Skip navigation
Disciplinary Self-Help Litigation Manual - Header

Center for Economic and Policy Research - Ex‐offenders and the Labor Market, Schmitt & Warner, 2010

Download original document:
Brief thumbnail
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
Ex‐offenders  
and the Labor Market 
John Schmitt and Kris Warner 
 
November 2010 

Center for Economic and Policy Research 
1611 Connecticut Avenue, NW, Suite 400 
Washington, D.C. 20009 
202‐293‐5380 
www.cepr.net 

CEPR

Ex-offenders and the Labor Market

I

Contents 
Executive Summary ...........................................................................................................................................1
Introduction........................................................................................................................................................2
Estimating the Size of the Ex-offender Population .....................................................................................3
The Effects of Imprisonment and Felony Conviction on Subsequent Employment and Wages .........8
Longitudinal Surveys of Individuals ...........................................................................................................8
Employer Surveys..........................................................................................................................................9
Audit Studies................................................................................................................................................10
Aggregated Geographic Data ....................................................................................................................11
Administrative Data....................................................................................................................................11
Assessment of Employment Effects ........................................................................................................11
Estimating the Impact of the Ex-offender Population on Total Employment and Output................12
Conclusion ........................................................................................................................................................14
Appendix...........................................................................................................................................................15
Releases.........................................................................................................................................................15
Lifetime Probability ....................................................................................................................................18
References .........................................................................................................................................................19

About the Authors
John Schmitt is a Senior Economist and Kris Warner is a Program Assistant at the Center for
Economic and Policy Research, in Washington, D.C.

Acknowledgments
The authors thank Dean Baker for helpful comments. CEPR thanks the Ford Foundation, the
Public Welfare Foundation, and the Arca Foundation for generous financial support.

CEPR

Ex-offenders and the Labor Market

1

Executive Summary 
We use Bureau of Justice Statistics data to estimate that, in 2008, the United States had between 12
and 14 million ex-offenders of working age. Because a prison record or felony conviction greatly
lowers ex-offenders’ prospects in the labor market, we estimate that this large population lowered
the total male employment rate that year by 1.5 to 1.7 percentage points. In GDP terms, these
reductions in employment cost the U.S. economy between $57 and $65 billion in lost output.
Our estimates suggest that in 2008 there were between 5.4 and 6.1 million ex-prisoners (compared
to a prison population of about 1.5 million and a jail population of about 0.8 million in that same
year). Our calculations also suggest that in 2008 there were between 12.3 and 13.9 million ex-felons.
In 2008, about one in 33 working-age adults was an ex-prisoner and about one in 15 working-age
adults was an ex-felon. About one in 17 adult men of working-age was an ex-prisoner and about one
in 8 was an ex-felon.
An extensive body of research has established that a felony conviction or time in prison makes
individuals significantly less employable. It is not simply that individuals who commit crimes are less
likely to work in the first place, but rather, that felony convictions or time in prison act
independently to lower the employment prospects of ex-offenders.
Given our estimates of the number of ex-offenders and the best outside estimates of the associated
reduction in employment suffered by ex-offenders, our calculations suggest that in 2008 the U.S.
economy lost the equivalent of 1.5 to 1.7 million workers, or roughly a 0.8 to 0.9 percentage-point
reduction in the overall employment rate.
Since over 90 percent of ex-offenders are men, the effect on male employment rates was much
higher, with ex-offenders lowering employment rates for men by 1.5 to 1.7 percentage points.
Even at the relatively low productivity rates of ex-offenders (they typically have less education than
the average worker), the resulting loss of output that year was likely somewhere between $57 and
$65 billion.
The rise in the ex-offender population – and the resulting employment and output losses –
overwhelmingly reflects changes in the U.S. criminal justice system, not changes in underlying
criminal activity. Instead, dramatic increases in sentencing, especially for drug-related offenses,
account for the mushrooming of the ex-offender population that we document here.
Substantial scope exists for improvement. Since high levels of incarceration are not the result of high
levels of crime, changes in sentencing today can greatly reduce the size of the ex-offender
population in the future. Moreover, the high cost in terms of lost output to the overall economy also
suggests the benefits of taking action to reduce the substantial employment barriers facing exoffenders.
In the absence of some reform of the criminal justice system, the share of ex-offenders in the
working-age population will rise substantially in coming decades, increasing the employment and
output losses we estimate here.

CEPR

Ex-offenders and the Labor Market

2

Introduction 
Federal, state, and local governments in the United States currently hold about 2.3 million people in
prisons and jails 1 and supervise another 5.1 million people on parole or probation. 2 As recent
research from the Pew Center on the States has emphasized, these figures translate to about one in
100 American adults 3 behind bars and about one in 33 American adults 4 under some form of
correctional control. 5 In this report, we examine an even larger population connected to the criminal
justice system – the growing number of ex-offenders (ex-prisoners and ex-felons) – most of whom
are not currently in prison or jail nor on probation or parole. (See Figure 1 below.) 6
An extensive body of research has established that a felony conviction or time in prison makes
individuals significantly less employable. This effect is not simply that individuals who commit
crimes were less likely to work in the first place. Rather, the best available evidence suggests that
felony convictions or time in prison has an independent impact that further lowers the employment
prospects of ex-offenders. Given the number of ex-offenders and the best estimate of the associated
reduction in employment suffered by this population, our calculations suggest that in 2008 the US
economy lost the equivalent of 1.5 to 1.7 million workers, or roughly a 0.8 to 0.9 percentage-point
reduction in the overall employment rate. Since over 90 percent of ex-offenders are men, the effect
on male employment rates was much higher, with ex-offenders lowering employment rates for men
by 1.5 to 1.7 percentage points. Even at the relatively low productivity rates of ex-offenders (they
typically have much less education than the average worker), the resulting loss of output that year
was likely somewhere between $57 and $65 billion.
The rise in the ex-offender population – and the resulting employment and output losses –
overwhelmingly reflect changes in the U.S. criminal justice system, not changes in underlying
criminal activity. In 2008, for example, both violent and property crimes were below their 1980 rates,
about when the current incarceration boom got underway. Instead, dramatic increases in sentencing
probabilities and sentence lengths, especially for drug-related offenses, account for both the increase
in the incarcerated population and the mushrooming of the ex-offender population that we
document here. 7

1 Data on prison and jail inmates for 2008 from Sabol, West, and Cooper (2009). Prisons are state and federal facilities,
usually run by the government, but sometimes on a contract basis by private companies, that usually hold convicted
criminals with sentences of a year or longer; jails are local facilities, usually run by local governments, but sometimes by
contractors, that usually hold convicted criminals with sentences of less than one year or unconvicted individuals
awaiting trial.
2 Data for probation and parole for 2008 from Glaze and Bonczar (2009). “Probation is a court-ordered period of
correctional supervision in the community generally as an alternative to incarceration. In some cases probation can be
a combined sentence of incarceration followed by a period of community supervision. Parole is a period of conditional
supervised release in the community following a prison term” (p. 1).
3 Public Safety Performance Project (2008). Separately, Schmitt, Warner, and Gupta (2010) estimate that one of every 48
working-age men was in prison or jail in 2008.
4 Public Safety Performance Project (2009).
5 For a recent overview of incarceration and crime in the United States, see Schmitt, Warner, and Gupta (2010). For a
modern history of incarceration in the United States, see Abramsky (2007).
6 High and low estimates for ex-prisoners and ex-felons vary according to assumptions about recidivism; see text below
for details.
7 See Schmitt, Warner, and Gupta (2010), pp. 7-9.

CEPR

Ex-offenders and the Labor Market

3

While we cannot undo the felony convictions and the prison sentences that have created today’s
large ex-offender population, we have substantial scope for improvement going forward. Since high
levels of incarceration are not the result of high levels of crime (and since the research consensus
also suggests that incarceration has a relatively small effect on lowering crime 8 ), changes in
sentencing today can greatly reduce the size of the ex-offender population in the future. The high
cost in terms of lost output to the overall economy also suggests the benefits of taking action to
reduce the substantial employment barriers facing current ex-offenders. 9
FIGURE 1
Estimates of Correctional Populations, 2008
16
High (13.9)

14
12

Low (12.3)

Millions

10
8
High (6.1)

6

5.1
Parole (0.8)

4
2.3
2
0

Low (5.4)

Prob. (4.3)

Jail (0.8)
Prison (1.5)
Prison and Jail
Inmates

Parolees and
Probationers

Ex-Prisoners

Ex-Felons

Source: Authors’ analysis of BJS data. High and low estimates for ex-prisoners and ex-felons vary according to
assumptions about recidivism.

Every indication is that, in the absence of some reform of the criminal justice system, the share of
ex-offenders in the working-age population will rise substantially in coming decades. Even if
imprisonment rates remain at current levels, the aging of today’s younger cohorts – which have
much higher rates of imprisonment than the older cohorts they will eventually replace – will
continue to raise the share of ex-offenders in the total population. In either case, the future
employment and output losses would be higher than what we have estimated here for 2008.

Estimating the Size of the Ex‐offender Population 
Unfortunately, there are no publicly available data that can provide a direct estimate of the exoffender population. Instead, we produce two indirect estimates of the formerly imprisoned
population, and then use these calculations as the basis for estimating the ex-felon population.
8 See, among others, Austin, et al. (2007), Irwin, Schiraldi, and Ziedenberg (1999), Public Safety Performance
Project (2007), Schmitt, Warner, and Gupta (2010), and Stemen (2007).
9 See Center for Employment Opportunities (2006) and Emsellem (2010).

CEPR

Ex-offenders and the Labor Market

4

Table 1 reports our final estimates of the ex-offender population using these various approaches.
The first two columns show the result of an analysis of administrative data that count the number of
prisoners released from state and federal prisons in each year from 1962 through 2008. Beginning
with the cohort of prisoners released in 1962, we “track” each year’s cohort of released prisoners
over time, assuming that the released population has an age and gender structure identical to the
incarcerated prison population in the year they were released. We use the assumed age structure to
“age” ex-prisoners out of the working-age population as they turn 65; similarly, we apply an estimate
of the annual death rate, by age, using a death rate that reflects this high-risk population. Finally, we
apply a high and a low estimate of the recidivism rate (about 40 percent of ex-prisoners return to
prison within three years of release), which substantially reduces our estimate of the ex-offender
population. 10 (For further details, see the Appendix.) This first procedure suggests that the exprisoner population in the United States in 2008 stood at between 5.4 million (using a high estimate
of recidivism) and 6.1 million (using a low estimate of recidivism).
TABLE 1
Estimated Size of Ex-offender Population, Age 18-64, 2008
(thousands)
Ex-prisoners
Ex-felons
Release data
Release data
Recidivism
Recidivism
Lifetime
Lifetime
Low
High
probability
Low
High
probability
6,094
5,427
5,504
13,851
12,333
12,508
Notes: Ex-felon population estimated from ex-prisoners, assuming: 90% of
prisoners are state prisoners, 10% are federal prisoners, 42% of felons
convicted in state courts are sentenced to prison, 62% of felons convicted in
federal courts are sentenced to prison (authors’ analysis of BJS data, 19922006, see text for details).

The third column shows the results of a separate analysis, which draws on survey data rather than
administrative records. We took the total number of men of each age from 18 to 64 in 2008 (from
the 2008 Current Population Survey), applied an age-specific estimate of their probability of ever
having been imprisoned, and then summed these across all age levels to produce an estimate of the
total male population age 18 to 64 that had ever been imprisoned. We estimated the age-specific
probability of ever having been imprisoned by age 30-to-34 from estimates 11 at three periods of time
(for birth cohorts from 1945-49, 1965-69, and 1975-79) and used linear interpolation for years where
no direct data were available. (For further details, see the Appendix.) The resulting estimate is 5.5
million. 12
The first three estimates in Table 1 are fairly close to each other. They are also close to two
independent estimates made by earlier researchers. Thomas Bonczar (2003) of the Bureau of Justice
Statistics (BJS), for example, used generation life table techniques (which were also the basis for the
Western (2006) and Western and Pettit (2010) estimates that we use here) to estimate the lifetime
probability of imprisonment in the U.S. population. Assuming age-specific rates of first incarceration
10 Our approach is similar in spirit to Uggen, Manza, and Thompson (2006), who generated estimates of the exprisoner and ex-felon populations through 2004.
11 Western and Pettit (2010) and Western (2006).
12 This estimate is based on data for men only, which we have scaled up for a total ex-prisoner population, using the
average share of men in the prison population in Table 2.

CEPR

Ex-offenders and the Labor Market

5

remained at 2001 levels, Bonczar concluded that the number of adults having ever served time in
prison would rise to 7.3 million by 2008. Given that about 1.6 million people were in state and
federal prisons in 2008, Bonczar’s forecast implies a 2008 ex-prisoner population of about 5.7
million. 13 Uggen, Manza, and Thompson (2006), using an approach similar to the one we have used
here, concluded that there were about 4.0 million ex-prisoners in 2004. 14
The final three columns of Table 1 give estimates of the ex-felon population. The ex-felon
population is larger than the ex-prisoner population because not all felons serve prison terms (more
than half are sentenced to jail or probation only). To produce these estimates, we used
administrative data 15 on the share of all felons sentenced to state or federal prison (about 44 percent)
in order to “scale up” the ex-prisoner population in the first half of Table 1 to the implied ex-felon
population. 16 (For details, see the Appendix.) This procedure suggests that the total ex-felon
population in 2008 was somewhere between 12.3 million and 13.9 million people. Again, our
estimates for 2008 are broadly consistent with the only other estimate of the ex-felon population of
which we are aware, by Uggen, Manza, and Thompson (2006), for 2004. 17 They used a more
elaborate estimation procedure than we do and concluded that there were about 11.7 million exfelons in the United States in 2004.
The share of ex-offenders in the working-age population will likely rise substantially in coming
decades. The most recent available data, for example, suggest that about 9.7 percent of 30-to-34
year-old men today have been in prison – the highest rate recorded since these kinds of data became
available in the 1970s. 18 As this cohort – and younger cohorts with as high or higher lifetime
probabilities of imprisonment – age, they will replace older cohorts that have lower rates, which will
raise the share of ex-offenders in the working-age population above current levels. Moreover, the
BJS has estimated that about 11.3 percent of males born in 2001 will be imprisoned at some point
during their lifetime, compared to just 3.6 percent of those born in 1974. 19 These higher projected
imprisonment rates for cohorts below age 30 imply large additional increases in the ex-offender
population over time.
We are also interested in the basic demographic characteristics of the ex-offender population. We
approximate these characteristics by applying demographic data for prisoners to the ex-prisoner and
ex-felon populations estimated in Table 1 and making some adjustments to reflect differences in
recidivism and sentencing across different populations. Table 2 provides some basic demographic
information on prisoners in 1960, 1980, 2000, and 2008. Across the entire period, men are the
overwhelming majority of prisoners (in excess of 90 percent). The prison population is also generally
far less educated than the general population. In 2008, more than one-third of prisoners had less
13 Bonczar (2003), p. 7. Bonczar estimated the ex-prisoner population to be about 4.3 million in 2001, up from about
1.6 million in 1974 (see Bonczar, 2003, Table 1).
14 Uggen, Manza, and Thompson (2006), Table 1.
15 Langan and Graziadei (1995), Langan (1996), Langan and Brown (1997), Brown and Langan (1997), Brown and
Langan (1999a), Brown and Langan (1999b), Durose, Levin, and Langan (2001), Durose and Langan (2003), Durose
and Langan (2004), Durose and Langan (2007), and Rosenmerkel, Durose, and Farole (2009).
16 For example, if we had estimated that there were 440,000 ex-prisoners, we would assume that these 440,000 exprisoners represented 44 percent of the total ex-felon population (the rest were sentenced to jail or probation), giving
us an estimate of total ex-felons of one million (440,000 / 0.44 = 1,000,000).
17 Uggen, Manza, and Thompson (2006), Table 2.
18 Calculation based on Western and Pettit (2010), Table 1.
19 Bonczar (2003), p. 1.

CEPR

Ex-offenders and the Labor Market

6

than a high school degree (compared to just over 10 percent in the non-institutional population 20 );
just over half had only a high school degree (compared to about 30 percent in the non-institutional
population); and only about 11 percent of prisoners had had any college-level education (compared
to almost 60 percent of the non-institutional population). In 1960, a large majority of prisoners were
white (about 62 percent, but this figure includes Latinos). The share of whites, however, has fallen
steadily to about only one-third in 2008. Meanwhile, African Americans have become heavily overrepresented in prison, making up almost 40 percent of prisoners in 2008 (compared to less than 15
percent of the non-institutional population). The share of Latinos in the total prison population has
been rising since at least 1980 and Latinos are also now substantially over-represented in the prison
population (just over 20 percent of prisoners in 2008, compared to about 15 percent of the noninstitutional population in the same year). Finally, prisoners are generally much younger than the
non-institutional population.
TABLE 2
Estimated Prisoner Demographics, 1960-2008
(percent)
Female
Male

1960
3.8
96.2

c. 1980
4.0
96.0

c. 2000
6.7
93.3

c. 2008
8.5
91.5

Men only
Less than high school
High school
Any college

85.5
10.6
3.9

51.0
35.0
14.0

39.6
49.3
11.1

36.6
52.0
11.4

62.3
36.1
-1.6

42.9
42.5
12.3
2.2

35.3
46.3
16.7
1.7

33.3
39.2
20.6
6.9

White
Black
Latino
Other

18-19
6.3
4.9
2.7
1.7
20-24
19.7
22.7
16.1
14.5
25-29
18.2
19.7
18.8
17.2
30-34
16.2
15.6
18.9
16.6
35-39
12.7
12.9
17.2
15.8
40-44
8.6
9.6
12.1
14.1
45-49
6.2
5.8
6.7
9.5
50-54
4.2
3.4
3.7
5.3
55-59
2.8
1.9
1.8
2.7
60-64
1.4
1.0
0.9
1.3
18-64
96.4
97.6
98.9
98.7
Notes: In 1960, Latino was not treated as a separate category (all 1960 data from Tables 4 and 25 of U.S. Bureau of
the Census, 1961); for 1980, race is for jail and prison inmates (Ewert, Pettit, and Sykes, 2010, Table 1); 1980
gender from BJS (1981b); all education data is for state prisoners only, federal prisoners (10% of all prisoners) are
slightly more educated; 1980 education is for 1979 (Pettit and Western, 2004, Table 2); 2000 education is for 1997
(Harlow, 2003, Table 6); 2008 education is for 2004 (Glaze and Maruschak, 2008, Appendix Table 16); 1980 age
groups interpolated from 1970 and 1991 data (U.S. Bureau of the Census, 1972 and Gilliard and Beck, 1998);
remaining data for 2000 from Harrison and Beck (2001), with age groups 45-49 and 50-54 broken out from 45-54
age group, age groups 55-59 and 60-64 broken out from 55+ age group, based on proportions from closest year
available (2007, from West and Sabol, 2008); remaining data for 2008 from Sabol, West, and Cooper (2009).

20 Authors’ analysis of the CEPR extract of the CPS Outgoing Rotation Group.

CEPR

Ex-offenders and the Labor Market

7

Table 3 gives estimates of the basic demographic characteristics of the ex-offender population. We
have assumed that the ex-offender population has the average demographic characteristics as the
prison population in Table 2 and have made adjustments to reflect the higher recidivism rate and the
higher likelihood of imprisonment conditional on committing a felony for black and Latino
offenders. 21 We use these estimates of the characteristics of the ex-offender population below when
we analyze the impact of the large ex-offender population on various labor-market outcomes. While
we believe that these estimates for subgroups can provide a general idea of the differences in the
sizes of relative subgroups in the ex-offender population, these calculations have a larger range of
uncertainty than our calculations for total ex-offender or the total male ex-offender populations and
should be taken as broadly indicative rather than as definitive estimates.
TABLE 3
Estimated Size of Ex-offender Population, Age 18-64, by Education and Race or Ethnicity, 2008
(thousands)
Ex-prisoners
Ex-felons
Release data
Release data
Recidivism
Recidivism
Lifetime
Lifetime
Low
High
probability
Low
High probability
All
6,094
5,427
5,504
13,851
12,333
12,508
Female
449
400
406
1,021
909
922
Male
LTHS
HS
College

5,645

5,026

5,098

12,829

11,424

11,585

2,743
2,073
829

2,443
1,846
738

2,477
1,872
748

6,235
4,712
1,883

5,552
4,195
1,677

5,630
4,255
1,701

White
2,230
1,985
2,013
5,867
5,224
Black
2,251
2,004
2,033
4,593
4,090
Latino
874
778
789
1,784
1,588
Notes: Authors’ analysis, using data in Tables 1 and 2. Race and ethnicity categories exclude the
“other” category in Table 2.

5,298
4,148
1,611

21 We use the average over the period 1980-2008. Based on BJS data for 1983 and 1994, we assume that, on average,
African American ex-prisoners have a recidivism rate that is about 9 percent higher than the average, Latinos about
14 percent higher than average, and whites about 9 percent lower than average. In the absence of good data on
recidivism rates by education, we apply the overall African American rate to less than high school educated workers,
the average rate to high school educated workers, and then adjust the population of ex-offenders with more
education so that all three educational categories sum to the total ex-offender population as calculated using the
average rate across all groups. A higher recidivism rate lowers estimates of the ex-offender population. Based on BJS
data covering 1992 through 2006, we assume that blacks (and Latinos, for which no separate data are available,
though Harris, Steffensmeier, Ulmer, and Painter-Davis (2009) suggest that Latinos face at least the same
disadvantage as blacks) have about an 11 percent higher chance than the average of being sentenced to prison after
committing a felony and that whites (which includes Latinos in these data) have about a 14 percent lower probability
than average of being sentenced to prison after a felony conviction. A higher probability of being sentenced to
prison after committing a felony reduces our estimates of ex-felons. The resulting calculations changes the mix of
ex-offenders across the three groups, but yields a total for these groups that is quite close to what the average
imprisonment rate for ex-felons implies; the small remaining discrepancy changes the share of the “other” racial and
ethnic group in the total male ex-offender population (not shown in Table 3). We also adjust the overall male and
female shares to adjust for the lower recidivism rate among women; but in the absence of data, we make no
additional adjustment by gender for the probability of being sentenced to prison conditional on a felony conviction.

CEPR

Ex-offenders and the Labor Market

8

The Effects of Imprisonment and Felony Conviction on 
Subsequent Employment 
A felony conviction or a prison or jail term can have a substantial negative impact on future job
prospects. 22 Researchers have identified several distinct channels for this effect. 23 Time behind bars
can lead to deterioration in a worker’s “human capital,” including formal education, on-the-job
experience, and even “soft skills” such as punctuality or customer relations. Incarceration can also
lead to the loss of social networks that can help workers find jobs; and, worse, provide former
inmates with new social networks that make criminal activity more likely. Incarceration or a felony
conviction can also impart a stigma that makes employers less likely to hire ex-offenders. In many
states, a felony conviction also carries significant legal restrictions on subsequent employment, 24
including limitations on government employment and professional licensing.
Quantifying the impact of incarceration or a felony conviction on subsequent labor-market
outcomes, however, is challenging. Many ex-prisoners and ex-felons struggled in the labor market
before their convictions and likely would have continued to have had problems even without trouble
with the law. 25 Nevertheless, a fairly large body of research has attempted to isolate the labor-market
effects of prison time and felony convictions. The five most common approaches have used: (1)
surveys of individuals that track offenders before and after their incarceration; (2) surveys of
employers attitudes about ex-felons; (3) “audits” that compare the employment prospects of
otherwise identical job applicants with and without felony convictions; (4) aggregate state- or citylevel data that compare labor-market outcomes across demographic groups with different
experiences of incarceration; and (5) administrative data that track offenders before and after their
incarceration. Taken together, this research consistently shows a substantial negative effect of a
felony conviction or time in prison or jail on the employment prospects of ex-offenders.
Longitudinal Surveys of Individuals 
Probably the most influential research on the impact of incarceration on subsequent labor-market
outcomes has used survey data that follows a large sample of individuals over time, and is thus able
to compare both offenders to non-offenders, and offenders before and after their time in prison and
jail. Almost all of this research has used the National Longitudinal Survey of Youth (NLSY), a
nationally representative sample of about 13,000 young men and women who were 14 to 22 years
old when they were first interviewed in 1979; these respondents were re-interviewed every year until

22 Incarceration may also work to improve an offender’s labor-market prospects if the time in prison or jail has a
rehabilitative effect or leads to the acquisition of additional education or training. High recidivism rates and the
results of empirical investigations (see below) suggest that this countervailing effect is likely to be small.
23 For recent reviews of research on the labor-market consequences of a criminal or prison record, see: Pager (2007,
Chapter 3); Holzer (2007, pp. 10-29); Western (2006, Chapter 5); The Pew Charitable Trusts (2010, pp. 9-17).
24 Ex-prisoners may also be less likely to work because some face high implicit marginal tax rates stemming from childsupport obligations, which generally accumulate while they are in prison. Between ongoing obligations and prisonrelated arrears, some ex-prisoners have large sums deducted from their pay checks for child support (Holzer 2007;
Holzer, Raphael, and Stoll 2004; Mincy and Sorenson 1998; and Garfinkel 2001.
25 The direction of causality is complicated. As Western (2002) notes: “men with few economic opportunities may turn
to crime” (p. 526) but also “desistance from crime is associated with social attachments and the normative bonds of
regular employment” (pp. 526-27).

CEPR

Ex-offenders and the Labor Market

9

1994 and every-other year since then. 26 Unlike most large, nationally representative surveys, the
NLSY interviews respondents even when they are in prison and jail, and notes the location of the
interview, which allows researchers to identify those initial respondents who were subsequently
incarcerated. Since the share of young women interviewed in prison or jail was small, all of the
NLSY studies reviewed here focused exclusively on men.
The earliest research using the NLSY identified large employment effects of incarceration. Freeman
(1991), for example, found that incarceration led to a 15 to 30 percentage-point decline in
subsequent employment rates. Grogger (1992) concluded that differences in incarceration rates
between young white and young black men accounted for about one third of the black-white
employment gap in the NLSY data.
Later research based on the NLSY, generally using more refined estimation techniques, has found
somewhat smaller, but still large, employment effects. Western and Beckett (1999) estimated that
incarceration reduced ex-offenders’ average annual weeks of work by about five weeks, relative to a
baseline of 42 weeks (about a 12 percent decline in employment). The effect diminishes over time,
but remains statistically significant over the seven year period they studied. 27 Western (2006) found
that time in jail or prison cut employment by about five weeks per year (9.7 percent) for young white
men; about eight weeks per year (15.1 percent) for young black men; and about eight weeks per year
(13.7 percent) for young Hispanic men. 28 Raphael (2007) concluded that incarceration lowered
annual weeks worked by 6-11 weeks (which, assuming an average of 48 weeks per year, is roughly a
13 to 23 percent decline in employment). The most recent analysis of the NLSY data, by the Pew
Charitable Trusts (2010), found that incarceration reduced the average number of weeks worked by
a 45 year old male by about 9 weeks (about 19 percent). 29
Geller, Garfinkel, and Western (2006) have used another (short) longitudinal study, the Fragile
Families and Child Wellbeing Survey (FFCWS), 30 to produce independent estimates that also suggest
a large effect of incarceration on employment. The FFCWS follows the families of a cohort of
almost 5,000 children born in 20 U.S. cities between 1998 and 2000, including the children’s married
and unmarried parents. The survey has detailed information on the parents’ economic situation as
well as their incarceration status. 31 They conclude that “...employment rates of formerly incarcerated
men are about 6 percentage points lower than for similar men who have not been incarcerated.”
Employer Surveys  
Harry Holzer and collaborators (Holzer, 1996; Holzer, Raphael, and Stoll, 2004, 2006, 2007) have
asked employers directly about their attitudes toward hiring job applicants with criminal records.
They interviewed about 3,000 employers in four large metropolitan areas (Atlanta, Boston, Detroit,
and Los Angeles) over the period 1992-94, and then did a follow-up study of 600 employers in Los
26
27
28
29

For more details on the NLSY, see the Bureau of Labor Statistics, http://www.bls.gov/nls/nlsy79.htm.
See Western and Beckett (1999), Figure 3 and discussion on pages 1050-51.
See Western (2006), Figure 5.1 and discussion on p. 119.
See Pew Charitable Trusts (2010), Figure 4. Bruce Western and Becky Pettit conducted the data analysis for The Pew
Charitable Trusts.
30 For more information, see The Fragile Families and Child Wellbeing Study at
http://www.fragilefamilies.princeton.edu/.
31 Geller, Garfinkel, and Western also focus exclusively on men; they use fathers’ self-reported incarceration status as
well as independent reports of fathers’ incarceration status provided by mothers.

CEPR

Ex-offenders and the Labor Market

10

Angeles in 2001. 32 In the initial survey, they sought to “gauge... employer willingness to hire a variety
of workers with various stigmas – such as having a criminal record, being a welfare recipient, having
an unstable work history, etc. – into the job filled by the last worker hired at the firm” (Holzer, 2007,
pp. 11-12) In the follow-up survey, they “also asked about actual hiring of ex-offenders, as well as
self-reported willingness to do so; and asked a more detailed set of questions about employer
perceptions of offenders and their willingness to hire them” (Holzer, 2007, p. 12).
Employers reported that they were much less likely to hire ex-offenders. The vast majority of
employers (80 to 90 percent), for example, said that they would ‘definitely’ or ‘probably’ hire
“former welfare recipients, workers with little recent work experience or lengthy unemployment, and
other stigmatizing characteristics” (Holzer, 2007, p. 14). By contrast, only about 40 percent of
employers would ‘definitely’ or ‘probably’ hire applicants with criminal records, especially for jobs
that involved dealing with customers or handling money. 33
Audit Studies 
Another group of studies goes beyond employers’ self-reported attitudes to examine actual employer
behavior using “audit” studies. In audit studies, researchers present actual employers filling actual
vacancies with carefully constructed job applications or specially trained actors posing as applicants.
In both situations, the basic features of the applications or the applicants are designed to be identical
on all relevant hiring dimensions except the feature of interest, in this case, criminal record.
Researchers can then measure the extent of barriers facing ex-offenders by examining differences in
“call back” and job offer rates across the two groups.
The earliest research used “correspondence” studies, involving letters or paper applications rather
than in-person applicants. Richard Schwartz and Jerome Skolnick (1962), for example, had a
researcher, posing as an employment agent, present prospective employers with files on possible
employees. The researcher then asked if the employer would be interested in hiring the candidate.
Employers were less likely to express interest in applicants with a criminal record. Moreover, as
Pager (2007, p. 50) notes: their “findings ... suggest that mere contact [emphasis in original] with the
criminal justice system can have significant repercussions, with records of “arrest,” “conviction,”
and “incarceration” conveying a stigma differing in degree but not kind.” 34

32 For more information on the “Multi-City Study of Urban Inequality” see
http://www.sociology.emory.edu/MCSUI/.
33 The research by Holzer, Raphael, and Stoll also suggests that employers may engage in statistical discrimination
against members of groups (particularly, young less-educated black men) with high incarceration rates. As Holzer
(2007) points out, this may reduce estimates of the effect of incarceration on subsequent wages and employment in
studies using the NLSY or the FFCWS.
34 Pager (2007) has identified two other related studies using correspondence methodologies with broadly consistent
findings. The first is R.H. Finn and Patricia A. Fontaine (1985, p. 353): “Employment applications were prepared for
20 fictitious job applicants, and were then rank ordered by 225 undergraduate students enrolled in personnel
management classes based on perceived suitability for employment in an entry-level job. Job applicants differed from
each other on three treatments: type of crime allegedly committed, judicial outcome, and sex. Employability scores
were derived for each applicant by converting the rank orders to a normal distribution with a given mean and
standard deviation. Analysis of the data revealed a clear bias against all applicants who had allegedly committed a
crime. The magnitude of the bias was related to the type of crime allegedly committed, and to the judicial outcome.”
And Dov Cohen and Richard E. Nisbett (1997).

CEPR

Ex-offenders and the Labor Market

11

More recently, Devah Pager (2003) carried out a large, carefully designed in-person audit of 350
employers in Milwaukee in 2001 (see also Pager 2007). She sent out pairs of otherwise identical
white applicants (100) and otherwise identical black applicants (250), where one member of each
matched pair had a criminal record. She found that respondents with a criminal record were less
than half as likely as those with no criminal record to get a call back. 35
Aggregated Geographic Data 
Harry Holzer, Paul Offner, and Elaine Sorensen (2005) have used variation in incarceration rates
across the U.S. states to measure the effect of incarceration on employment rates of young black
men. They conclude that “...post-incarceration effects ... contribute to the decline in employment
activity among young black less-educated men in the past two decades, especially among those age
25-34” (p. 1). More specifically, they estimate that “the roughly 3 percentage point increase in lagged
incarceration from the early 1980s through the year 2000 reduced employment of young black men
by 2-4 percentage points” (pp. 19-20).
Administrative Data 
A final set of studies uses administrative data, rather than survey data, to track the labor-market
outcomes of inmates before and after their time in prison or jails. Typically, these studies link state
or federal data on released prisoners to their employment record (including employment status and
earnings), as revealed by their participation in state unemployment insurance systems. In strong
contrast to the other research reviewed so far, these studies have generally found little or no negative
effect of incarceration on employment (see Waldfogel (1994); Needels (1996), using data from
Georgia; Grogger (1992), using data from California; Cho and LaLonde (2005), using data from
Illinois; Kling (2006), using data from California and Florida; Pettit and Lyons (2007), using data
from Washington; and Sabol (2007), using data from Ohio).
Holzer (2007), however, reviews most of these studies and argues: “a number of problems plague all
of these studies based on administrative data ... [unemployment insurance] records only capture
earnings in formal jobs ... [and] would automatically exclude public sector jobs, any employment that
occurs in another state, any self-employment, and most importantly – any casual and informal work
for cash... “ 36 (p. 22). He notes that the employment rates, both before and after incarceration, tend
to be “dramatically lower” in the administrative-data studies than they are in the NLSY studies (p.
23). He continues: “Another problem arises from the absence of a clear control or comparison
group of non-offenders in at least some of these studies. Simple pre-post incarceration comparisons
of employment and earnings outcomes may tell us little about the counterfactual situation that
would have existed in the absence of incarceration” (p. 23). He concludes that: “These
considerations suggest that the studies based on administrative data might well understate the
negative impacts of incarceration on subsequent earnings or employment” (p. 25).
Assessment of Employment Effects 
The preceding review suggests that most of the available research finds that incarceration or a
criminal record has moderate to large effects on subsequent employment levels. The wide range of
35 See Pager, 2007, Figure 5.1, p. 91.
36 Holzer (2007, p. 22) notes that: “part-time and casual employment likely characterize much work among offenders and ex-offenders,
both pre- and post-incarceration.” [emphasis in original]

CEPR

Ex-offenders and the Labor Market

12

research techniques, different populations studied, and metrics used to express the employment
impact, however, present a specific challenge for our purposes. We are interested in estimating the
likely employment impact of the large ex-offender population on the employment rate of all
working-age men. But, the preceding estimates of effects generally do not translate directly to the
exercise we are conducting here. The estimates based on longitudinal surveys of individuals come
closest to capturing the effect we are trying to measure, and generally suggest moderate to large
employment effects. The findings based on aggregate state-level data are consistent with small to
moderate effects, but are not as directly applicable. The employer surveys and audit studies are
generally consistent with large, negative effects of incarceration on subsequent employment, but are
even more difficult to translate to the kinds of reduction in employment probabilities that are most
useful for our analysis. The administrative studies, which find at most only small negative effects of
incarceration on employment, are more in spirit methodologically with the longitudinal studies, but
have technical difficulties and are inconsistent with all the other available data.
In the analysis we conduct in the next section of the paper, therefore, we use three separate
estimates of the employment effects of incarceration. In the low-effects scenario, we assume that exprisoners or ex-felons pay an employment penalty of five percentage points (roughly consistent with
the largest effects estimated using administrative data and the lower range of effects estimated using
the aggregate data and survey data). In the medium-effects scenario, we assume that the employment
penalty faced by ex-prisoners and ex-felons is 12 percentage points, which is consistent with the
bulk of the survey-based studies. In the high-effects scenario, we assume that the employment
penalty is 20 percentage points, which is consistent with the largest effects estimated in the surveybased studies, as well as, arguably, the findings of the employer surveys and audit studies.

Estimating the Impact of the Ex‐offender Population on 
Total Employment and Output 
In this section, we use our estimates of the size of the ex-offender population (Section II) and the
outside estimates of the impact of having a prison or felony record on subsequent employment
probabilities (Section III) to assess the likely impact of the large and growing ex-offender population
on key labor-market outcomes.
As a first step, we compare the size of the ex-offender population to the total non-institutional
working-age population. Table 4 expresses the estimated ex-offender population in Table 3 as a
percent of the total non-institutional working-age population. 37 In 2008, ex-prisoners were 2.9 to 3.2
percent of the total working-age population (excluding those currently in prison or jail), or about
one in 33 working-age adults. Ex-felons were a larger share of the total working-age population: 6.6
to 7.4 percent, or about one in 15 working-age adults. Ex-offenders were heavily concentrated
among men. Between 5.4 and 6.1 percent, or about one in 17 working-age adults, were ex-prisoners;
between 12.3 and 13.9 percent, or about one in 8 working-age adults, were ex-felons. AfricanAmerican men and men with less than a high school degree had the highest concentration of exoffenders.

37 Total population age 18 to 64 from the 2008 CPS.

CEPR

Ex-offenders and the Labor Market

13

TABLE 4
Estimated Ex-offender Population as Share of Civilian Non-institutional Population, 2008
(percent)
Ex-prisoners
Ex-felons
Release data
Release data
Recidivism
Lifetime
Recidivism
Lifetime
Low
High probability
Low
High probability
All
3.2
2.9
2.9
7.4
6.6
6.7
Female
0.5
0.4
0.4
1.1
1.0
1.0
Male
LTHS
HS
College

6.1

5.4

5.5

13.9

12.3

12.5

25.2
6.8
1.6

22.4
6.1
1.4

22.7
6.1
1.5

57.2
15.5
3.7

50.9
13.8
3.3

51.6
14.0
3.3

White
3.6
3.2
3.3
9.5
8.5
Black
21.4
19.1
19.4
43.8
39.0
Latino
6.0
5.3
5.4
12.2
10.9
Notes: Authors’ analysis of Table 3 and Current Population Survey data for population.

8.6
39.5
11.0

Next, we use the relative size of the ex-offender population in Table 4 to estimate the impact on
employment rates in 2008. Table 5 shows the impact on the working-age male population assuming
a low, medium, and high effect of imprisonment or a felony record on subsequent employment.
Assuming a low effect (a reduction of about 5 percentage points relative to a comparable worker
without prison time or a felony conviction), in 2008, the ex-offender population lowered overall
male employment about 0.3 to 0.7 percentage points. Assuming a mid-range effect (a 12-percentagepoint employment penalty), ex-offenders lowered overall male employment between 0.7 and 1.7
percentage points. Finally, assuming a large effect (a 20-percentage-point penalty), ex-offenders cut
male employment rates 1.1 to 2.8 percentage points.
TABLE 5
Estimated Reduction in Employment-to-Population Rate, All Males 2008
Ex-prisoners
Ex-felons
Release data
Release data
Recidivism
Recidivism
Lifetime
Lifetime
Low
High probability
Low
High probability
(a) Assuming 5-percentage-point employment penalty for ex-offenders
0.3
0.3
0.3
0.7
0.6
0.6
(b) Assuming 12-percentage-point employment penalty for ex-offenders
0.7
0.7
0.7
1.7
1.5
(c) Assuming 20-percentage-point employment penalty for ex-offenders
1.2
1.1
1.1
2.8
2.5
Notes: Authors’ analysis of Table 4.

1.5

2.5

Table 6 presents results of a similar exercise for additional groups of workers, using only the midrange estimate of the employment penalty (a 12-percentage-point employment penalty for exoffenders). According to these estimates, in 2008, the ex-offender population reduced employment

CEPR

Ex-offenders and the Labor Market

14

rates for 18 to 64 year olds as a whole by 0.3 to 0.9 percentage points. The impact was biggest for
African-American men, lowering employment rates between 2.3 and 5.3 percentage points. Men
with less than a high school education were also especially hard hit, with an estimated decline in
employment rates of 2.7 to 6.9 percentage points as a result of the large ex-offender population; to
put this decline in context, between 1979 and 2008, employment rates for less than high school
educated men in the non-institutional population fell a total of 9.3 percentage points.
TABLE 6
Estimated Decline in Employment Rates in 2008
(Percentage points; Assuming 12-percentage-point Employment Penalty)
Ex-prisoners
Ex-felons
Release data
Release data
Recidivism
Recidivism
Lifetime
Lifetime
Low
High probability
Low
High probability
All
0.4
0.3
0.4
0.9
0.8
0.8
Female
0.1
0.1
0.1
0.1
0.1
0.1
Male

0.7

0.7

0.7

1.7

1.5

1.5

3.0
0.8
0.2

2.7
0.7
0.2

2.7
0.7
0.2

6.9
1.9
0.4

6.1
1.7
0.4

6.2
1.7
0.4

White
0.4
0.4
0.4
Black
2.6
2.3
2.3
Latino
0.7
0.6
0.6
Notes: Authors’ analysis of Tables 4 and 5.

1.1
5.3
1.5

1.0
4.7
1.3

1.0
4.7
1.3

LTHS
HS
College

Ex-offenders, of course, bear the direct cost of these lower employment rates, in the form of lower
lifetime earnings. But, the economy as a whole also pays a price in reduced output of goods and
services. Using the estimated reduction in total employment rates of 0.8 to 0.9 percentage points
(from columns four and five of the first row of Table 6), and assuming that ex-offenders produce
only one-half the output of the average worker, we estimate that the large ex-offender population
cost the United States about 0.4 to 0.5 percentage points of GDP in 2008, or roughly $57 to $65
billion.

Conclusion 
Bruce Western and Katherine Beckett (1999) have rightly called the criminal justice system a U.S.
labor-market institution. Our estimates suggest that ex-offenders lower overall employment rates as
much as 0.8 to 0.9 percentage points; male employment rates, as much as 1.5 to 1.7 percentage
points; and those of less-educated men as much as 6.1 to 6.9 percentage points. These employment
losses hit ex-offenders hardest, but also impose a substantial cost on the U.S. economy in the form
of lost output of goods and services. In GDP terms, we estimate that in 2008 these employment
losses cost the country $57 to $65 billion per year.

CEPR

Ex-offenders and the Labor Market

15

Appendix 
We use two methods to estimate the number of ex-prisoners ages 18-64 in the United States. The first is
based on annual releases from state and federal prisoners in the United States. The second is based upon
lifetime probabilities of being incarcerated.
Releases 
We start with published data on total prisoners, total admissions, and from 1977 on, total releases, in each
year. Before 1977, we estimate that total number of releases each year from federal and state prison
admissions using data on total prison admissions and the total prison population. For example, in 1962 (the
first year we look at because an 18-year-old released in that year would be 64 in 2008), there were 89,082
admissions; adding this figure to the total number of prisoners in 1961 (220,149) results would give a total
of 309,231 prisoners in 1962. However, we know from data on the total prisoner population in 1962 that
there were only 218,830 prisoners in that year, which implies that there were 90,401 releases in 1962. We do
a similar calculation for each year up through 1976, and from 1977 we use the directly reported release
figure. (See Appendix Table 1).
APPENDIX TABLE 1
Total, Admitted, and Released Prisoners, 1962-2008
Year
Total Admissions
Releases
Year
Total Admissions
Releases
1961
220,149
N/A
N/A
1985
480,568
271,366
234,496
1962
218,830
89,082
90,401
1986
522,084
304,858
263,181
1963
217,283
87,826
89,373
1987
560,812
339,762
305,098
1964
214,336
87,578
90,525
1988
603,732
379,742
336,822
1965
210,895
87,505
90,946
1989
680,907
460,798
385,479
1966
199,654
77,857
89,098
1990
739,980
474,128
419,783
1967
194,896
77,850
82,608
1991
789,610
480,046
436,991
1968
187,914
72,058
79,040
1992
846,277
495,756
447,105
1969
196,007
75,277
67,184
1993
932,074
518,562
456,408
1970
196,429
79,351
78,929
1994
1,016,691
541,434
456,942
1971
198,061
89,395
87,763
1995
1,085,022
562,724
491,858
1972
196,092
99,440
101,409
1996
1,137,722
555,992
504,289
1973
204,211
109,484
101,365
1997
1,194,334
584,177
528,848
1974
218,466
119,529
105,274
1998
1,248,370
615,226
561,020
1975
240,593
129,573
107,446
1999
1,304,188
614,985
574,804
1976
262,833
146,388
124,148
2000
1,329,367
625,219
604,858
1977
285,456
163,203
147,895
2001
1,345,217
638,978
628,626
1978
294,396
162,574
154,484
2002
1,380,516
661,712
630,176
1979
301,470
172,753
166,132
2003
1,408,361
686,437
656,384
1980
315,956
182,617
169,826
2004
1,433,793
699,812
672,202
1981
353,673
212,264
174,955
2005
1,462,866
733,009
701,632
1982
395,516
230,834
188,435
2006
1,504,660
749,798
713,473
1983
419,346
250,061
225,856
2007
1,532,850
742,875
721,161
1984
443,398
246,260
221,768
2008
1,540,036
739,132
735,454
Source: 1962-1970, 1975 admissions from Cahalan (1986) p. 36 (data for missing years was not
available, and so was interpolated from adjacent years); 1961-1976 total from Cahalan, p. 35; 1977-1998
data
from
the
BJS’s
National
Corrections
Reporting
Program,
available
at
http://bjs.ojp.usdoj.gov/content/dtdata.cfm; 1999-2008 data from BJS, various years.

Next, we break down these annual releases into age groups (18-19, 20-24, 25-29, 30-34, 35-39, 40-44, 45-49,
50-54, 55-59, and 60-64) assuming that the released prisoners have same age structure as inmates (see Table
2) in that year; where no direct estimates of the population structure were available for a particular year, we
use linear interpolation between available estimates. Appendix Table 2 shows the resulting estimates of the

CEPR

Ex-offenders and the Labor Market

16

age structure in each year. (Within each age group, for simplicity, we further assume that individuals are
evenly spread across each age; for example, we assume that one-fifth of the 20-24 year olds are 20, one-fifth
21, and so on). We use the one-year age cohorts to “age-out” ex-prisoners after they turn 65.
APPENDIX TABLE 2
Released Prisoners by Age Group, 1962-2008
Year
18-19
20-24
25-29
30-34
35-39
40-44
45-49
50-54
55-59
60-64
1962
5,668 18,879 16,830 14,305 11,103
7,741
5,465
3,602
2,402
1,239
1963
5,585 19,239 16,813 13,927 10,785
7,626
5,328
3,478
2,298
1,195
1964
5,638 20,070 17,207 13,889 10,730
7,697
5,320
3,439
2,250
1,180
1965
5,645 20,749 17,464 13,735 10,585
7,705
5,269
3,371
2,182
1,155
1966
5,512 20,901 17,284 13,242 10,178
7,521
5,087
3,220
2,061
1,101
1967
5,093 19,910 16,186 12,079
9,260
6,949
4,647
2,909
1,840
993
1968
4,856 19,559 15,641 11,367
8,690
6,625
4,380
2,710
1,693
924
1969
4,114 17,058 13,426
9,501
7,243
5,610
3,666
2,242
1,381
763
1970
4,816 20,548 15,927 10,972
8,339
6,567
4,241
2,560
1,555
869
1971
5,222 22,487 17,853 12,509
9,433
7,335
4,685
2,818
1,722
957
1972
5,879 25,567 20,794 14,812 11,085
8,513
5,379
3,223
1,982
1,095
1973
5,722 25,139 20,950 15,162 11,265
8,547
5,341
3,188
1,973
1,084
1974
5,782 25,675 21,930 16,118 11,892
8,916
5,511
3,276
2,041
1,115
1975
5,737 25,763 22,557 16,829 12,333
9,140
5,587
3,309
2,075
1,127
1976
6,440 29,258 26,266 19,883 14,477 10,606
6,413
3,782
2,388
1,288
1977
7,446 34,246 31,531 24,207 17,516 12,690
7,589
4,457
2,833
1,519
1978
7,542 35,137 33,187 25,829 18,579 13,313
7,874
4,604
2,947
1,571
1979
7,858 37,103 35,960 28,362 20,283 14,378
8,410
4,897
3,156
1,671
1980
7,773 37,230 37,036 29,591 21,044 14,761
8,538
4,950
3,213
1,691
1981
7,741 37,635 38,440 31,101 21,999 15,272
8,736
5,042
3,296
1,723
1982
8,051 39,760 41,708 34,162 24,038 16,518
9,344
5,368
3,536
1,836
1983
9,305 46,727 50,359 41,741 29,224 19,882 11,121
6,359
4,220
2,177
1984
8,798 44,970 49,809 41,767 29,100 19,605 10,843
6,171
4,126
2,114
1985
8,946 46,587 53,049 44,991 31,199 20,817 11,385
6,448
4,345
2,210
1986
9,639 51,204 59,967 51,421 35,495 23,461 12,686
7,150
4,856
2,453
1987
10,709 58,104 70,015 60,686 41,706 27,311 14,602
8,189
5,605
2,811
1988
11,309 62,761 77,844 68,183 46,657 30,276 16,004
8,929
6,162
3,068
1989
12,354 70,243 89,717 79,391 54,101 34,792 18,182 10,092
7,022
3,470
1990
12,814 74,768 98,385 87,935 59,682 38,044 19,656 10,852
7,614
3,734
1991
12,673 76,036 103,130 93,079 62,927 39,766 20,310 11,153
7,892
3,841
1992
12,817 76,604 101,865 93,668 66,693 42,922 22,272 12,230
8,035
3,911
1993
12,929 77,165 100,876 94,326 70,042 45,698 24,012 13,186
8,172
3,977
1994
12,788 76,374 98,396 93,399 71,756 47,303 25,114 13,791
8,160
3,971
1995
13,612 81,377 103,583 99,602 78,693 52,334 28,019 15,386
8,776
4,271
1996
13,937 83,475 104,566 101,334 81,806 54,466 29,298 16,088
8,833
4,299
1997
14,279 83,558 98,895 101,539 92,548 63,991 35,163 19,309
9,270
4,512
1998
14,944 88,501 104,887 108,009 98,148 67,743 37,363 20,517
9,859
4,798
1999
15,103 90,532 107,441 110,963 100,527 69,264 38,345 21,056 10,127
4,929
2000
15,795 94,954 113,039 116,747 105,713 73,070 40,518 22,250 10,748
5,231
2001
17,253 104,310 120,160 121,423 108,986 72,938 40,080 22,011
9,984
4,859
2002
17,211 103,723 120,660 122,258 109,566 73,455 39,993 21,957
9,896
4,815
2003
12,249 102,187 114,251 110,944 106,286 92,174 57,380 31,487 14,917
7,261
2004
11,533 104,454 118,613 110,690 107,877 93,484 59,406 32,577 17,397
8,473
2005
13,206 110,736 124,902 116,018 108,815 96,858 63,140 34,631 16,964
8,255
2006
10,971 99,265 124,010 116,316 110,711 102,352 71,443 39,173 20,237
9,846
2007
11,621 103,272 123,267 120,256 115,693 103,413 68,644 37,639 19,149
9,316
2008
11,843 105,010 125,306 122,536 118,190 105,249 70,198 38,585 19,675
9,646
Source: 1960: U.S. Bureau of the Census (1961), Table 4; 1970: U.S. Bureau of the Census (1972), Table
3; 1991, 1997, 1999-2008: BJS Prisoner Series reports. For years where no data was available for
prisoners’ ages, we interpolate data from adjacent years. 1991, 1997, 1999-2006 age-groups 45-49 and 5054 estimated from 45-54 age group, 55-59 and 60-64 from 55+ age group, based on 2007 proportions.

CEPR

Ex-offenders and the Labor Market

17

Next, we apply age-group-specific three-year recidivism rates (with an upward adjustment to account for any
recidivism beyond the initial three-year period after release; see Appendix Table 3) to the share of releases
in each year that eventually returned to prison. This procedure helps us to avoid double counting (which
would occur if released prisoners were released in one year, readmitted to prison later, and released again in
a subsequent year).
APPENDIX TABLE 3
Age-group-specific Recidivism Rates, 1962-2008
Year 18-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64
(a) 3-year Recidivism Rate plus 10 percentage points (high recidivism estimate)
1962-1983
54.9
54.9
53.2
53.0
46.5
40.7
35.7
35.7
35.7
35.7
1984
55.5
55.5
54.0
54.1
47.9
42.5
37.1
37.1
37.1
37.1
1985
56.2
56.2
54.9
55.1
49.3
44.2
38.5
38.5
38.5
38.5
1986
56.8
56.8
55.7
56.2
50.7
46.0
39.8
39.8
39.8
39.8
1987
57.5
57.5
56.6
57.3
52.1
47.7
41.2
41.2
41.2
41.2
1988
58.1
58.1
57.4
58.4
53.5
49.5
42.6
42.6
42.6
42.6
1989
58.8
58.8
58.3
59.4
55.0
51.2
44.0
44.0
44.0
44.0
1990
59.4
59.4
59.1
60.5
56.4
53.0
45.4
45.4
45.4
45.4
1991
60.1
60.1
60.0
61.6
57.8
54.7
46.8
46.8
46.8
46.8
1992
60.7
60.7
60.8
62.7
59.2
56.5
48.1
48.1
48.1
48.1
1993
61.4
61.4
61.7
63.7
60.6
58.2
49.5
49.5
49.5
49.5
1994-2008
62.0
62.0
62.5
64.8
62.0
60.0
50.9
50.9
50.9
50.9
(b) 3-year Recidivism Rate plus 5 percentage points (low recidivism estimate)
1962-1983
49.9
49.9
48.2
48.0
41.5
35.7
30.7
30.7
30.7
30.7
1984
50.5
50.5
49.0
49.1
42.9
37.5
32.1
32.1
32.1
32.1
1985
51.2
51.2
49.9
50.1
44.3
39.2
33.5
33.5
33.5
33.5
1986
51.8
51.8
50.7
51.2
45.7
41.0
34.8
34.8
34.8
34.8
1987
52.5
52.5
51.6
52.3
47.1
42.7
36.2
36.2
36.2
36.2
1988
53.1
53.1
52.4
53.4
48.5
44.5
37.6
37.6
37.6
37.6
1989
53.8
53.8
53.3
54.4
50.0
46.2
39.0
39.0
39.0
39.0
1990
54.4
54.4
54.1
55.5
51.4
48.0
40.4
40.4
40.4
40.4
1991
55.1
55.1
55.0
56.6
52.8
49.7
41.8
41.8
41.8
41.8
1992
55.7
55.7
55.8
57.7
54.2
51.5
43.1
43.1
43.1
43.1
1993
56.4
56.4
56.7
58.7
55.6
53.2
44.5
44.5
44.5
44.5
1994-2008
57.0
57.0
57.5
59.8
57.0
55.0
45.9
45.9
45.9
45.9
Source: Beck and Shipley (1989) and Langan and Levin (2002). Data on recidivism of the
overall prison population is only available for years 1983 and 1994. For years before 1983, the
rate in 1983 was used; for years after 1994, the rate in 1994 was used; for years between these
two, data was estimated by interpolation.

Finally, we also apply an age-specific mortality rate using general population rates, adjusted up by 20
percent, following Bonczar (2003, p. 11), to allow for this higher-risk population. This procedure lowers our
estimate of the ex-prisoner population slightly.
Appendix Table 4 shows our final estimates of the ex-prisoner population, by age, in 2008, excluding those
in prison in that same year.

CEPR

Ex-offenders and the Labor Market

18

APPENDIX TABLE 4
Ex-Prisoners in 2008 by Age Group, Estimated from Releases
18-19
20-24
25-29
30-34
35-39
40-44
45-49
50-54
55-59
60-64
Total
High Recidivism Rate
Number
6,707 142,320 364,643 542,658 709,261 863,463 906,816 795,535 624,589 470,638 5,426,630
% of total
0.1
2.6
6.7
10.0
13.1
15.9
16.7
14.7
11.5
8.7
100.0
Low Recidivism Estimate
Number
7,590 161,046 412,867 615,743 804,392 975,935 1,018,434 887,251 692,225 518,865 6,094,349
% of total
0.1
2.6
6.8
10.1
13.2
16.0
16.7
14.6
11.4
8.5
100.0

Lifetime Probability 
Separately, we also estimated the lifetime probability of imprisonment using estimates of the share of men
ages 30-34 that had ever spent time in prison at three points in time (men born in 1945-49, 1965-69, and
1975-79), drawn from Western (2006) and Western and Pettit (2010), based on BJS estimates (including
Bonczar, 2001).
For each year between 1944 and 2008, we calculated the approximate share of 30-34 year olds that had ever
been incarcerated; where available, we used estimates that corresponded to the year; when no direct
estimates were available we used linear interpolation; for birth years that had not yet reached 30-34 in 2008,
we adjusted the 1974-79 birth-year rates down based on the ratio of incarceration rate for each age younger
age group to the rate for the 30-34 year old cohort, using Bonczar (2003), Table 3.
We then multiplied these probabilities by the corresponding male population of each age in 2008 (from
published Census estimates), and summed these to produce an estimate of the total population that had ever
been incarcerated. This estimate is based on data for men only, which we have scaled up for a total exprisoner population, using the average share of men in the prison population in Table 2.
For more details, spreadsheets of the data and calculations described here are available upon request.

CEPR

Ex-offenders and the Labor Market

19

References 
Abramsky, Sasha. 2007. American Furies: Crime, Punishment, and Vengeance in the Age of Mass Imprisonment.
Boston: Beacon Press.
James Austin, Todd Clear, Troy Duster, David F. Greenberg, John Irwin, Candace McCoy, Alan Mobley,
Barbara Owen, and Joshua Page. 2007. “Unlocking America.” Washington, DC: The JFA Institute.
http://www.jfa-associates.com/publications/srs/UnlockingAmerica.pdf
Beck, Allen J. 2000. “Prisoners in 1999.” Washington, DC: U.S. Department of Justice, Bureau of Justice
Statistics. http://bjs.ojp.usdoj.gov/content/pub/pdf/p99.pdf
Beck, Allen J. and Bernard E. Shipley. 1989. “Recidivism of Prisoners Released in 1983.” Washington,
DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/rpr83.pdf
Bonczar, Thomas P. 2003. “Prevalence of Imprisonment in the U.S. Population, 1974-2001.”
Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/piusp01.pdf
Brown, Jodi M. and Patrick A. Langan. 1997. “Felony Sentences in the United States, 1994.” Washington,
DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/fsus94.pdf
Brown, Jodi M. and Patrick A. Langan. 1999a. “Felony Sentences in State Courts, 1996.” Washington,
DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/fssc96.pdf
Brown, Jodi M. and Patrick A. Langan. 1999b. “Felony Sentences in the United States, 1996.”
Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/fsus96.pdf
Bureau of Justice Statistics. 1981a. “Prisoners in 1980.” Washington, DC: U.S. Department of Justice,
Bureau of Justice Statistics.
Bureau of Justice Statistics. 1981b. “Prisoners in Midyear 1981.” Washington, DC: U.S. Department of
Justice, Bureau of Justice Statistics.
Cahalan, Margaret Werner. 1986. “Historical Corrections Statistics in the United States, 1850-1984.”
Rockville, MD: U.S. Department of Justice, Bureau of Justice Statistics.
http://www.ncjrs.gov/pdffiles1/pr/102529.pdf
Center for Employment Opportunities. 2006. “The Power of Work.” New York: Center for Employemnt
Opportunities. http://www.mdrc.org/publications/426/full.pdf
Cho, Rosa and Robert J. LaLonde. 2005. “The Impact of Incarceration in State Prison on the
Employment Prospects of Women.” IZA Discussion Paper No. 1792. Bonn: Institute for Labor
Studies.
Cohen, Dov and Richard E. Nisbett. 1997. “Field Experiments Examining the Culture of Honor: The
Role of Institutions in Perpetuating Norms about Violence.” Personality and Social Psychology Bulletin, vol.
23, no. 11, pp. 1188 -99.
Durose, Matthew and Patrick A. Langan. 2003. “Felony Sentences in State Courts, 2000.” Washington,
DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/fssc00.pdf
Durose, Matthew and Patrick A. Langan. 2004. “Felony Sentences in State Courts, 2002.” Washington,
DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/fssc02.pdf
Durose, Matthew and Patrick A. Langan. 2007. “Felony Sentences in State Courts, 2004.” Washington,
DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/fssc04.pdf
Durose, Matthew R., David J. Levin, and Patrick A. Langan. 2001. “Felony Sentences in State Courts,

CEPR

Ex-offenders and the Labor Market

20

1998.” Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/fssc98.pdf
Emsellem, Maurice. 2010. “Collateral Consequences of Criminal Convictions: Barriers to Reentry for the
Formerly Incarcerated.” Hearing before the House Committee on the Judiciary, 111th Congress.
Ewert, Stephanie, Becky Pettit, and Bryan Sykes. 2010. “The Degree of Disadvantage: Incarceration and
Racial Inequality in Education.” Unpublished manuscript, University of Washington.
Finn, R.H. and Patricia A. Fontaine. 1985. “The Association between Selected Characteristics and
Perceived Employability of Offenders.” Criminal Justice and Behavior, vol. 12, no. 3, pp. 353-65.
Freeman, Richard B. 1991. “Crime and the Employment of Disadvantaged Youths.” NBER Working
Paper No. 3875, Cambridge, MA: National Bureau of Economic Research.
Geller, Amanda, Irwin Garfinkel, and Bruce Western. 2006. “The Effects of Incarceration on
Employment and Wages: An Analysis of the Fragile Families Survey.” Center for Research on Child
Wellbeing, Working Paper # 2006-01-FF. http://crcw.princeton.edu/workingpapers/WP06-01-FF.pdf
Gilliard, Darrell K. and Allen J. Beck. 1998. “Prisoners in 1997.” Washington, DC: U.S. Department of
Justice, Bureau of Justice Statistics. http://bjs.ojp.usdoj.gov/content/pub/pdf/p97.pdf
Glaze, Lauren E. and Thomas P. Bonczar. 2009. “Probation and Parole in the United States, 2008.”
Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/ppus08.pdf
Glaze, Lauren E. and Lauren M. Maruschak. 2008. “Parents in Prison and Their Minor Children.”
Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/pptmc.pdf
Grogger, Jeffrey. 1992. “Arrests, Persistent Youth Joblessness, and Black/White Employment
Differentials.” Review of Economics and Statistics, vol. 74, no. 1, pp. 100-06.
Harlow, Caroline Wolf. 2003. “Education and Correctional Populations.” Washington, DC: U.S.
Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/ecp.pdf
Harris, Casey T., Darrell Steffensmeier, Jeffrey T. Ulmer, and Noah Painter-Davis. 2009. “Are Blacks and
Hispanics Disproportionately Incarcerated Relative to Their Arrests? Racial and Ethnic
Disproportionality Between Arrest and Incarceration.” Race and Social Problems, vol. 1, no. 4, pp. 187199. http://www.springerlink.com/content/807350w4104g7151/fulltext.pdf
Harrison, Paige M. and Allen J. Beck. 2001. “Prisoners in 2000.” Washington, DC: U.S. Department of
Justice, Bureau of Justice Statistics. http://bjs.ojp.usdoj.gov/content/pub/pdf/p00.pdf
Harrison, Paige M. and Allen J. Beck. 2002. “Prisoners in 2001.” Washington, DC: U.S. Department of
Justice, Bureau of Justice Statistics. http://bjs.ojp.usdoj.gov/content/pub/pdf/p01.pdf
Harrison, Paige M. and Allen J. Beck. 2003. “Prisoners in 2002.” Washington, DC: U.S. Department of
Justice, Bureau of Justice Statistics. http://bjs.ojp.usdoj.gov/content/pub/pdf/p02.pdf
Harrison, Paige M. and Allen J. Beck. 2004. “Prisoners in 2003.” Washington, DC: U.S. Department of
Justice, Bureau of Justice Statistics. http://bjs.ojp.usdoj.gov/content/pub/pdf/p03.pdf
Harrison, Paige M. and Allen J. Beck. 2005. “Prisoners in 2004.” Washington, DC: U.S. Department of
Justice, Bureau of Justice Statistics. http://bjs.ojp.usdoj.gov/content/pub/pdf/p04.pdf
Harrison, Paige M. and Allen J. Beck. 2006. “Prisoners in 2005.” Washington, DC: U.S. Department of
Justice, Bureau of Justice Statistics. http://bjs.ojp.usdoj.gov/content/pub/pdf/p05.pdf
Holzer, Harry. 1996. What Employers Want: Job Prospects for Less-Educated Workers. New York: Russell Sage
Foundation.
Holzer, Harry J. 2007. “Collateral Costs: The Effects of Incarceration on the Employment and Earnings
of Young Workers.” IZA Discussion Paper No. 3118. Bonn: Institute for Labor Studies.
Holzer, Harry J., Paul Offner, and Elaine Sorensen. 2005. “Declining Employment among Young Black
Less-Educated Men: The Role of Incarceration and Child Support.” Journal of Policy Analysis and
Management, vol. 24, no. 2, pp. 329-350. Spring.
Holzer, Harry J., Steven Raphael, and Michael Stoll. 2004. “Will Employers Hire Former Offenders?

CEPR

Ex-offenders and the Labor Market

21

Employer Preference, Background Checks and their Determinants,” in Mary Pattillo, David Weiman,
and Bruce Western (eds.), Imprisoning America: The Social Effects of Mass Incarceration. New York: Russell
Sage Foundation.
Holzer, Harry J., Steven Raphael, and Michael Stoll. 2006. “Perceived Criminality, Criminal Background
Checks and the Racial Hiring Practices of Employers.” Journal of Law and Economics, vol. 49, pp. 45180.
Holzer, Harry J., Steven Raphael, and Michael Stoll. 2007. “The Effect of an Applicant’s Criminal
History on Employer Hiring Decisions and Screening Practices: Evidence from Los Angeles,” in
Shawn Bushway, Michael A. Stoll, and David F. Weiman (eds.), Barriers to Reentry? The Labor Market for
Released Prisoners in Post-Industrial America. New York: Russell Sage Foundation.
Irwin, John, Vincent Schiraldi, and Jason Ziedenberg. 1999. “America’s One Million Nonviolent
Prisoners.” Washington, DC: Justice Policy Institute.
http://www.justicepolicy.org/images/upload/99-03_REP_OneMillionNonviolentPrisoners_AC.pdf
Kling, Jeffrey. 2006. “Incarceration Length, Employment, and Earnings.” American Economic Review, vol.
96, no. 3, pp. 863-76.
Langan, Patrick A. 1996. “Felony Sentences in the United States, 1992.” Washington, DC: U.S.
Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/fsus92.pdf
Langan, Patrick A. and Jodi M. Brown. 1997. “Felony Sentences in State Courts, 1994.” Washington, DC:
U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/fssc94.pdf
Langan, Patrick A. and Helen A. Graziadei. 1995. “Felony Sentences in State Courts, 1992.” Washington,
DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/Felsent.pdf
Langan, Patrick A. and David J. Levin. 2002. “Recidivism of Prisoners Released in 1994.” Washington,
DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/rpr94.pdf
Mincy, R. B. and E. J. Sorensen. 1998. “Deadbeats and Turnips in Child Support Reform.” Journal of
Policy Analysis and Management vol. 17 no. 1, pp. 44-51.
Needels, Karen. 1996. “Go Directly to Jail and Do Not Collect? A Long-Term Study of Recidivism,
Employment, and Earnings Patterns among Prison Releasees.” Journal of Research in Crime and
Delinquency, vol. 33, no. 4, pp. 471-96.
Pager, Devah. 2003. “The Mark of a Criminal Record.” American Journal of Sociology, vol. 108, no. 5, pp.
937-75.
Pager, Devah. 2007. Marked: Race, Crime, and Finding Work in an Era of Mass Incarceration. Chicago:
University of Chicago Press.
Pettit, Becky and Christopher Lyons. 2007. “Status and the Stigma of Incarceration: The Labor-Market
Effects of Incarceration, by Race, Class, and Criminal Involvement,” in Shawn Bushway, Michael A.
Stoll, and David F. Weiman (eds.), Barriers to Reentry? The Labor Market for Released Prisoners in PostIndustrial America. New York: Russell Sage Foundation.
Pettit, Becky and Bruce Western. 2004. “Mass Imprisonment and the Life Course: Race and Class
Inequality in U.S. Incarceration.” American Sociological Review, vol. 69, pp. 151-69.
Pew Charitable Trusts. 2010. “Collateral Costs: Incarceration’s Effect on Economic Mobility.”
Washington, DC: The Pew Charitable Trusts.
Public Safety Performance Project. 2007. “Public Safety, Public Spending: Forecasting America’s
Prison Population, 2007-2011.” Washington, DC: Pew Center on the States, Pew Charitable Trusts.
http://www.pewcenteronthestates.org/uploadedFiles/Public%20Safety%20Public%20Spending.pdf
Public Safety Performance Project. 2008. “One in 100: Behind Bars in American 2008.”
Washington, DC: Pew Center on the States, Pew Charitable Trusts.
http://www.pewcenteronthestates.org/uploadedFiles/8015PCTS_Prison08_FINAL_2-1-1_FORWEB.pdf

CEPR

Ex-offenders and the Labor Market

22

Public Safety Performance Project. 2009. “One in 31: The Long Reach of American Corrections.”
Washington, DC: Pew Center on the States, Pew Charitable Trusts.
http://www.pewcenteronthestates.org/uploadedFiles/PSPP_1in31_report_FINAL_WEB_3-26-09.pdf
Raphael, Steven. 2007. “Early Incarceration Spells and the Transition to Adulthood,” in Danziger,
Sheldon and Cecilia Rouse (eds.), The Price of Independence. New York: Russell Sage Foundation.
Rosenmerkel, Sean, Matthew Durose, and Donald Farole, Jr. 2009. “Felony Sentences in State Courts,
2006 - Statistical Tables.” Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/fssc06st.pdf
Sabol, William J. 2007. “Local Labor-Market Conditions and Post-Prison Employment Experiences of
Offenders Released from Ohio State Prisons,” in Shawn Bushway, Michael A. Stoll, and David F.
Weiman (eds.), Barriers to Reentry? The Labor Market for Released Prisoners in Post-Industrial America. New
York: Russell Sage Foundation.
Sabol, William J., Heather Couture, and Paige M. Harrison. 2007. “Prisoners in 2006.” Washington, DC:
U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/p06.pdf
Sabol, William J., Heather C. West, and Matthew Cooper. 2009. “Prisoners in 2008.” Washington, DC:
U.S. Department of Justice, Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/content/pub/pdf/p08.pdf
Schmitt, John, Kris Warner, and Sarika Gupta. 2010. “The High Budgetary Cost of Incarceration.”
Washington, DC: Center for Economic and Policy Research.
http://www.cepr.net/documents/publications/incarceration-2010-06.pdf
Schwartz, Richard D. and Jerome Skolnick. 1962. “Two Studies of Legal Stigma.” Social Problems, vol. 10,
pp. 133-142.
Stemen, Don. 2007. “Reconsidering Incarceration: New Directions for Reducing Crime.” New
York: Vera Institute of Justice. http://www.vera.org/download?file=407/veraincarc_vFW2.pdf
Uggen, Christopher, Jeff Manza, and Melissa Thompson. 2006. “Citizenship and Reintegration: The
Socioeconomic, Familial, and Civic Lives of Criminal Offenders.” Annals of the American Academy of
Political and Social Science, 605 (May), pp. 281-310.
U.S. Bureau of the Census. 1961. “U.S. Census of the Population: 1960, Vol 2: Subjects Reports, Part 8A:
Inmates of Institutions.” Washington, DC: U.S. Government Printing Office.
http://www2.census.gov/prod2/decennial/documents/41927948v2p8a-8c_ch02.pdf
U.S. Bureau of the Census. 1972. “U.S. Census of the Population: 1970, Vol 2: Subject Reports, Persons
in Institutions and Other Group Quarters.” Washington, DC: U.S. Government Printing Office.
http://www2.census.gov/prod2/decennial/documents/42045398v2p4d4ech5.pdf
Waldfogel, Joel. 1994. “The Effect of Criminal Conviction on Income and the Trust ‘Reposed in the
Workmen.’“ Journal of Human Resources, vol. 29, pp. 62-81.
West, Heather C. and William J. Sabol. 2008. “Prisoners in 2007.” Washington, DC: U.S. Department of
Justice, Bureau of Justice Statistics. http://bjs.ojp.usdoj.gov/content/pub/pdf/p07.pdf
Western, Bruce. 2002. “The Impact of Incarceration on Wage Mobility and Inequality.” American
Sociological Review, vol. 67, pp. 526-546.
Western, Bruce. 2006. Punishment and Inequality in America. New York: The Russell Sage Foundation.
Western, Bruce. 2008. “From Prison to Work: A Proposal for a National Prisoner Reentry Program.”
Washington, DC: The Hamilton Project / The Brookings Institution.
Western, Bruce. 2008. “Reentry,” Boston Review (July-August),
http://bostonreview.net/BR33.4/western.php.
Western, Bruce and Katherine Beckett. 1999. “How Unregulated Is the U.S. Labor Market? The Penal
System as a Labor Market Institution.” American Journal of Sociology, vol. 104, no. 4, pp. 1030-60.
Western, Bruce and Becky Pettit. 2010. “Incarceration & Social Inequality.” Daedalus: Journal of the
American Academy of Arts and Sciences vol. 139 no. 3, pp. 8-19. Summer.

 

 

Disciplinary Self-Help Litigation Manual - Side
Advertise here
The Habeas Citebook: Prosecutorial Misconduct Side