streets. He was released from his last term at age 44 in
the winter of 2008.
Upon his release Frank sought out temporary employment agencies to try to begin
building a work history. He expressed concerns about adjusting successfully to the
world of work.

	“You get these ideas that, well, ain’t nobody going to give me a
chance because of my criminal background and my criminal
record. It upsets you and it puts you in a bad place in your mind,
and you get to thinking, maybe I should do this, or maybe I could
pick up a bag and start working at it again.”

“If you try to do it by yourself with a background like mine, it’s
depressing. It’s not good, and you’ve got to take a lot of no’s. But,
if you can get networking with a group of people, whether it be
churches, organizations that offer re-entry programs, you’ve got a
base of people that’s trying to work at the same goal, trying to
help you. So, that would be a better shot.”

He balanced his comments about how important this social connection was with
discussion of the staff in the TJRD-sponsored program. As he stated:


“I was already kind of teetering. My thoughts were teetering. I
didn’t actually put any physical acts in, but I was starting to have
bad ideas or bad thoughts. So, without [the TJRD program] and
the direction that they’re pointing me in, I don’t think it would
have been good.”

which programs
are participating in
the project?
In mid-2006, the Joyce Foundation

However, there are also important

conducted a competition and ultimately

differences in the transitional jobs models.

selected four sites to participate in the

In Detroit and St. Paul, TJ participants are

project. Each site received about $600,000

employed directly by the Goodwill agency

over three years, and the grantees were also

running the program, and they work in

expected to raise funds from state or local

existing Goodwill enterprises. In Detroit,

agencies to support their programs. The

most work in a light manufacturing plant,

Illinois, Michigan, Minnesota, and Wisconsin

and in St. Paul most work in jobs related

Departments of Corrections are all active

to collecting, processing, and selling

partners in the project and are providing

merchandise in the agency’s retail stores.

funding to support the employment

In Chicago, most of the Safer Foundation


TJ participants work in garbage recycling


Table 3 shows the organizations that are
operating the transitional jobs and job
search assistance programs in each city.
In three of the cities, separate organizations
are serving the TJ and JS groups, while in
Chicago, the same organization serves
both groups.
There are some basic similarities across
the transitional jobs programs. All provide
participants with temporary, minimum-wage
jobs that offer 30 to 40 hours of paid work
each week; all aim to identify and address
behavior or performance issues that emerge
at the work site; all provide a range of ancillary
services and supports to participants; and
all help participants look for unsubsidized
jobs to follow the transitional jobs, often
with the help of job developers who reach
out to employers to identify job openings

plants operated by Allied Waste Industries
under contract to the City of Chicago;4 they
are directly employed by Pivotal Staffing
Services, a staffing company established
by Safer. In all three of these sites, the
transitional jobs are in enterprises that earn
revenue for the sponsoring agency, partly
offsetting the cost of wages for TJ workers.
The Milwaukee program uses a “scattered
site” model: the New Hope Project is the
employer of record and pays all wages,
but TJ participants are placed in various
nonprofit organizations and businesses
in the community. The worksites are not
asked to pay for the TJ workers, but they are
expected to provide supervision and to stay
in close contact with the New Hope staff,
who are responsible for identifying and
addressing workplace problems.

for participants.


3 Initially a fifth site was selected but research there was 	

4 The Allied Waste Industries contract ended in 2008

	 discontinued in 2007.

	 and some of the TJRD participants moved to another
	 transitional job as Pivotal employees.

In addition to these differences in the

The job search assistance programs also

transitional jobs models, two of the four

differ from each other in some key respects,

sites – Milwaukee and St. Paul – offer

but, at a minimum, all of them help partici-

relatively generous bonus payments to

pants prepare a resume, learn how to fill

participants who get and hold unsubsidized

out job applications and interview for jobs

jobs after working in a transitional job.

(including how to answer questions about

These payments are designed to supple-

their convictions), and identify job leads.

ment the earnings of participants who
obtain relatively low-paying jobs and to
encourage participants to keep working.

table 3: organizations operating transitional jobs and job search
assistance programs in the transitional jobs reentry demonstration.


transitional jobs program

job search assistance program

safer foundation

safer foundation

(through Pivotal Staffing Services)


goodwill industries of
greater detroit

detroit hispanic development corporation


new hope project

project return

st. paul

goodwill/easter seals minnesota

amherst h. wilder foundation


who are the tjrd
The TJRD project targets men age 18 or

As expected, almost all of the study partici-

older who were released from state prison

pants were under parole supervision when

within 90 days prior to enrollment in the

they enrolled in the study. They had served

study. It is widely believed that the first

an average of six years in prison over

weeks after people are released from prison

their lifetimes.

are a critical period in determining whether
their transition will be successful. Men with
all types of criminal histories were accepted
into the project, with no project-wide
restrictions based on the number or type
of previous offenses (there were some
limitations in individual sites).

are generally similar from site to site,
but there are some key differences. For
example, the St. Paul site is serving a larger
proportion of white men, and a much larger
proportion of the study participants there
were living in halfway houses when they

The sites recruited men into the study from

entered the study. In Chicago, about 40

January 2007 through September 2008.

percent of the study participants had no

Slightly more than 1,800 men entered the

high school diploma or GED, compared with

study in all, with the site totals ranging

20 to 25 percent in the other sites. Michigan

from about 375 to 500. Table 4 provides

study participants had served more than

a snapshot of the study participants across

four years in prison, on average, during their

all four sites at the time they entered

most recent stay, compared with about two

the project.

years in the other sites.

As the table shows, the study participants
were 35 years old on average when they
enrolled, and a large majority are African
American. About half are fathers, though
few lived with their children (a substantial
proportion of the fathers owed $5,000 or
more in back child support). Most reported
that they had worked at some point, but
only half had ever held a steady job. Only
about one in four participants had a high
school diploma, but nearly half had a
General Education Development (GED)
certificate; it seems likely that some of the
men earned a GED while incarcerated.


The characteristics of the study participants

table 4: characteristics of tjrd study participants at the time of enrollment

average age


race/ethnicity (%)


black/african american




has children (%)


has high school diploma or ged (%)


living arrangements (%)
owns/rents house/apartment


lives with friends/relatives


transitional housing




ever worked 6 consecutive months for one employer (%)


on probation or parole (%)


average total time spent in prison (months)


average time spent in prison in most recent spell (months)



what are the early
The research team visited the transitional
jobs and job search programs several times
to interview staff and participants, visit


transitional jobs and job search assistance

Despite the instability in
the participants’ lives and
living situations, the programs
were able to place a very high
percentage of the men in the
TJ group – about 85 percent
– into transitional jobs.

programs as designed.

In most cases, the programs sought

transitional jobs worksites, and observe
program activities. Although the grantees
had varying amounts of experience
working with former prisoners and faced
some operational challenges, they were
able, for the most part, to operate the

Key early observations and lessons include
the following:

to place participants in transitional
jobs very quickly – usually within a
week or less after enrollment – in
order to rapidly engage a highly

The programs worked closely
with corrections agencies
to recruit participants.
The programs recruited men by
holding information sessions in
prisons for men about to be released,
building linkages with parole officers
who could refer their clients to TJRD,
and by posting flyers and posters
in parole offices and other locations
in the community. It was sometimes
challenging to find men who had
been released from prison very
recently – many former prisoners
do not seek assistance immediately
after release – but, ultimately, the
programs were able to meet the
study’s enrollment targets.


mobile group of clients. Because
the number of enrollees varied from
week to week, this required having
a flexible pool of transitional jobs.
Also, in most cases, the programs did
not seek to match participants with
particular transitional jobs based on
their skills or interests. On average,
participants worked in transitional
jobs for about four months.



Most of the transitional jobs
are designed to teach general
employability skills, not to
train participants in specific

It has been challenging
for programs to place
participants in second
(post-TJ) jobs, particularly
with the weakening economy.

The transitional jobs model gives

As noted earlier, many former

program staff an opportunity to

prisoners face a range of obstacles

observe participants in a work

to finding jobs, including both

environment in order to identify and

personal factors, such as lack of work

address workplace problems – for

experience, and systemic issues, such

example, lateness, difficulty taking

as discrimination by employers. Thus,

direction or criticism, or inappro-

it is not surprising that many of the

priate interactions with co-workers.

transitional jobs and job search

Normally these issues might cause

assistance programs have struggled

an employee to be fired, but in a

to place participants in permanent

TJ worksite they are used to teach

jobs, particularly jobs that pay

employability skills. All of the

substantially above the minimum

programs provide this type of job

wage. This challenge is particularly

coaching, though in different ways.

daunting in a weak labor market.

However, most of the project’s

The project’s random assignment

transitional jobs are not designed

research design ensures that the

to provide training in a particular

TJ and JS groups are experiencing

occupation. Most of the work is quite

the same labor market conditions.

basic and requires minimal skills.

However, extremely high unemploy-

One site (St. Paul) offers some

ment rates could potentially affect

opportunities for paid training in

the study results by dramatically

construction, automotive skills, and

reducing the availability of jobs for

other occupations (other sites may

men in both groups.

refer participants to training provided


when will the results
be available and how
will they be used?
The research team is tracking the TJ and
JS groups using data from state agencies to
measure both employment and recidivism
during a period of at least one year. The
employment data will measure earnings
in jobs covered by state unemployment
insurance programs, and the criminal justice
data will measure arrests, convictions,
and admissions to state prisons. A report
describing the programs’ effects on
employment and recidivism, their implementation and costs, and the key findings
from the ethnographic interviews, will be
completed and released in summer 2010.
The TJRD project will provide the strongest
and most reliable kind of evidence to inform
the design of policies and programs for
former prisoners. For example, the impact
results and cost estimates may shape future
federal and state funding for reentry
services. At the local level, the information
on program implementation and impacts
will be a valuable resource for those who
design and operate reentry programs. The
Joyce Foundation and the research team
will work together with other key partners
to disseminate and explain the results to
policymakers and program operators in the
region and nationwide.


The TJRD project will
provide the strongest
and most reliable
kind of evidence to
inform the design of
policies and programs
for former prisoners.


the joyce foundation
The Joyce Foundation supports efforts to protect the Great Lakes, to
reduce poverty and violence in the region, and to ensure its residents
have access to good schools, decent jobs, a strong democracy, and a
diverse and thriving culture.

research partners
MDRC is a nonprofit, nonpartisan education and social policy research
organization dedicated to learning what works to improve programs and
policies that affect the poor.

the urban institute
Established in 1968, The Urban Institute is a nonprofit, nonpartisan
policy research organization that examines social, economic, and
governance issues.

university of michigan’s gerald r. ford
school of public policy
The University of Michigan’s Gerald R. Ford School of Public Policy is
one of the nation’s top-ranked policy schools, offering undergraduate,
master’s, and doctoral degrees in public policy.




Stop Prison Profiteering Campaign Ad 2
Advertise Here 3rd Ad
The Habeas Citebook Ineffective Counsel Side