1

Criminal history as an employment barrier for female

TANF recipients

Pamela C. Ovwigho, Ph.D.

Valerie Head, M.P.P.

Correne Saunders, M.P.P.

Catherine E. Born, Ph.D.

Paper presented at the 47th annual workshop of the National Association for Welfare Research and Statistics

This research was funded by the Maryland Department of Human Resources and the Office of the Assistant Secretary for Planning and Evaluation, US Department of Health and Human Services

Wednesday, August 22, 2007

2

background

• Research demonstrates the prevalence of employment barriers among TANF recipients, yet there remains little research regarding criminal history.

• A criminal record appears to be a barrier to obtaining and maintaining employment. However, little is known regarding its impact on the employment outcomes among disadvantaged workers in general. Even less is known about its impact on female workers.

• Compared to their male counterparts, women with a criminal history face more difficult economic circumstances. At the time of their arrest:

• 30% reported receiving welfare assistance• 3 out of 5 women were unemployed • 37% reported incomes of less than $600• More likely to be custodial parents

Source: Greenfeld, L.A., & Snell, T. L. (December 1999). Women offenders (Bureau of Justice Statistics No. NCJ 175688). Washington D.C.: U.S. Department of Justice.

3

Research Questions

1. What proportion of the Maryland TCA caseload consists of female caseheads who report having a criminal background?

Among those female caseheads who report a criminal history:

2. What are their demographic and case characteristics? How do they compare to other female caseheads?

3. What are their welfare participation patterns pre- and post- survey? How do they compare to other female caseheads?

4. What are their employment experiences pre- and post-survey? How do they compare to other female caseheads?

5. What are their reported barriers to employment? How do employment barriers relate to their employment experiences? How do they compare to other female caseheads?

4

Methods: Sample

• Random sample of single adults with children who received a TANF grant in Maryland in June 2002 (n=1,146).

• For this study, we include only those females who responded to a telephone survey between August and October 2002 (n=790).

• Weighted to represent true proportions of Baltimore City and non-

Baltimore City cases in the June 2002 caseload.

5

• Groups were based on responses to the following survey question:

• [Do you] have a past criminal record?

• Weighted responses yielded the following groups:

• No Report (n=689)

• Report (n=104)

• Total (n=791)

• Chi-square and ANOVA tests were used to determine differences between the two groups.

Methods: design

6

Methods: design

7

• Maryland TANF Caseload Survey, sponsored by ASPE

Included questions re: the prevalence and degree of various employment barriers over the previous year (June 2001 – June 2002)

Conducted for FWRTG through CATI by Mathematica Policy Research, Inc. (MPR)

• Maryland State Administrative Systems

Client Information System Maryland Unemployment Insurance System

Methods: data sources

8

Research question 1

What proportion of the Maryland TCA caseload consists of female caseheads who report having

a criminal background?

9

Prevalence of criminal history

Report 13 .1%

(n=104)

No Report 86 .9%

(n=687)

For a sizeable minority of welfare recipients, a criminal background may pose a barrier to obtaining and maintaining employment.

10

Research question 2

Among those who report a criminal history, what are their demographic and case characteristics?

How do their characteristics compare to other female caseheads?

11

Demographics

• In terms of demographics, there are no significant differences between non-reporters and reporters.

• On average, reporters tend to be older than non-reporters (31.7 years versus 30.1 years, respectively).

• Reporters are less likely to be African American.

12

Case characteristics

Children Less Than Age 18 Living Outside the Home***

Reporters are three times more likely to have children living outside of the home than non-reporters.

20 .8%(n =22)

7 .2% (n =50)

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

N o R eport R eport

*p<0.05, **p<0.01, ***p<0.001

13

Case characteristics

Number of Children Living Outside the Home***

Among reporters with children outside the home, more than half have 3 or more children living outside the home.

11.5

25.7

62.8

52.2

16.4

31.4

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

1 2 3 or m ore

N o R eport R eport

No Report

Report

Mean 1.8 2.5

Median 1.0 3.0

Standard Deviation

1.8 2.5

Range 1-8 1-4

*p<0.05, **p<0.01, ***p<0.001

14

Research question 3

Among those who report a criminal history, what are their pre- and post-survey welfare

participation patterns?

How do their patterns compare to other female caseheads?

15

Welfare receipt history

Pre-Survey TANF Receipt

*p<0.05, **p<0.01, ***p<0.001

• Reporters are significantly more likely to

have a longer welfare receipt history

prior to the survey (i.e. >36 months).

• Reporters average 7.1 more months of

time-limited TANF than non-reporters

prior to survey.

33.5

26.4

0

10

20

30

40

N o R ep o rt R ep o rt

Mean Number of Months of Receipt Counted Towards the 60-Month Limit**

16

Welfare follow-up

30.138.6

53.3

93.6

33.0

53.9

66.1

96.1

0.0%

20.0%

40.0%

60.0%

80.0%

100.0%

1st Fo llow -U p Year

2nd Fo llow -U p Year**

3rd Fo llow -U p Year**

4th Fo llow -U p Year

N o R eport R eport

No Report

Report

Mean ** 15.9 18.7

Median 14.0 19.0

Standard Deviation

12.5 12.6

Range 0-48 0-48

*p<0.05, **p<0.01, ***p<0.001

Number of Months of Receipt During Entire 4-Year Follow-Up

% Received TANF During Follow-Up

Post-Survey TANF Receipt

Reporters are more likely to receive TANF for significantly longer periods during the follow-up period.

17

Research question 4

Among those who report a criminal history, what are their pre- and post-survey UI-employment

experiences?

How do their experiences compare to other female caseheads?

18

Employment experiences

• Reporters were more likely to have been employed in Maryland UI-covered jobs than non-reporters. 2 7 .0

5 7 .7

7 3 .8

3 4 .1

7 5 .76 1 .5

0 .0%

20.0%

40.0%

60.0%

80.0%

8 Q u arte rsPr io r toSu rv e y

4 Q u arte rsPr io r toSu rv e y

Q u arte r o fSu rv e y

Pre-Survey Employment

% Employed

Mean Number of Unique Employers

• On average, reporters held significantly more UI-Covered jobs than non-reporters.

2.8

1.9

3.4

2.3

0.0

1.0

2.0

3.0

4.0

8 Quarters P rio r toS urvey*

4 Quarters P rio r toS urvey*

N o R eport R eport

19

Employment experiences

Pre-Survey Earnings

Though reporters were more likely to be employed prior to survey, they typically earned less in terms of median total earnings.

*p<0.05, **p<0.01, ***p<0.001

$877.45

$2,413.28

$5,168.78

$677.98

$2,111.62

$3,572.75

$0.00

$2,000.00

$4,000.00

$6,000.00

8 Q u arte rs P r io r toS u rvey

4 Q u arte rs P r io r toS u rvey

Q u arte r o f S u rvey

N o R e port R e port

Median Total Earnings Prior to Survey

20

Employment FOLLOW-UP

• Reporters are less likely to be UI-employed than non-reporters during the follow-up period.

5 7 .05 7 .35 8 .55 5 .6 5 5 .45 2 .35 2 .7 5 1 .9

0.0%

20.0%

40.0%

60.0%

80.0%

1st YearP ost

S urvey

2nd YearP ost

S urvey

3rd YearP ost

S urvey

4th YearP ost

S urvey

Post-Survey Employment

% Employed

Mean Number of Unique Employers

• On average, reporters hold significantly more UI-covered jobs than non-reporters during the follow-up period.

3 .94 .3

1 .0

2 .0

3 .0

4 .0

5 .0

No Rep o rt Rep o rt

21

Employment Follow-Up

Post-Survey Earnings

Reporters typically earn less than non-reporters in terms of median total earnings. However, reporters’ earnings continue to increase, while non-reporters’ earnings

plateau toward the end of the follow-up period.

$6,692.58$6,855.00$6,375.44

$3,615.39

$7,188.87

$4,395.92$3,808.49

$1,552.06

$0.00

$2,000.00

$4,000.00

$6,000.00

$8,000.00

1st Year P ost S urvey 2nd Year P ost S urvey 3rd Year P ost S urvey 4th Year P ost S urvey

N o R e port R e port

Median Total Earnings

22

Among those who report a criminal history, what are their reported barriers to employment

and how do these barriers relate to subsequent employment experiences?

How do the employment experiences of reporters compare to other female caseheads?

Research question 5

23

24.720.4

40.135.6

26.4

54.2

0 .0%

20 .0%

40 .0%

60 .0%

80 .0%

100 .0%

L ess T h an Hig hS ch o o l/G E D **

No Wo rk E xp er ien ce P erfo rm ed L ess T h an 4C o m m o n Jo b T asks**

No Rep o rt

Rep o rt

Employment Barriers

Human Capital Deficits

Reporters are significantly less likely to have at least a high school education than non-reporters. Additionally, reporters are

significantly less likely to perform basic job skills.

*p<0.05, **p<0.01, ***p<0.001

24

5.513.9

13.64.7

28.632.0

20.7

2.7

20.831.8

7.2

32.7

44.5

19.1

0 .0%

20 .0%

40 .0%

60 .0%

80 .0%

100 .0%

P h ys ica lHea lth

P ro b lem

C h ild o rF am ily

M em b er w ithHea lth o rS p ec ia lNeed ***

M en ta l Hea lthP ro b lem

C h em ica lD ep en d en ce

S evereP h ys ica lD o m esticV io len ce*

P o ss ib leP resen ce o f a

L earn in gD isab ility

P reg n an t

No Rep o rt

Rep o rt

Employment Barriers

Personal and Family Challenges

Reports are significantly more likely to have a child or family member with a health or special need or experience severe physical

domestic violence.

*p<0.05, **p<0.01, ***p<0.001

25

Employment Barriers

Logistical and Situational Challenges

Reporters are significantly more likely to report transportation, housing, discrimination, and neighborhood challenges.

16.4

54.6

18.3

40.8

24.7 27.3

67.9

34.943.8

34.7

0 .0%

20 .0%

40 .0%

60 .0%

80 .0%

100 .0%

T ran sp o rta tio n * C h ild C are Un stab leHo u s in g ***

P erce ivedP ro b lem

Neig h b o rh o o dC h arac te r is tics*

D iscr im in a tio n **

No Rep o rt

Rep o rt

*p<0.05, **p<0.01, ***p<0.001

26

Employment Barriers

Total Employment Liabilities***

On average, reporters have significantly more employment liabilities than non-reporters. Half of reporters have more than 4 liabilities.

4.3

3.2

0

1

2

3

4

5

No Report Report

No Report

Report

Mean*** 3.2 4.3

Median 3.0 4.0

Range 0-10 0-12

Mean Number of Total Employment Liabilities

*p<0.05, **p<0.01, ***p<0.001

27

Employment follow-up

• When these reported employment barriers are considered, employment patterns are notably different between the two groups.

• Reporters who possess • 2 or more human capital deficits

are more likely to be employed than non-reporters who too face those challenges.

• Reporters who face • 2 or more logistical or situational challenges • 2 or more personal and family challenges

are less likely to be employed than non-reporters who also face those challenges.

• However, in all cases, reporters earn less in terms of median quarterly earnings.

Reported Barriers & Employment

28

Employment follow-up

Reported Barriers & Employment

When reported employment barriers are considered, employment patterns are notably different between the reporters and non-reporters

67.8%71.3%

81.9% 78.4%68.7% 66.9%

74.7% 74.4%

0.0%

20.0%

40.0%

60.0%

80.0%

100.0%

2 or More H um anC ap ita l D efic its

2 o r More P ersonal &Fam ily C hallenges

2 or More Log istical o rS ituational C hallenges

2 or More To talL iab ilities

N o R e port R e port

$1,577.40

$1,236.00

$1,582.65

$1,218.44

$1,933.86

$1,578.77

$1,769.00

$1,633.76

% Employed & Median Quarterly Earnings During Entire Follow-Up Period

29

Conclusions

• Only slight demographic differences between reporters and non-reporters.

• Reporters are long-term TANF recipients.

• Work effort among reporters is high, but there is some indication that they lack job stability.

• Reporters are more likely to have other barriers to employment, and to be more disadvantaged overall.

30

limitations

• UI-employment data is limited to a select group of Maryland employers. To the extent that reporters are more or less likely to work in these types of jobs, employment data will be skewed.

• Survey data is self-reported. It is unknown whether some with a criminal background choose not to report it.

31

implications

• There is a need to address criminal record status as part of regular client assessment within TANF, specifically:

• As an indicator of other potential barriers.

• In order to refer to appropriate jobs and/or re-entry services, especially those services rendering job skill development.

• Further research is needed regarding:

• The number of hours worked within a quarter, and job stability overall.

• Family and child well-being, related to the finding which indicates a disproportionate number of children living outside the home among reporters.

32

THANK YOU

The full report will be available shortly on our website:

http://www.familywelfare.umaryland.edu