Order Code RL30882
CRS Report for Congress
Received through the CRS Web
Welfare Reform Research: What Do We Know
About Those Who Leave Welfare?
March 13, 2001
Christine Devere
Analyst in Social Legislation
Domestic Social Policy Division
Congressional Research Service ˜ The Library of Congress

---

Welfare Reform Research: What Do We Know About
Those Who Leave Welfare?
Summary
Cash welfare caseloads have declined 57% since peaking in 1994 at 5.1 million
cases, with the largest declines occurring since the passage of the Personal
Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA).
PRWORA ended the entitlement program of Aid to Families with Dependent Children
(AFDC), and replaced it with a block grant program to the states, the Temporary
Assistance for Needy Families (TANF) program. Unlike AFDC, under TANF most
adults are required to work as a condition of receiving benefits, and assistance from
federal funds is time-limited.
States have initiated and completed studies of former welfare recipients to
understand how those who exit welfare are faring. These leaver studies use a
combination of administrative data and survey data, and the results differ by the type
of data used. The methods used by these studies also vary, as do their outcomes.
Employment and other income-related closures are the most common reasons
a recipient leaves cash assistance. To date, no information is available on those who
reach the federal 5-year lifetime limit as the first recipients will begin reaching this
time limit later in 2001. Some recipients have reached shorter state-imposed time
limits, but these recipients do not appear to face increased difficulties or hardship.
Among the studies reviewed in this report, employment rates range considerably,
but the majority report employment rates between 55% and 64% within 3 months of
exit or at the time of the survey. A larger percentage, between 63% and 91% of
leavers, have been employed (ever employed) for some period of time since exit. The
average hourly wage reported among welfare leavers ranged from $5.50 to $8.80 per
hour.
The leaver studies indicate that while the majority of welfare leavers are
employed, most remain poor and rely on other types of assistance such as health
insurance, food stamps, and child care to supplement their wages. However,
declining participation in these programs has raised concerns. The leaver studies
illustrate that some welfare leavers did not think they were eligible for food stamps
or Medicaid, while others indicated that they did not need these services or that it was
too much hassle to receive them. Receipt of child care subsidies is somewhat low,
although some indicate a “lack of need” for these subsidies. Less than half of leavers
also appear to be experiencing “hardships” such as difficulties paying bills, difficulties
acquiring medical care, and experiencing a time with no way to buy food.
Employment is the reason the majority of recipients leave welfare, and lack or
loss of employment is the most common reason individuals return to welfare. Among
leavers who remained off welfare for at least 2 months, between 18% and 35% had
returned for a period of time since exit. Although lack of employment is the most
common reason individuals return to welfare, there is some evidence that those who
leave for income-related reasons are less likely to return than those who leave welfare
for failing to comply with program requirements.

---

Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Overall Well-Being of Welfare Leavers . . . . . . . . . . . . . . . . . . . . . . . 3
Ending Dependence on Government Benefits . . . . . . . . . . . . . . . . . . . 4
Issues in Monitoring Welfare Leavers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Different exit cohorts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Studies funded by the Department of Health and Human Services
(HHS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Reason for exit from cash welfare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Closures due to time limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Closures due to sanctions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Employment-Related Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Employment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Earnings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Occupations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Finding employment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Employment-related benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Results by time of exit from cash welfare . . . . . . . . . . . . . . . . . . . . . 12
Results by reason for cash welfare exit . . . . . . . . . . . . . . . . . . . . . . . 14
Employment Retention and Advancement . . . . . . . . . . . . . . . . . . . . 14
Barriers to employment and reasons for unemployment . . . . . . . . . . 15
Overall Well-Being of Welfare Leavers . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Income/Poverty Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Measures of Hardship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Hardships before and after exit from cash welfare . . . . . . . . . . . . . . . 17
Child Welfare Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Receipt of Other Types of Assistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Health Insurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Child Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Food Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Child Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Earned Income Tax Credit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Returns to Welfare Among Welfare Leavers . . . . . . . . . . . . . . . . . . . . . . 23
List of Figures
Figure 1. Employment Rate: Number of Study Cohorts Within Each Range . . 11
List of Tables
Table 1. Earnings and Employment Outcomes Among Leaver Studies That
Examine Leavers Based on When They Exit Cash Welfare . . . . . . . . . . . . 13
Table 2. Measures of Hardships as Reported in State Leaver Studies . . . . . . . 17
Appendix Table A-1. Administrative Data: Employment and Earnings of
Welfare Leavers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

---

Appendix Table A-2. Survey Data: Employment and Earnings of Welfare
Leavers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Appendix Table B-1. Selected Outcomes from Welfare Leaver Studies . . . . . . 30

---

Welfare Reform Research: What Do We
Know About Those Who Leave Welfare?
This report presents information that is part of a larger review of welfare reform
research, which has been published in the 2000 Edition of the House Committee
on Ways and Means Green Book. That review, entitled “Appendix L: Monitoring
the Effects of Pre- and Post-TANF Welfare Reform Initiatives,” includes three
complementary analyses. One component is an analysis of income, work and
welfare trends among single female-headed families with children, using data
from the Current Population Survey and the Consumer Expenditure Survey (an
update of the analysis from the Current Population Survey is also published in
CRS Report RL30797, Trends in Welfare, Work and the Economic Well-Being of
Female-Headed Families with Children). Another is a review of state impact
evaluations of welfare reform initiatives (also published in CRS Report RL30724,
Welfare Reform Research: What Have We Learned Since the Family Support Act
of 1988?). The third component, a review of state studies of former welfare
recipients (state “leaver” studies), is presented in an updated format in this CRS
report, adding studies available between January 2000 and January 2001.
Introduction
There is considerable interest in the circumstances of families who have left cash
welfare (the “leavers”). Since the peak of 5.1 million welfare cases in 1994, the
caseload has fallen 57% with the fastest declines occurring since the passage of the
Personal Responsibility and Work Opportunity Reconciliation Act of 1996
(PRWORA, herein referred to as the welfare reform law).1 The welfare reform law
ended the entitlement program of Aid to Families with Dependent Children (AFDC)
and replaced it with a block grant program to the states, the Temporary Assistance
for Needy Families (TANF) program. Unlike AFDC, under TANF most adults are
required to work as a condition of receiving benefits, and assistance is time-limited.
States also have broad flexibility (within minimal federal guidelines) to design their
state programs within the goals of ending dependence on welfare, promoting
marriage, and reducing nonmarital births.2
The unexpected caseload decline has led many to explore reasons for this
dramatic drop. The number of cases receiving cash welfare expanded greatly during
1 The caseload decline reflects not only a growing number of individuals who have left
cash welfare, it also reflects a declining number of individuals who enroll in welfare.
2 For more information on the state TANF programs, see CRS Report RL30695,
Welfare Reform: State Programs of Temporary Assistance for Needy Families (TANF), by
Emilie Stoltzfus, Vee Burke, and Gene Falk.

---

the late 1980s until reaching a peak in 1994. Contributing to the increase in the
caseload was an increase in single mothers, but the decline in the caseload has not
been accompanied by a decline in the number of single mothers (which has remained
between 9.4 million and 10 million since 1994).3 While most researchers agree that
the caseload decline has been due to a combination of the economic expansion,
welfare reform, increased earnings supplements provided through the Earned Income
Tax Credit (EITC), and the expansion of other services (such as child care), there is
no broad agreement on the weight of each factor.4
While the debate continues on reasons for the dramatic caseload decline,
attention has turned to examining the overall well-being of those who exit welfare.
A growing number of states are conducting studies to examine the characteristics and
circumstances of former welfare recipients (“leaver” studies). Part of this interest is
sparked by new welfare policies seeking to end dependence of needy parents on
government benefits. Under TANF, the nature of cash welfare has changed. Welfare
recipients are subject to time limits on assistance, which include the federal lifetime
limit of 5 years, and in many instances, shorter state-imposed time limits. Recipients
are also subject to work requirements and increased work sanctions. To date, a
relatively small share of the national caseload has been removed from the rolls because
of time limits or reported as sanctioned off the rolls. The majority of former welfare
recipients attribute their exit from cash welfare to employment-related reasons, as was
the case under AFDC.
This report reviews state leaver studies that focus on former recipients who
exited from TANF programs or from programs with requirements similar to new
TANF initiatives (such as sanctions or time limits). State leaver studies available as
of January 2001 are included. Appendix Table B-1 presents a state-by-state table of
the studies reviewed in this report. This report begins with a brief overview, outlines
issues that states face in examining welfare leavers, and explores how information
from the leaver studies addresses the following questions:
! Why are people leaving welfare? What is the reason for case closing?
! Where are former welfare recipients going? To answer this question, states
have examined information on earnings, employment rates, length of
employment, and occupations. Additionally, among leavers who are
unemployed, the studies examine barriers to employment and reasons for
unemployment.
! How are former welfare recipients doing? Some studies examine family
income of welfare leavers. Outcomes include earnings and unearned income
of the welfare leaver as well as other household members (e.g., spouses).
These studies also look at other types of assistance such as health care, child
3 See CRS Report RL30797, Trends in Welfare, Work, and Economic Well-Being of
Female-Headed Families with Children, by Thomas Patrick Gabe.
4 See for example, The Effects of Welfare Policy and the Economic Expansion on
Welfare Caseloads: An Update, August 3, 1999, a report by the Council of Economic
Advisors; and Ziliak, James P., David N. Figlio, Elizabeth E. Davis, & Laura S. Connolly.
(2000). Accounting for the Decline in AFDC Caseloads: Welfare Reform or the Economy?
Journal of Human Resources, 35(3), p.570-586.

---

care, food assistance, child support, and the EITC. The studies also examine
subjective measures of hardship, such as difficulty paying bills, difficulty
acquiring housing, and difficulty acquiring health care. Finally, for a small
number of studies, information is available on measures of child welfare,
including removal of the child from home and substantiated instances of abuse
and neglect.
! Do former recipients return to welfare? And, why do they return to welfare?
The studies examine rates of return among those who leave welfare as well as
their reasons for return.
Overview
The large decline in welfare caseloads that has occurred since 1994 has been
welfare reform’s biggest surprise. States have initiated and completed studies of
welfare leavers to better understand how those who exit welfare are faring. These
leaver studies have used administrative data or survey data (and in some instances,
both types of data), and the results differ by the type of data used. The studies that
use surveys to track welfare leavers have been able to gather added information of
interest to state administrators and policymakers than what administrative data alone
provide. The outcomes collected by these survey-based studies vary, but they all
illustrate the difficulty in finding individuals once they leave welfare.
Key points from this review of leaver studies are:
! Among those who leave welfare, employment and income-related closures are
the most common reasons for exit from cash assistance (their “reason for case
closing”).
! Employment rates range considerably, but the majority of studies report
employment rates between 55% and 64% within 3 months of exit or at the time
of the survey. Among individuals ever employed since exit from welfare, the
rates increase to between 63% and 91% of respondents.
! Although the majority of welfare leavers are employed, based on wages alone,
welfare leavers in a number of states remain poor. These leavers continue to
rely on programs such as food stamps and Medicaid, although declining
participation rates in these programs have been noted.
! Between 18% and 35% of leavers have returned to welfare for a period of time
since exit, and lack or loss of employment is the most common reason cited for
individual return to welfare.
Overall Well-Being of Welfare Leavers. Many states have gathered
information beyond income and wages to more comprehensively examine the well-
being of welfare leavers, including measures such as income and wages of other
household members, support from other government programs (such as food stamps
and Medicaid), and “hardships” experienced by those who leave welfare (such as
difficulty paying bills or accessing health care). The leaver studies indicate that while
the majority of leavers are employed, most remain poor and rely on other types of
assistance such as health care, child care, and food stamps to supplement their wages.
However, declining participation in these programs has raised concerns. In a number
of studies, respondents indicated that they did not think they were eligible for
Medicaid and food stamps. However, some also indicated they did not need these

---

services or that it was too much hassle to receive them. A small percentage of welfare
leavers are receiving child care subsidies, while a number of respondents indicate no
need for these subsidies. Child support also appears to be an important source of
income among welfare leavers. While a large number of individuals are leaving
welfare for work, less than half of welfare leavers indicate that they have filed for or
received the EITC.
State leaver studies indicate that less than half of leavers appear to be
experiencing hardships after exit from welfare. Whether these problems have
increased or decreased since exit varies by state and by the outcome measured. A few
states have reported increases in instances where leavers experienced difficulties
paying bills and/or arranging housing since exit, while other states have reported
decreases in these hardship measures since exit. However, problems acquiring
medical care and, on occasion, food have increased among leavers since cash welfare
exit.
Some contended that children would be harmed by welfare reform, but to date
little information has been collected on outcomes of children among families who
leave welfare. In a small number of studies, between 3% and 19% of children in these
states were reported to have left a former welfare home since cash welfare exit, but
little additional information was given to explain why these children left the home
(i.e., were they removed by Child Protective Services, did they go to live with their
father, etc.).
Ending Dependence on Government Benefits. A stated goal of TANF
is to end dependence on government benefits by promoting job preparation, work,
and marriage. While a number of welfare leavers are employed, most remain poor and
rely on food stamps and Medicaid to supplement their wages. Some argue that low
participation rates illustrate that leavers have replaced these government benefits with
other types of support, thereby ending their dependence on government benefits.
However, others argue that recipients are leaving welfare for low-wage work and
therefore remain in need of health care and food assistance. They believe that eligible
leavers are not participating in these programs and therefore, outreach to increase
enrollment should be the emphasis.5
The role of employment in enabling individuals to leave welfare also raises
concerns. Individuals who have left welfare for low-wage work may be especially
vulnerable to losing their employment in the case of an economic downturn. In this
case, it is unlikely that many would be eligible to receive benefits under
Unemployment Insurance (UI). A recent report by the U.S. General Accounting
Office (GAO) concluded that the role of UI as a safety net for low-wage workers is
5 See: Dion, M. and L. Pavetti,(2000). Access to and Participation in Medicaid and
the Food Stamp Program: A Review of the Recent Literature. Washington, DC;
Mathematica Policy Research, Inc. See also: Food Stamp Program: Various Factors Have
Led to Declining Participation. Report by the U.S. General Accounting Office (1999)
(RCED-99-185) and Medicaid Enrollment: Amid Declines, State Efforts to Ensure Coverage
After Welfare Reform Vary. Report by the U.S. General Accounting Office (1999) (HEH-99-
163).

---

limited. The report found that, nationwide, few states have adjusted their UI
programs to eliminate practices that may present difficulties to low-wage workers.
For example, voluntarily quitting a job for personal financial problems often
disqualifies claimants from UI benefits. GAO reports that if a worker currently
available for work quit his or her job because child care was temporarily unavailable,
the worker would not qualify for UI benefits in 32 states.6
Researchers have also explored the circumstances of welfare leavers in
comparison to those who have not recently received welfare but can be classified as
low-income. Many of the provisions of TANF not only encourage recipients to leave
welfare, they might also discourage enrollment in welfare in the first place. The result
is a lower cash welfare caseload, and perhaps a larger number of low-income working
parents. A report completed by the Urban Institute, which focused on an earlier
group of welfare leavers (those who stopped receiving benefits at some point between
1995 and 1997), found considerable similarities in employment and wages among
welfare leavers and low-income working mothers.7
Issues in Monitoring Welfare Leavers
The studies undertaken by the states to track those who left welfare are
fundamentally different from impact evaluations of state welfare reform initiatives.8
Impact evaluations are designed to measure the difference between two sets of
policies to determine how policy changes have affected reported outcomes. Studies
of welfare leavers are monitoring studies, which are not designed to determine the
impact, or effect, of a policy initiative.
Leaver studies track those who left welfare and provide information on why they
left welfare and how they are currently doing. These studies inform state
administrators and policymakers of the circumstances of former welfare recipients.
Because these studies have been initiated by the states, the information collected and
the outcomes reported vary considerably. These studies track different exit cohorts
of welfare leavers, they examine different study populations, and they follow leavers
for different periods of time.
Different exit cohorts. Prior to the welfare reform law of 1996, states were
experimenting with welfare policy initiatives intended to promote work and end
dependence on government benefits through waivers of AFDC rules. Many states
incorporated a number of these waivers into their TANF programs. Therefore,
6 See Unemployment Insurance: Role as Safety Net for Low-Wage Workers is Limited
(GAO-01-181). Report by the U.S. General Accounting Office to Congressional Requesters,
December 2000.
7 See Loprest, P. (1999). Families Who Left Welfare: Who are they and how are they
doing? Washington, DC; Urban Institute. Available at
[http://newfederalism.urban.org/pdf/discussion99-02.pdf]
8 For a review of state impact evaluations of welfare reform initiatives, see CRS Report
RL30724, Welfare Reform Research: What Have We Learned Since the Family Support Act
of 1988?, by Christine Devere, Gene Falk, and Vee Burke.

---

welfare reform has been occurring among the states for varying lengths of time. The
leaver studies examine different exit cohorts, as each state defines its study cohorts
by determining which group of former recipients to monitor. The leaver studies
reviewed in this report follow recipients who left welfare at different times from 1995
to 2000. This may affect how outcomes are interpreted.
Different study populations. It is important to understand that the
populations vary across these studies when interpreting reported outcomes. The
studies vary in how they define their leaver population. Examples include:
! While the vast majority of the caseload is comprised of single mothers, states
have chosen also to include two-parent families and child-only cases in
studying welfare leavers (although most exclude child-only cases).
! States have chosen to look at cases closed for different periods of time. That
is, some states define a leaver as an individual who left for any period of time,
while others define leavers as those who remain off welfare for at least 2
months.
! Some states have defined their leavers by their reason for case closing. This
includes following those who leave welfare for work, those who leave welfare
because of a time limit, or those who leave welfare because of a sanction.
! Finally, states have differed in the geographic location they choose to examine.
While some limit their study to only a few counties in the state, others examine
a large metropolitan area (such as Los Angeles). The majority examine a
group of leavers from across the state.
Different interview dates/follow-up periods. The leaver studies vary in
the length of time between the cash welfare exit and when the leaver is examined in
the study. Those that use administrative data follow individuals over a period of time,
reporting quarterly (3 months) data on employment, earnings, and participation in
other assistance programs (such as the Food Stamp program and Medicaid). The
survey studies use information at a point in time, and the length of time between exit
and the survey varies. The majority of the studies have monitored leavers over 12
months, with a few examining those off welfare for a longer period of time (up to 30
months).
Types of data used. State leaver studies have used administrative data or
survey data (and in some instances, both types of data) to monitor former welfare
recipients. One of the major challenges in following welfare leavers lies in tracking
these individuals once they leave cash assistance. Studies that use administrative data
generally track a much larger number of former recipients than surveys, but may miss
segments of the population. For example, most individuals who leave welfare for
employment appear in the state’s Unemployment Insurance (UI) wage files.
However, these UI files do not cover all types of employment (e.g., they do not
include any federal government employment), and UI files cannot track individuals
who move and/or work out-of state. W
9
hen using administrative data, evaluators are
9 Because of these concerns, the District of Columbia leaver study used data available
from the National Directory of New Hires (NDNH) to track employment and earnings of
(continued...)

---

also limited to the information that is collected. For example, the UI wage files
collect quarterly earnings information, but it is not possible to identify how many
hours of work have occurred over the quarter, or the average hourly wage.
To collect information on outcomes not available through administrative data
(such as the hourly wage rate), states have also conducted surveys of welfare leavers.
These surveys allow states to ask many different types of questions to gather
information that is of interest to state administrators and policy makers. However,
survey samples are typically much smaller than those used in administrative data, the
information may be more expensive to collect, and surveys often take more time and
resources. Efforts are made to minimize these costs by identifying a representative
sample of welfare leavers to survey, with the goal of generalizing these results to the
entire population of leavers in the state.
However, studies that use survey data have found that locating and monitoring
welfare leavers is difficult. Information collected through a survey represents the
situations of those who respond to the survey. If the individuals who respond to the
survey systematically differ from those who do not respond, the findings may not be
representative of the survey population, let alone the state’s entire population of
welfare leavers. Previous reviews of state leaver studies have not included results
from survey studies with low response rates, given the difficulty in generalizing these
results to a larger population.10
Studies funded by the Department of Health and Human Services
(HHS). With the goal of increasing the ability to compare outcomes across leaver
studies, the Office of the Assistant Secretary for Planning and Evaluation (ASPE) at
HHS awarded funding to nine states, the District of Columbia, and three large
counties or consortia of counties to examine welfare leavers. Grants were awarded
to Arizona, the District of Columbia, Florida, Georgia, Illinois, Massachusetts,
Missouri, New York, Washington, Wisconsin, Cuyahoga County (Ohio), Los Angeles
County (California), and a consortium of San Mateo, Santa Clara and Santa Cruz
counties in California. Somewhat comparable studies were funded in Wisconsin,
through the Institute for Research on Poverty (IRP), and in South Carolina, resulting
in a total of fifteen studies of former recipients funded by ASPE in FY1998. In
addition, Iowa received a grant from ASPE in FY1999 to examine welfare leavers.
Studies available as of January 2001 are reviewed for this report.
The ASPE leaver studies use a common definition of the “leaver” population,
studying those who leave cash assistance and remain off the rolls for at least 2
months. Most studies are reporting on outcomes for at least two cohorts of leavers,
an early cohort that left welfare in 1996 or 1997, and a later cohort that left in 1998
or 1999, after the transition from AFDC to TANF. Additionally, to address a number
of the concerns raised by the source of data (as discussed above), the researchers are
(...continued)
9
welfare leavers.
10 See Welfare Reform: Information on Former Recipients’ Status, Report by the U.S.
General Accounting Office, HEHS-99-48, April 28, 1999.

---

using a combination of linked administrative data sets and surveys to provide
information on the overall well-being of welfare leavers.11
Reason for exit from cash welfare
State administrators and policymakers want to understand why individuals leave
welfare. However, this outcome has proven difficult to compare across the studies
given a number of differences in reporting among the states. The states vary in which
reasons for exit they report and how they collect this information. This report makes
efforts to place the categories reported across the studies into groupings for purposes
of providing an overview. In general, the studies show:
! Employment/income-related reasons account for the largest number of case
closings, with the majority of states attributing over 40% of welfare exits to
these reasons.
! Case closures for failure to redetermine eligibility/failure to report for re-
certification were reported by a number of welfare leavers, including 26% of
leavers in Maryland and 24% of leavers in New York.
! Marriage accounted for a small number of case closings, ranging from 2%
(Florida) to 9% (Arizona).
! A small number of individuals didn’t want to be on welfare
anymore/requested case closure, ranging from 4% (Kentucky) to 15% (New
York).
! A small number of states also asked if “hassle” was a reason for case closing.
The percentage varied by state, ranging from 2% (Florida) to 9%
(Pennsylvania).
Methodological questions cloud interpretation of this outcome. Among the
studies, survey data report a higher percentage of case closures because of
employment compared to studies that rely on administrative data. Administrative
12
data may not capture all reasons for exit (such as employment) from cash welfare as
reasons for exit are based on information as reported by the case worker. If
individuals do not return to the welfare office or contact the welfare caseworker, their
case may be closed for failure to cooperate by the caseworker. For example, Arizona
attributed a substantial number of case closings to failure to comply with program
procedures (37%), loss of contact (3%), and voluntary withdrawal (7%), using
administrative data. However, using a survey to collect additional information on
these leavers, researchers found that almost half of the leavers with the above case
closures (48%) reported that their case closed because they either got a job, or
increased their earnings.
11 For more information, see: A Cross-State Examination of Families Leaving Welfare:
Findings from the ASPE-Funded Leaver Studies, by Julia B. Isaacs and Matthew R. Lyon,
Office of the Assistant Secretary for Planning and Evaluation (ASPE), HHS, revised
November 6, 2000 available at: [http://aspe.hhs.gov/hsp/leavers99/cross-state00/index.htm]
12 Two examples are studies from Arizona and Illinois. Arizona reported that 19% of
the closures in the administrative data were employment-related compared to 54% in the
survey data. Illinois reported a similar result when examining employment-related closures
in administrative data and survey data (33% vs. 53%, respectively).

---

Closures due to time limits. With the federal time limit of five years and
shorter state-imposed time limits, time limits may be affecting individuals in two
ways. First, they may encourage
13
those on welfare to leave more quickly, perhaps
to “bank” their time for future use. In Illinois, when asked to elaborate on other
factors that influenced their decision to leave welfare, 24% of leavers responded that
they were influenced to leave welfare by perceived time limits on welfare receipt.
Second, time limits may end cash assistance for those who exhaust their time on
welfare. Thus far under TANF, the majority of leavers attribute their cash welfare
exit to employment. To date, no information is available on individuals who leave
welfare because of the 5-year lifetime limit on federal cash welfare receipt (recipients
may begin reaching this time limit in some states beginning in 2001). However, in a
small number of states, leaver studies have focused on individuals whose reason for
exit from cash welfare was a shorter, state-imposed time limit. In general, these
leavers do not appear to be experiencing increased hardships or difficulties when
compared to those whose reason for exit was not a time limit.14
Individuals who have reached shorter state-imposed time limits may not be
indicative of individuals who will reach the federal 5-year lifetime limit. Individuals
whose exit from welfare is caused by a state time limit may be considered by some as
more ready for employment, as these individuals did not qualify for any state
exemption or exception to this time limit. In the Connecticut study, almost all of the
individuals in the sample were employed when they hit the 21-month time limit.
However, Connecticut does grant an extension to any individual not employed when
(s)he reached this time limit (assuming the individual is complying with the program
requirements).
15
Closures due to sanctions. TANF law requires states to penalize families
if an adult recipient refuses to engage in required work without good cause. For a
first work violation, 19 jurisdictions end TANF family benefits until compliance (full
13 For more information on time limits under TANF, see CRS Report RL30838, Welfare
Reform: Time Limits Under TANF, by Gene Falk, Emilie Stoltzfus, Holly Goodliffe, and
Courtney Schroeder.
14 North Carolina, and Virginia have studies that examined individuals whose reason for
exit was a state time limit. In Connecticut, Massachusetts, and Utah, while the study focused
on a larger population of welfare leavers, results were also reported separately for those whose
reason for exit was a time limit. Detailed information on each of these studies is available in
Appendix Table B-1.
15 The Connecticut study illustrates an example of the interaction of the state time limit
with incentives to work in the states by offering increased disregards of earnings. That is, a
number of those in this study maintained eligibility for TANF while working due to a TANF
state policy that disregarded a higher percentage of their earnings than would have been
available under AFDC. See: Melton, L. and D. Bloom (2000). Connecticut Jobs First
Program: An Analysis of Welfare Leavers. New York, NY; Manpower Demonstration
Research Corporation.

---

family sanction) while the remainder of the jurisdictions either remove the parent’s
share of the benefit or reduce the family’s benefit (partial family sanction).16
Tennessee is one of the states that ends TANF family benefits for the first work
violation. In Tennessee, which provided separate data for sanctioned welfare leavers,
it appeared that individuals understood they would be sanctioned for failure to comply
with program requirements. The state reported that 67% of leavers who were
sanctioned for refusal to sign a personal responsibility plan and 84% of those who
were sanctioned for noncompliance knew that this sanction meant less money or no
benefit.17
A small number of states reported the number of exits attributed to sanctions.
In Maryland, a state with a full family sanction for the first work violation, 11% of
leavers had their cases closed for work sanctions. Arizona and New Mexico are both
states that do not impose a sanction until the third work violation, and the sanction
is a partial benefit reduction. Approximately 20% of leavers in Arizona and 11% of
leavers in New Mexico had their cases closed because of a sanction. In both states,
sanctioned leavers were less educated when compared to the full sample of welfare
leavers. Those sanctioned in Arizona were also more likely to have reached the 24-
month time limit and to have less income from wages.18
Employment-Related Outcomes
As discussed earlier, the majority of welfare recipients attribute their exit to
employment-related activities. Although leavers may be employed when they leave
cash welfare, it may not be the case that they are employed after they leave cash
welfare.
Employment. Consistent with the finding that most leavers exit for
employment-related reasons, the leaver studies reported that from 63% to 91% of
those who left welfare had been employed (ever employed) for a period of time since
exit from welfare. The studies also report the percent employed at a certain point in
time, such as those employed when the study is being completed. A number of states
explored multiple groups of leavers who left cash welfare at varying times (i.e., those
who left in 1997 and those who left in 1998) and for different reasons (i.e., those who
left for employment and those who left due to a time limit). Because some states
16 For more information on sanctions in state welfare programs, see CRS Report
RL30767, Welfare Reform: Work Activities and Sanctions in State TANF Programs, by Vee
Burke.
17 Iowa, also a state with a full family sanction for the first work violation, completed
a study of the state’s “Limited Benefit Plan,” where benefits were discontinued in the 7th
month for failure to comply with procedural requirements. This program operated prior to
TANF, but the results are included in Table B-1.
18 Cases that reach the 24-month time limit in Arizona are neither considered to be
sanctioned, nor are they closed. Upon reaching the time limit, the family’s benefit it reduced
as the adult member of the household is “removed” from the cash assistance grant. The adult
continues to be eligible for supportive services. However, the cases may be subsequently
closed for other reasons.

---

examined multiple groups of welfare leavers, these states reported outcomes for more
that one study cohort. Figure 1 illustrates the number of leaver study cohorts that
reported employment rates within these defined ranges. Figure 1 illustrates results
using survey data and administrative data for each cohort (group)examined in the
leaver studies. The employment rates from the administrative data represent the
percent employed in the 1 quarter after exit (3 months after the quarter of exit); for
st
survey data, they represent the percent employed at the time of the survey (which
varies from 3 months to 30 months after exit).
Figure 1. Employment Rate: Number of Study Cohorts Within Each Range
20
15
10
5
0
30% to 44%
45% to 54%
55% to 64%
65% to 74%
75% or higher
Employment Rate Range
Survey data
Administrative data
Source: Congressional Research Service (CRS) based on information contained in the state leaver
studies.
Figure 1 illustrates three points:
! Among the study cohorts, the majority report employment rates between 55%
and 64% among welfare leavers.
! The employment rates vary considerably among the studies. Administrative
data (26 cohorts in 14 states and the District of Columbia) illustrate that from
45% to 72% of leavers are employed in the 1 quar
st
ter after exit.19 Survey data
(41 cohorts in 35 states and the District of Columbia) illustrate that between
34% and 77% are employed at the time of the survey.
! Studies that use the survey data reported a higher number of recipients
employed. Employment rates from administrative data come from the state
Unemployment Insurance (UI) wage files, which do not capture all types of
employment, such as “off-the-books” employment, employment outside of the
state, and federal employment. For this reason, some states have
supplemented their administrative data with survey data. Four states and the
District of Columbia reported both survey and administrative data results and
19 The Connecticut study reported a much higher employment rate (87%) for a sample
of leavers who left due to the state time limit.

---

found that employment rates in the survey results were higher (Arizona,
Illinois, Missouri, and Montana).20
Earnings. As with the employment rates, the earnings among employed leavers
vary dramatically among the states. The studies that use administrative data (26
cohorts in 14 states and the District of Columbia) report that quarterly earnings in the
1
st quarter after exit (period of 3 months) range from $1,999 to $3,868. Survey data
(17 cohorts in 15 states and the District of Columbia) illustrate that the average hourly
wage varies from $5.50 to $8.80. While the average wage reported show
21
s
substantial variation, a number of states also examined the distribution of the wages
and found that the majority of workers were making between $5.15 and $7.00 an hour
at the time of the survey. Further examination of wage information based on full-time
employment and part-time employment shows higher wage rates among those
employed full-time.
Occupations. In general, the studies indicate that most leavers are employed
in retail or clerical occupations. A study in New Mexico reported that persons
employed in retail jobs and restaurant jobs were more likely to work non-traditional
hours (defined as work that starts before 6 a.m. or ends after 6 p.m.) than persons
employed in office jobs, schools, or in similar situations. North Carolina (using two
separate studies) reported that a sizable number of welfare leavers “usually had to
work weekends”: 43% of those who left due to the time limit and 55% of its sample
of welfare leavers in 8 counties (who left for varying reasons).
Finding employment. A few surveys also asked leavers about how they
found their jobs. The majority of respondents in these surveys reported friends,
family, or their own search as the source of their job.
Employment-related benefits. The information available on employment
related benefits is limited, as only a few states that administered surveys collected
data. It is possible some individuals have to wait for employer-related benefits for a
few months after beginning employment. Therefore, the length of time between exit
and the interview may affect the results reported among the studies. Among
employed leavers in Missouri who were interviewed approximately 30 months after
exit, 53% were in jobs that offered health insurance, 52% had paid vacation days,
43% had dental benefits, 40% had paid sick days, 32% of the jobs had job training or
tuition reimbursement, and 4% offered paid child care. Approximately 71% of
welfare leavers interviewed in 8 North Carolina counties between 2 to 12 months after
exit were working for employers that offered health insurance. However, only 1 in
3 of these individuals (35%) were participating in employer-provided health care.22
20 A study in Washington reported that the employment rate in the first quarter after exit
(administrative data) was higher (62%) than the employment rate at the time of the survey
(59%).
21 Assuming the leaver worked 30 hours in all 13 weeks of the quarter, quarterly
earnings at these hourly wage rates would range from $2,145 to $3,432.
22 This issue is discussed in more detail in the section on health care among welfare
(continued...)

---

Results by time of exit from cash welfare. State leaver studies have
turned to administrative data to examine employment and earnings outcomes among
those who have left welfare at different points in time. That is, the studies have
examined whether those who left welfare during earlier years report higher
employment rates or earnings when compared to a cohort of leavers from a later
period of welfare reform.
Table 1 reports the results among four state studies that use administrative data
to examine leavers based on when they exit cash welfare. For each state, Cohort I is
the earliest group of welfare leavers as defined by the state. For example, in
Maryland, Cohort 1 is comprised of the first group of individuals who left welfare
during the study. Therefore, Cohort 3 is comprised of the most recent group of
welfare leavers in the study.
Table 1. Earnings and Employment Outcomes Among Leaver
Studies That Examine Leavers Based on When They Exit Cash
Welfare
Employment rate in 1st
Quarterly earnings in
State
quarter after exit
1 quarter after exit
st
Maryland
Cohort I
53%
$2,474
Cohort II
49%
$2,560
Cohort III
45%
$2,412
North Carolina
Work First Cohort I
68%
$2,251
Work First Cohort II
71%
$2,135
Work First Cohort III
70%
$1,999
Washington
Cohort I
51%
$2,778
Cohort II
57%
$2,714
Cohort III
62%
$2,387
Wisconsin
Cohort I
69%
$2,545
Cohort II
70%
$2,081
Source: Congressional Research Service (CRS) based on information in the state leaver studies.
Many contend that those who were among the first to exit under welfare reform
(the early leaver study cohorts) were perhaps the most ready for employment, and
therefore that individuals in these early cohorts would report higher employment rates
and higher earnings. This pattern emerges in Maryland, and also is consistent with
earnings data from the three other studies. However, as Table 1 illustrates,
(...continued)
22
leavers.

---

employment rates in the later cohorts in North Carolina, Washington, and Wisconsin
are higher. Therefore, a larger percentage of these leavers are working in the first
23
3 months after their exit, but these leavers report lower average quarterly earnings.
These lower earnings may reflect lower wage rates, or they may reflect fewer hours
of work (i.e., part-time employment rather than full-time employment).
Results by reason for cash welfare exit. States have used leaver studies
to explore outcomes among those who leave welfare for different reasons, such as
time limits and sanctions. The low average hourly wage of $5.50 was reported in
Tennessee, where the sample was comprised of those whose case was closed by a
sanction. Among the states’ studies that examined individuals whose reason for exit
was a state time limit, the average hourly wage in Virginia was $5.99 while the
majority of those employed in North Carolina (55%) were earning between $5.00 and
$6.99 per hour. Massachusetts reported a much higher hourly wage among those
who left due to time limit, $8.21 per hour. Some of the variation may reflect the
differing economic conditions as the minimum wage in Massachusetts is above the
federal minimum wage.
The highest employment rates at the time of the survey were reported in
Connecticut, Massachusetts, Virginia, and North Carolina (88%, 73%, 71%, and
66%, respectively), all of which were studies of individuals who left due to state time
limits on cash welfare. The state-imposed time limits often exempt individuals who
may not be considered “job-ready,” such as individuals with younger children or other
barriers to work. States may aid families beyond the federal time limit of 5 years, as
federal funds may be used for up to 20% of a state’s caseload without regard to the
time limit or states may use their own funds to continue aid to these families.
However, these studies of state-imposed time limits may not be representative of
individuals who may hit the federal time limit of 5 years (which will begin occurring
in some states later in 2001).
Employment Retention and Advancement. The data illustrate that
between 63% and 91% of welfare recipients have been employed for a period of time
since exit. However, the percentage employed at any given time is lower which
indicates that a number of leavers are not continuously employed. While this has
generated interest in examining employment patterns among welfare leavers, there is
very little information on employment retention and advancement among this
population.
Based on administrative data, between 35% and 58% of leavers in five states
were employed in four post-exit quarters. It is important to understand that these
24
23 The North Carolina study was divided into four exit cohorts: one cohort of recipients
who left AFDC (AFDC exit cohort), and three cohorts of recipients who left Work First
(TANF exit cohorts). Work First began as a waiver program in July of 1996, and when the
welfare reform law was signed in August of 1996, Work First was established as the state’s
TANF program with only minor programmatic changes needed to comply with the new law.
Among the three cohorts who left Work First, earnings have declined over time.
24 This is based on information for Los Angeles (35%), Illinois (39%), New York (40%),
(continued...)

---

data may not reflect continuous employment in these quarters, as it is not possible,
using these data, to determine the length of employment, the hours of employment,
or whether individuals were continuously employed. The UI wage files used to
provide employment and earnings information are not designed to provide information
on continuous employment or wage advancement. For example, the studies show that
quarterly earnings in the 4 quarter after exit
th
are higher than quarterly earnings in the
1 quarter after exit. However, these data do not illustrate wages among the sam
st
e
population of welfare leavers. Rather, they illustrate wages among leavers employed
in the 1 quarter after exit and leavers employed in the 4
st
quarter after exit. Th
th
e
composition of these populations is likely to be different, as some employed in the 1st
quarter after exit may not be employed in the 4 quarter after exit (and vice-versa).
th
A few studies have also used surveys to provide more subjective information.
In New Mexico, 58% of employed respondents were very satisfied with their job and
55% said it was very likely that they would stay in their current job. In Ohio, 42% of
welfare leavers were employed by a single employer for 6 months after leaving.
Barriers to employment and reasons for unemployment. The leaver
studies indicated that the majority of leavers are employed for a period of time since
exit, and between 55% and 64% are employed at the time of the study. Among those
leavers who are unemployed at the time of the study, a number of studies examined
barriers to employment or reasons for unemployment. Some of the studies asked for
the “main reason” individuals were unemployed, while in other studies individuals
could provide more than one response. Highlights include:
! Health/Physical/Mental Illness: Health, physical illness, or mental illness of
the respondent was the number one reason unemployed individuals gave when
asked for the main reason for unemployment in Arizona (23%), Missouri
(26%) and Kansas (43%), and accounted for the most responses in North
Carolina (18%) and Illinois (32% expressed physical illness and 18% expressed
mental illness). Among the 19% of unemployed leavers in New Mexico who
indicated that their health, physical or mental illness was a reason for
unemployment, 78% said that the disability was permanent. The health,
physical illness, or mental illness of others (e.g., dependents or other members
of the household) was seen as a reason for unemployment by 35% of the
respondents in Utah, but other studies (6 states) indicated a much smaller
percentage, ranging from 4% (North Carolina time limit study) to 9% (South
Carolina).
! Couldn’t find a Job/Couldn’t find a Job that Pays Enough: When asked
for the main reason leavers were unemployed (6 states), between 6%
(Washington) and 25% (Idaho) said that they couldn’t find a job or that they
couldn’t find a job that pays enough. Among those who could offer multiple
reasons for unemployment (5 states), between 8% (Pennsylvania) and 46%
(Illinois) indicated that difficulty finding employment was a reason for
unemployment. In Missouri, 4% said they did not want to work. Among
those who left due to the time limit in North Carolina, 42% expressed that they
(...continued)
24
Wisconsin (40%), and Maryland (58%).

---

couldn’t find a job at the time of the second survey, up from 29% at the time
of the first survey.
! Child Care Problems: Between 3% (Washington) and 22% (Arizona) of
unemployed leavers indicated that child care problems were the main reason
for unemployment. Child care was the number one reason (21%) leavers quit
their jobs in New Mexico. Child care problems were also one potential reason
for unemployment, ranging from 11% to 15% (3 states). In Illinois, 52%
indicated that they had a problem finding child care for work hours. In four
states, unemployed leavers chose to stay home and care for their children: this
was the number one reason for unemployment in Washington (21%) and
accounted for a sizable number of responses in Pennsylvania (6%) and New
Mexico (19%).
! Transportation Problems: From 1% in Washington to 42% in Utah of
unemployed leavers indicated that transportation problems were a reason for
unemployment.
! Domestic Violence: A small number of the state leaver studies collected
information on whether domestic violence was a barrier to employment. In
Oklahoma, 15% of leavers reported domestic violence or alcohol problems that
interfered with work. In Illinois, 27% of leavers were ever in an abusive
relationship. Of these, 35% said that this abusive relationship had been a
problem in getting/keeping a job.
Overall Well-Being of Welfare Leavers
Many states have gathered information beyond income and wages from
employment to more comprehensively examine the well-being of welfare leavers,
including measures such as income and wages of other household members, support
from other government programs (such as food stamps and Medicaid), and
“hardships” experienced by those who leave welfare (such as difficulty paying bills or
accessing health care). Some administrators and policymakers are concerned about
whether those who are eligible for assistance under other government programs such
as food stamps and Medicaid are continuing to receive these benefits upon exit from
cash assistance, especially given declines in these non-cash programs caseload.
Income/Poverty Status. Between 55% and 64% of leavers were employed
at the time of the studies, and an even larger majority had been employed for some
time since exit. However, although the majority of welfare leavers were employed,
based on wages alone, most welfare leavers remained poor.
Among single-parent families with a child under the age of 18 in Missouri, 34%
of the families were in extreme poverty (defined as 50% of the poverty line or less),
43% were below the federal poverty level, and 92% were below or near the federal
poverty level. In addition to the earnings of the welfare leavers, measures of family
income typically include earned and unearned income of other household members
(e.g., spouses) which may be important sources of support for welfare leavers.
Measures of Hardship. Table 2 illustrates information from state leaver
studies on measures of hardships, as reported. These studies use survey data to ask
families to provide more subjective assessments of the hardships they have faced since

---

leaving welfare. Table 2 includes information on four areas: paying bills, health care,
food insecurity, and housing arrangements. The table also provides the range of
results (i.e., percent of leavers who reported experiencing a particular hardship) and
indicates the number of states included in this range of results.

---

Table 2. Measures of Hardships as Reported in State Leaver
Studies
Percent of respondents
“Hardship” measure
reporting hardship
Paying bills
Behind on utility bill
11% to 59% (11 states)
Electricity shut off
2% to 36% (6 states)
Phone cut-off
22% to 48% (6 states)
Help from family or friends
20% (2 states)
Health care
Needed medical care but couldn’t get it
13% to 54% (8 states)
Food insecurity
Had a time when no way to buy food
13% to 44% (8 states)
Housing arrangements
Moved or evicted since cash welfare exit
5% to 21% (11 states)
People move in to share expenses
1% to 17% (4 states)
Move in with people to share expenses
8% to 30% (8 states)
Had been or were homeless
1% to 4% (8 states)
Source: Congressional Research Service (CRS) based on information in state leaver studies.
Hardships before and after exit from cash welfare. Table 2 illustrates
that individuals experienced difficulties after leaving welfare. However, a number of
states also asked welfare leavers to provide “hardships” experienced while receiving
welfare as well as after exit from welfare. Findings include:
! Paying bills: While leavers in Illinois reported greater problems in paying
utilities before leaving welfare (27% before compared to 14% after), well-over
twice as many leavers in South Carolina reported a problem paying utilities
after leaving welfare (4% before compared to 11% after). A similar result was
found among those who left due to the time limit in North Carolina, where 4%
reported a problem paying utilities before leaving welfare compared to 11%
who reported the problem after leaving welfare.
! Health care: There is some evidence that welfare recipients experience more
difficulty acquiring health care after leaving welfare (3 states). Illinois and
New Mexico both reported an increase in the number of individuals who
needed medical care but couldn’t get it after leaving welfare, among their
overall sample of welfare leavers (9% needed medical care but couldn’t get it
before leaving welfare compared to 19% after leaving welfare in New Mexico;
26% needed medical care but couldn’t get it before leaving welfare compared
to 31% after leaving welfare in Illinois). North Carolina also reported an
increase in the number of adults who needed medical care but couldn’t get it
after leaving welfare among those who left welfare due to the time limit (4%
before compared to 24% after), but reported no difference among children.
! Food insecurity: There also is some evidence that the problem of food
insecurity in some places has increased after exit from welfare. Wisconsin,

---

New Mexico, and South Carolina all reported a larger percentage of
respondents with this hardship after leaving welfare compared to when the
respondents were receiving welfare (32% after leaving welfare compared to
22% while receiving welfare in Wisconsin; 25% after leaving welfare compared
to 15% in New Mexico; 13% after leaving welfare compared to 6% while
receiving welfare in South Carolina).25
! Housing arrangements: Illinois, New Mexico and Wisconsin reported that
the incidence of problems with housing arrangements was lower since their exit
than it had been while the respondents were on welfare, although in Illinois this
varied by employment status. While those currently employed reported greater
problems in housing before leaving welfare (43% before leaving welfare
compared to 29% after), those currently unemployed reported greater
problems in housing after leaving welfare (54% after leaving welfare compared
to 47% before leaving welfare). In two North Carolina studies, leavers
indicated that they were more likely to get behind on rent after leaving welfare
(10% before leaving compared to 22% after leaving welfare among those who
left due to the time limit; 21% before leaving compared to 31% after leaving
welfare among leavers in 8 selected counties).
Child Welfare Outcomes. Some contended that children would be harmed
by welfare reform, but to date little information has been collected on outcomes of
children among families who leave welfare. State leaver studies report that in 3%
26
to 19% of households, a child(ren) left the home since cash welfare exit. However,
no additional information was provided to determine where these children were
placed, such as whether the child was removed from the home by a child welfare
agency, whether the child moved in with the father, or perhaps whether the child
moved out on their own. The sample sizes also were quite small, as the children in
these situations comprised a small portion of the overall sample of leavers. For
example, 74 children (19%) in Utah left the home since cash welfare exit. When
asked why the children left home in Utah, the majority of respondents (43%) gave
“other” as the reason, with no additional clarification. Additional reasons offered
were that 18% were removed by Child Protective Services (CPS), 12% indicated that
they couldn’t afford to care for their child, and 4% indicated that the child left to
establish his/her own household.
Both Arizona and Illinois reported no increase in the number of substantiated
child protective service (CPS) reports or out-of-home placements following exit from
cash assistance. In Utah, 50% reported a CPS referral, a slight increase over the 44%
who reported a CPS referral during an earlier survey. Maryland reported that 12%
25 In South Carolina, respondents were asked to give the reasons for their inability to buy
food after leaving welfare. The most common responses were “food stamps were not enough
to pay for food” (28%), “had unexpected or emergency expenses” (18%), “spent all money
on things other than food” (16%), “lost job or lost job that provided food” (16%), and “had
a hard time budgeting” (14%).
26 Child Trends, a nonprofit, nonpartisan research organization that studies children,
youth and families, recommended a set of child outcomes to be examined, in response to a
request of those awarded grants through ASPE to conduct leaver studies. Their
recommendations can be found at: [http://aspe.hhs.gov/hsp/leavers99/recommended.htm]

---

of the children in the sample had a pre-exit indication or confirmation of child abuse
or neglect; 90 days after exit, 1% reported an indication or confirmation of child
abuse or neglect compared to 2% in the 12 months after exit.
Receipt of Other Types of Assistance
A stated goal of TANF is to end dependence on government benefits by
promoting job preparation, work, and marriage. The leaver studies indicate that while
leavers are employed, most remain poor and may therefore rely on other types of
assistance to supplement their wages. For example, employed leavers may qualify for
(EITC).
After cash welfare exit, a number of leavers and their children also remain
eligible for health insurance, child care, and food stamps. However, declining
participation in these programs has raised concerns. Some argue that low
participation rates illustrate that leavers have replaced these government benefits with
other types of support, thereby ending their dependence on government benefits.
However, others argue that recipients are leaving welfare for low-wage work and
therefore remain in need of health insurance and food assistance. They believe that
eligible leavers are not participating in these programs and therefore, outreach to
increase enrollment should be the emphasis. Each of these areas, as examined in the
leaver studies, is briefly discussed below.
Health Insurance. When welfare recipients leave welfare for work, they do
not always replace public insurance with private or employer-provided health
insurance. In part, this may be because some who leave welfare are not employed or
because some are not offered health insurance through their employment. Among the
state leaver studies, participation in private health insurance ranged from 7% in South
Carolina to 21% in Illinois. However, this low rate of private insurance coverage may
also be due to the fact that some employer-provided health plans require a certain
period of time in employment before eligibility, the costs of the employer provided
health-insurance may be too high, and/or the insurance doesn’t meet the needs of the
individual. For example, 43% of employed leavers in New Mexico were participating
in an employer health plan if available; among those not participating, 32% had not
worked in the job long enough and 31% indicated that the cost was too high.
There is some evidence that likelihood of participation in employer-provided
health insurance may increase the longer the individual has been employed. Among
those in North Carolina who reached the time limit, 65% were participating in
employer-provided health insurance if available at the time of the second survey,
compared to 33% when the first survey was administered. This increase was due to
an increasing number of individuals who were eligible to participate in employer-
provided health insurance. Among those who were not participating, 44% were
working part-time, 28% hadn’t worked there long enough, and 12% indicated that the
cost was too high; among those not participating in employer-provided health
insurance if available, 80% were enrolled in Medicaid.
The state leaver studies indicate that health care coverage is an issue for many
welfare leavers. Prior to TANF, individuals qualifying for AFDC were automatically
eligible for, and in most states, automatically enrolled in Medicaid. The welfare

---

reform law of 1996 “delinked” Medicaid from cash assistance, and therefore TANF
eligibility does not confer automatic Medicaid eligibility. Rather, Medicaid
entitlement was retained for individuals who meet the requirements of the former
AFDC program in effect on July 16, 1996 (herein referred to as eligibility under
Section 1931) regardless of whether they are receiving TANF.
Because Medicaid eligibility is not based on TANF eligibility, states may not
deny Medicaid to any member of a family, if the family loses eligibility for TANF
because of employment, time limits, sanctions, or for any other reason. States are to
“exhaust all avenues to eligibility before denial or termination” of Medicaid
eligibility.
27 States are also required to provide transitional medical assistance (TMA)
for up to 12 months to families who, because of hours of work, income from
employment (or loss of the earned income disregard), or child support income, lose
their eligibility under Section 1931. Regardless of whether they lose TANF
assistance, it is the loss of Medicaid eligibility under Section 1931 that triggers
eligibility for TMA. Finally, states also have the option to provide continuous
Medicaid eligibility to children, and to terminate Medicaid coverage for a TANF
recipient whose assistance is terminated for refusing to cooperate with TANF work
requirements.28
Medicaid coverage rates, as reported in the studies, are difficult to interpret
given the structure by which those who leave welfare continue to qualify for Medicaid
(as discussed above). Results for each state are included in Appendix Table B-1.
Among those studies that ask if anyone in the household was receiving Medicaid since
exit or at the time of the survey (adults or children), the range was from 58% in New
Mexico (6 to 18 months after exit) to 87% in Wisconsin (6 to 9 months after exit).
Among those studies that examine rates of participation for adults and children,
the rates for children are higher regardless of the period of time since exit. For
adults, participation rates range from 15% in Colorado (2 years after exit) to 72% in
Arizona (ever received in the 12 months since exit). Among children, participation
rates range from 35% in New York (9 to 12 months after exit) to 77% in Washington
state (6 to 9 months after exit). Utah found that participation rates in Medicaid varied
by reason for case closing: among those who left due to income or other reasons,
participation rates in Medicaid were lower (64% and 63%, respectively) than those
who left due to the time limit (84%).
Compared to the children of respondents who leave welfare, the respondents
themselves are more likely to be uninsured. In Missouri, 11% of households had
uninsured children while 41% of the households had uninsured adults; the lowest rates
of uninsurance were found in households where the leaver had never worked. In
Illinois, 29% of children and 36% of respondents were uninsured.
27 For more information, see : Supporting Families in Transition: A Guide to Expanding
Health Coverage in a Post-Welfare Reform World, available at
[http://www.hcfa.gov/medicaid/welfare.htm]
28 Excluded from this option are pregnant women and children eligible under a poverty-
level group and minor children who are not heads-of-household under TANF. For more
information, see: Supporting Families in Transition.

---

Among those who are not receiving Medicaid, reasons for not participating vary.
In Arizona, 66% of those not receiving Medicaid thought they were not eligible and
13% considered it not worth the effort. In New Mexico, 23% were planning to apply,
20% were told they were not eligible, 18% didn’t believe they were eligible and 16%
said they no longer needed Medicaid; 9% thought it was not worth the effort. In
Illinois, among those who never applied and never received Medicaid, 45% did not
need Medicaid or had other health insurance; 17% didn’t believe they were eligible
and 16% believed it was too much hassle/too time consuming.
While a number of studies indicate that individuals think they are not eligible for
Medicaid and therefore are not participating, there is some evidence that individuals
may be receiving Medicaid services although they think the services have been
discontinued. Among the 22% of welfare leavers in Washington state who did not
know that they could continue to receive Medicaid when they exit cash assistance, 1
in 3 were enrolled in Medicaid.
To address issues related to Medicaid coverage for welfare leavers, on April 7,
2000, HHS ordered states to search computer files to identify families improperly
dropped from Medicaid and to reinstate them retroactively. HHS has also included,
beginning in TANF performance year FY2001, a Medicaid/State Children’s Health
Insurance Program (SCHIP) performance measure in awarding TANF bonus funds.
States have the option of competing on this performance measure, and may forego
reporting this measure but still compete on other TANF performance measures, such
as work and family formation measures. If the state chooses to compete, to be
eligible the state will have to follow certain qualifying procedures and will be judged
on the percentage of welfare leavers who continue to be enrolled in Medicaid or
SCHIP 6 months after leaving TANF.29
Child Care. Individuals who have left welfare have indicated some barriers to
child care, such as the cost of child care or difficulty obtaining child care. A
30
s
discussed earlier, some welfare leavers have also said that child care related problems
have been a barrier to employment. Among the studies that include child care
information, relatives and friends are reported as the primary source of care, but use
of government subsidies varies considerably from 3% in the New Orleans
metropolitan area to 55% in Missouri. When examining the percent who receive child
care subsidies among those who are eligible for subsidies, the percentages increase
but still vary considerably from 20% in Illinois to 67% in North Carolina.
Some of the leaver studies indicate that a major reason for the low receipt of
child care subsidies is that many individuals simply judge that they do not need child
care assistance, and many individuals do not want their children cared for by someone
else. However, a number of individuals also appear to be unaware of the availability
of child care assistance. For example, among those not using a subsidy in Arizona,
62% said they did not need the subsidy, while 21% said they were not eligible.
29 For more information, see CRS Report RS20552, Welfare Reform and Medicaid:
Brief Overview, by Vee Burke.
30 See: Schumacher, Rachel, and Mark Greenberg. Child Care After Leaving Welfare:
Early Evidence from State Studies. Center for Law and Social Policy, 1999

---

Among those who were not using subsidies in Washington, 39% had no need and
13% thought it was too much trouble to apply. Among those who were paying for
child care but were not receiving help in New Mexico and North Carolina,
respectively, 31% and 27% didn’t know they could get help and 18% and 20%
applied but did not qualify.
The type of child care used among welfare leavers also varied considerably. In
Washington state, leavers with subsidies were more likely to use day-care centers than
those without subsidies (39% vs. 8%); those without subsidies were more likely to
use grandparents or other relatives when compared to those with subsidies (49% vs.
20%). In New Mexico, 73% of individuals who pay for child care and receive help
from the county use informal child care. Leavers without subsidies in Missouri lived
in households with more adults, which may imply less need for outside care.
Food Support. Participation in food stamps fell from 25.5 million individuals
in fiscal year (FY) 1996 to 17 million in July 2000, a decline of 33%. A report
completed by the GAO found that the strong U.S. economy, tighter food stamp
eligibility requirements, and welfare reform initiatives are the primary reasons for the
decline in food stamp participation. However, the report also found that some
households, including those with eligible children, have had problems obtaining food
stamps because some state and local governments “have gone farther than the law
permits in limiting benefits.”31
According to the leaver studies, the number of individuals who received food
stamps after leaving cash assistance ranged from 30% in Kansas to a high of 83% of
children in Oklahoma at the time of the survey, with the majority of studies reporting
food stamp participation rates between 46% and 78%. In Missouri, eligible
households that were not receiving food stamps were better off financially than
households that did receive food stamps; median income was 40% higher and these
households were more likely to own their own home. In households that didn’t
receive food stamps in Missouri, 90% had someone receiving Medicaid.
As with Medicaid and child care, reasons for not participating in food stamps
vary among the studies. Among those not receiving food stamps in Arizona, 58%
believed they were not eligible and 19% said they did not need food stamps.
Individuals who left because of sanctions in Arizona were more likely to report that
they believed they were not eligible (69% vs. 58% overall). Among those not
receiving food stamps in New Mexico, 32% were told they were not eligible, 24% no
longer needed food stamps, 16% didn’t believe they were eligible, and 8% believed
it was not worth the effort.
There is some evidence that eligible individuals choose not to participate in food
stamps. In Illinois, of those who had not received food stamps since exit, 41%
thought they were eligible and 29% said a caseworker told them they were eligible.
In New Jersey, approximately 3 out of 4 of those who left TANF and weren’t
receiving food stamps (72%) knew that they could receive food stamps.
31 See: Food Stamp Program: Various Factors Have Led to Declining Participation.
Report by the U.S. General Accounting Office, RCED-99-195, July 1999.

---

Child Support. Among welfare leavers, a number of studies report child
support as an important source of income. Among the studies that reported the
percent who collected child support, the range varied from 8% of leavers in
metropolitan New Orleans to 41% of leavers in Idaho. In general, the studies also
show that less than half of those who have court-ordered child support arrangements
are actually receiving a payment. In Massachusetts and Washington, respectively,
41% and 44% report a child support agreement, but only 22% and 23% report
actually receiving child support. In New Mexico, 30% had orders while only 10%
received child support. However, there is also some evidence that leavers receive
child support even without court-ordered support payments. In Illinois, 20% of those
without court-ordered child support reported some child support payments.
Approximately 31% of the leavers in Utah were receiving child support, but this
number varied greatly by reason for case closing. Among those who left for income,
48% reported receiving child support, while the percentages among those who left
because of the time limit or for other reasons were smaller (23% and 25%,
respectively).
Earned Income Tax Credit. The Earned Income Tax Credit (EITC) offers
a refundable tax credit to low-income working parents who care for dependent
children (with smaller credits available for low-income individuals with no children).
Although EITC is available on an ongoing basis (e.g., supplement to their paycheck),
most receive it when they file for their federal income taxes. A small number of
surveys collected information on the EITC among welfare leavers. While a number
of leavers were aware of the EITC, a small percentage actually filed for the tax credit.
In Illinois, 85% of leavers had worked during the study period, but only 41% have
received the EITC. Part of this may be attributed to the fact that only 47% of the
leavers in the Illinois study had a clear idea of what the EITC was.
Returns to Welfare Among Welfare Leavers
Given the dramatic declines in the welfare caseload, many are interested in
whether individuals leaving welfare remain off welfare. A number of the studies
measured how many leavers return to cash welfare. In general, the studies find that
between 18% and 35% of those who leave welfare have returned in a period of time
since exit (which varies by study). For example, Arizona reported that 28% of leavers
had returned at least once in the 12 months since exit, while New York reported that
21% of leavers had returned at least once in the 12 months since exit.
Studies have found that many cases that were closed and re-opened within a
month had been closed in error. Others may have their cases closed on technical
grounds (e.g., failure to report income). These individuals may return to cash
assistance immediately (“churners”). To include “churners” in these studies may
overstate the incidence of recidivism among leavers. Maryland found that 40% of all
cases returned within 36 months of exit, but after excluding those who returned within
30 days of case closure (the “churners”), the number declined slightly to 36%.
Employment is the most common reason for leaving welfare, and lack or loss of
employment is also the most common reason individuals return to welfare. In Illinois,
survey data illustrate that 44% of welfare returns were because the individual couldn’t

---

find a job or lost a job; 12% were child-care related. In Arizona, 54% of those who
returned lost a job or had a reduction in earnings (40% for those who were sanctioned
and 58% of those who left for other reasons); 11% returned due to divorce. In
Washington, 47% of welfare returns were employment-related (25% were laid-off or
fired, 10% lost work hours, and 12% quit the job for health reasons); 13% were due
to a marriage or partnership breakup.
A few states asked respondents questions regarding the likelihood that they
thought they would return to welfare. In North Carolina, 43% thought it was very
unlikely they would reapply to welfare in the next 6 months while 15% found it very
likely that they would return. New Mexico reported a similar result: 57% thought
it was very unlikely they would reapply to welfare in the next 6 months while 15%
found it very likely that they would return. However, New Mexico respondents who
left because of benefit cut-off or a sanction were more likely to indicate that they were
“very likely” to reapply for welfare (38%); those who left because of employment
were less likely to indicate that they were “very likely” to think they would reapply
(9%).
Along with differences in employment and earnings, studies have reported
differences in rates of return to TANF among cohorts of welfare leavers. Some
expect that those who left welfare more recently may be harder to serve and therefore
may be more likely to return. Because these leavers didn’t leave welfare during the
early period of reform, they may have more barriers to employment that might impede
a successful transition to work. In Maryland, those who left in the later years were
far more likely to return to welfare. In general, those who left in Year 1 of the study
had lower rates of return than those who left in Year 2, who in turn had lower rates
of return than those who left in Year 3. Among all cases who left welfare (including
those who returned within 30 days), 24% of those who left in Year 1 returned within
12 months, 41% of those who left in Year 2 returned within 12 months, and 54% of
those who left in Year 3 returned within 12 months. Excluding those who returned
within 30 days (the “churners”), 18% of those who left in Year 1, 21% of those who
left in Year 2, and 34% of those who left in Year 3 returned within 12 months. It
appears that, not only were those who left welfare in later years returning at a higher
rate, they were more likely to return within 30 days of exit. This may also be another
indication the later cohorts are less ready to find and retain jobs after leaving welfare.
Why a family left welfare also appears to play a role in why a family returns to
welfare. Those who leave welfare for income reasons appear less likely to return than
those who exit because they failed to fulfill program requirements. In Maryland, while
25% of those who left because of income returned within 12 months of exit, 53% of
those whose cases were closed by family sanction and 52% of those whose cases were
closed for failure to reapply/redetermine eligibility returned within 12 months; in both
cases, approximately half of those who returned within 12 months did so within 30
days of exit (“churners”). In Illinois, 21% of those leaving for income reasons
returned within 12 months of exit compared to almost 40% of cases closed for non-
cooperation.
Finally, a number of state leaver studies asked those still receiving welfare what
they considered to be an important resource that would enable them to leave welfare.
In the 8-county study completed in North Carolina, respondents were asked what

---

types of services they felt they would need in the next 6 months to remain off welfare
(including services they were currently receiving and wished to continue receiving).
Child support was the service respondents felt they would need (55%), with child care
the next most identified service (50%).

---

Appendix Table A-1. Administrative Data: Employment and
Earnings of Welfare Leavers
Earnings
Employment
1st
4th
1st
4th
Ever
Quarter
Quarter
Quarter
Quarter
employed
State
after exit
after exit
after exit
after exit
since exit
Study examined a single group of leavers
Arizonaa
$2,211
$2,511
53%
50%
73%
Arkansas
$2,034
$2,299
50%
NR
66%
b
b
California (3 counties)
$3,373
57%
NR
Colorado
$2,400
$2,603
52%
51%
NR
Connecticut
$3,369
$3,516
49%
50%
63%
b
b
Connecticut (Time limit)
$2,966
88%
NR
Cuyahoga County (Ohio)a
$2,756
$2,952
59%
57%
NR
District of Columbia
$3,416
$3,934
62%
60%
NR
Georgiaa
$2,193
$2,389
64%
53%
74%
Illinois
$2,659
$2,961
54%
54%
69%
Los Angeles (California)a
$3,414
$3,576
47%
47%
NR
Missouri
$2,192
$2,698
58%
58%
80%
New Yorkc
$3,868
$4,230
55%
53%
66%
Study examined multiple groups of leavers based on their time of exit from cash welfare
Maryland
Cohort 1
$2,474
NR
53%
NR
NR
Cohort 2
$2,560
NR
49%
NR
NR
Cohort 3
$2,412
NR
45%
NR
NR
North Carolina
AFDC exit cohort
$2,075
$2,211
66%
62%
NR
Work First Cohort I
$2,251
$2,275
68%
65%
NR
Work First Cohort
$2,135
$2,374
71%
66%
NR
II
Work First Cohort
$1,999
$2,119
70%
63%
NR
III
Washington
Cohort I
$2,778
$3,340
51%
56%
NR
Cohort II
$2,714
$3,351
57%
58%
NR
b
b
Cohort III
$2,387
62%
NR
Wisconsina

Cohort I
$2,545
$2,940
69%
69%
81%
Cohort II
$2,081
$2,744
70%
68%
84%
Source: Table prepared by the Congressional Research Service (CRS) based on state leaver studies.
The results among four state studies that use administrative data to examine leavers are based on
when they exit cash welfare. For each state, Cohort I illustrates results for the earliest group of
welfare leavers (i.e., the first group of leavers) as defined by the state. For example, in Maryland,

---

Cohort 1 is comprised of the first group of individuals who left welfare during the study. Therefore,
Cohort 3 is comprised of the most recent group of welfare leavers in the study. Detail may differ from
data reported in Table B-1 due to rounding. NR (Not Reported) indicates that the state did not
report this information .
Study includes only single-parent welfare cases.
a
b Fourth quarter data were as not available.

c In New York, the authors attempt to adjust the earnings to reflect employment not captured by the
Wage Reporting System, which was found to understate quarterly earnings by 16% to 21%. If a 16%
adjustment were used, the 4th quarter earnings reported here would increase from $4230 to $5034.

---

Appendix Table A-2. Survey Data: Employment and Earnings of
Welfare Leavers
Percent
Percent
Length of time
employed
ever
between survey
Wage reported at
at time of
employed
State
and exit
time of survey
survey
since exit
Reporting average hourly wage
Alabama
7 to 12 months
$6.08
54%
NR
Arizona
12 to 18 months
$7.52
58%
NR
California (3
6 to 12 months
$8.80
56%
NR
counties)
District of
12 months
$8.74
60%
NR
Columbia
Illinois
6 to 9 months
$7.89
63%
85%
Massachusetts
6 to 16 months
$8.21
73%
NR
(Time limit)
Massachusetts
6 to 16 months
$8.62
71%
91%
(Non-time limit)
Michigan
1 to 21 months
$7.02
71%
NR
Mississippi
3 to 9 months
$5.77
35%
47%
Missouri
30 months
$6.00
65%
90%
Oklahoma
3 to 18 months
$6.51
50%
NR
South Carolina
12 to 15 months
$7.00
55%
81%
Tennessee -
1 to 9 months
$5.50
39%
NR
sanctioned
Texas
6 months
$6.28
55%
68%
Virginia
6 to 12 months
$5.99
71%
86%
Washington
6 to 9 months
$7.70
59%
86%
Wisconsin
6 to 9 months
$7.42
62%
83%
Reporting a distribution of the average hourly wage
a
Colorado
24 months
Less than $5.16 = 10%
65%
89%
$5.16 - $7.00 = 26%
Connecticut
18 months
Less than $7.50 = 26%
70%
NR
$7.50 - $11.99 =
47.9%
Idaho
1 to 5 months
Less than $5.25 = 27%
53%
NR
$5.25 - $7.00 = 51%
Indiana
12 to 18 months
Less than $6.00 = 39%
64%
84%
$6.00 - $7.99 = 41%
Kansas
12 months
Less than $5.01 = 10%
36%
NR
$5.01-$7.00 = 51%
Kentucky
12 to 20 months
Less than $5.16 = 13%
48%
NR
$5.16 - $6.99 = 46%
New Mexico
6 to 18 months
Less than $5.00 = 9%
63%
NR
$5.00 - $6.99 = 49%

---

Percent
Percent
Length of time
employed
ever
between survey
Wage reported at
at time of
employed
State
and exit
time of survey
survey
since exit
New York (New
6 months
Less than $5.75 = 10%
NR
67%
York City)
$5.75 - $7.00 = 30%
North Carolina - 8
2 to 12 months
Less than $7.00 = 39%
64%
NR
counties
North Carolina -
13 to 16 months
$5.00 to $6.99 = 55%
66%
NR
time limit
Wyoming
1 to 15 months
Less than $5.01 = 4%
61%
NR
$5.01-$7.50 = 47%
Source: Table prepared by the Congressional Research Service (CRS) based on information in the
state leaver studies. Detail may differ from data reported in Table B-1 due to rounding. NR (Not
Reported) indicates that the state did not report this information .
The percentage of employed respondents who earn within this range of wages is reported.
a

---

Appendix Table B-1. Selected Outcomes from Welfare Leaver Studies
Table B-1 presents a state-by-state table of the studies reviewed in this report. This table presents descriptive information
on the study population and methodology (such as the length of follow-up and the response rate) and highlights selected
outcomes among the state studies. For each state, the most recent study available is included. However, Table B-1 includes
more than one study for those states where methods of collecting the data were different among the studies (i.e., a study that
used administrative data versus a study that used survey data), the study populations were different, and/or the studies
reported different types of outcomes.
Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Alabama: Welfare Reform Follow-up Survey, January 2000
Universe: Sample of all cases closed
Earnings:
Health assistance:
between July and November of 1998
Average hourly wage: $6.08
Receiving Medicaid (self/children) at time of
that contained a valid Alabama phone
survey: 33.7%/72.8%
number.
Hours of employment:
Average hours per week: 33.9
Child care assistance:
Data type: Survey conducted between
Receiving help paying for child care at time of
May and July of 1999, 7 to 12 months
Employment rate:
survey: 20.4%
after exit.
At time of survey: 53.8%
Food assistance:
Response rate: 416 respondents
Recidivism:
Receiving food stamps at time of survey:
surveyed.
Receiving cash assistance at time of survey:
59.9%;
15.1%
Receiving WIC at time of survey: 34.9%
Arkansas: Evaluation of Arkansas’s Transitional Employment Assistance (TEA) Program, July 1999
Universe: Single and two-parent cases
Earnings (cases closed July 1997 to September
Health assistance:
closed between June 1996 and
1998):
Receiving Medicaid within 2 months of exit
December 1998.
1st quarter after exit: $2,034;
among cases closed July 1998 to September
4th quarter after exit: $2,299
1998 (adults/children): 56%/60%
Data type: Administrative data.
Employment rate (cases closed July 1996 to
Follow-up: Up to 4 quarters (varies).
September 1998):
1st quarter after exit: 50%;
Ever employed since exit: 66%
Recidivism:
Ever received cash aid from June 1996 to
December 1998: 23.2%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Arizona: Cash Assistance Exit Study: First Quarter 1998 Cohort Final Report, January 2000
Universe: Single-parent cases closed
Earnings:
Health assistance:
between January and March of 1998
1st quarter after exit: $2,211;
Ever received Medicaid in the 4 quarters since
that remained closed for at least 2
4th quarter after exit: $2,511
exit: 71.7%
consecutive months.
Employment rate:
Child care assistance:
Data type: Administrative data.
1st quarter after exit: 53.1%;
Ever received child care in the 4 quarters since
4th quarter after exit: 50.1%;
exit: 21.8%
Follow-up: 12 months.
Ever employed since exit: 73.3%
Food assistance:

Recidivism:
Ever received food stamps in the 4 quarters
Ever received cash aid in the four quarters since
since exit: 66.5%
exit: 27.7%
Arizona: Cash Assistance Exit Study: First Quarter 1998 Cohort Final Report, January 2000
Universe: Single-parent cases closed
Earnings:
Health assistance:
between January and March of 1998
Average hourly wage: $7.52;
Receiving Medicaid at the time of survey
that remained closed for at least 2
Average total monthly household income:
(adult/children): 58.7%/71.8%
consecutive months.
$1,467
Data type: Survey conducted 12 to 18
Hours of employment:
months after exit from cash
Average hours per week: 33.9
assistance.
Employment rate:
Response rate: 72%.
At time of survey: 58.4%
California: Examining Circumstances of Individuals and Families Who Leave TANF: Assessing the Validity of Administrative Data
- 12 Month Report, December 2000
Universe: Single and two-parent cases
Earnings (family):
Health assistance:
in three counties (San Mateo, Santa
1st quarter after exit: $3,373;
Received Medi-Cal in month 12 after exit:
Clara, and Santa Cruz) closed in the
3rd quarter after exit: $3,785
59.9%
last quarter of 1998 that remained
closed for at least 2 months.
Employment rate:
Food assistance:
1st quarter after exit: 57.2%;
Received non-assistance food stamps in month
Data type: Administrative data.
3rd quarter after exit: 57.6%;
12 after exit: 5.3%
Follow-up: 12 months.
Recidivism:
Respondent receiving cash aid in month 12 after
Note: Earnings reported are for the
exit: 19.9%
family, including all earnings in the
household, rather than simply

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
earnings of the respondent as reported
by the other studies.
Universe: Single and two-parent cases
Earnings:
Health assistance:
in three counties (San Mateo, Santa
Median hourly wage rate: $8.80
Health insurance at time of survey:
Clara, and Santa Cruz) closed in the
Medi-Cal: 63.6%;
last quarter of 1998 that remained
Employment rate:
Private/Government: 27.5%
closed for at least 2 months.
At time of survey: 55.6%
Child care:
Data type: Survey conducted 6 to 12
Adult relative as primary child care
months after exit.
arrangement at time of survey: 47.6%;
Respondent has out-of-pocket child care
Response rate: 66.4%.
expenses at time of survey: 22.4%
California: Los Angeles County Post-TANF Tracking Project: Quarterly Progress Report, January 2000
Universe: Single-parent cases who
Earnings:
Not reported.
exited cash assistance in the third
1st quarter after exit: $3,414;
quarter of 1996.
4th quarter after exit: $3,576
Data type: Administrative data.
Employment rate:
1st quarter after exit: 47.2%;
Follow-up: 12 months.
4th quarter after exit: 46.6%
Note: Data are preliminary.
Colorado: Evaluation of the Colorado Works Program: First Annual Report, November 1999
Universe: Single and two-parent cases
Earnings:
Health assistance:
closed between July 1997 and
1st quarter after exit: $2,400;
Enrolled in Medicaid in the first two months
December 1998.
4th quarter after exit: $2,603
after exit (adults/children): 29%/36%
Data type: Administrative data.
Employment rate:
1st quarter after exit: 52.4%;
Follow-up: 5 quarters (15 months).
4th quarter after exit: 51.0%
Recidivism:
Cases reopened by December 1998: 18.8%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Colorado: Evaluation of the Colorado Works Program: First Annual Report, November 1999
Universe: Sample of single and two-
Earnings:
Health assistance:
parent cases closed July to September
$5.15 or less: 9.6%;
Receiving Medicaid at time of survey
of 1997 that had not reopened as of
$5.16 - $7.00: 25.6%;
(adults/children): 14.9%/43.2%
December 1998.
$7.01 or more: 64.9%
Child care:
Data type: Survey conducted
Hours of employment:
Employed respondents receiving government
approximately two years after
Less than 35 hours per week: 25.7%;
or employer child care subsidy at time of
program exit, mostly between June
35 hours or more per week: 74.3%
survey: 14%
and August 1999.
Employment rate:
Food Assistance:
Response Rate: 78%.
At time of survey: 65%;
Eligible respondents receiving food stamps at
Ever employed since exit: 89%
time of survey: 47%
Connecticut: Connecticut’s Jobs First Program: An Analysis of Welfare Leavers, December 2000
Universe: Sample of single and two-
Earnings:
Food assistance:
parent cases closed because of the 21-
1st quarter after exit: $2,966;
Ever received food stamps:
month time limit on cash assistance as
3rd quarter after exit: $3,485
1st quarter after exit: 66.7%;
of March 1998.
3rd quarter after exit: 48.8%
Employment rate:
Data type: Administrative data.
1st quarter after exit: 87.8%;
3rd quarter after exit: 77.2%
Follow-up: 9 months after exit.
Recidivism:
Ever received welfare:
1st quarter after exit: 3.3%;
3rd quarter after exit: 4.1%
Universe: Sample of single and two-
Earnings:
Food assistance:
parent cases who left Jobs First for at
1st quarter after exit: $3,369;
Ever received food stamps:
least 2 consecutive months sometime
4th quarter after exit: $3,516
1st quarter after exit: 29.5%;
in the first 18 months of random
3rd quarter after exit: 31.7%
assignment into the program.
Employment rate:
1st quarter after exit: 48.7%;
Data type: Administrative data.
3rd quarter after exit: 49.6%
Follow-up: 12 months.
Recidivism:
Ever received welfare:
1st quarter after exit: 10.6%;
3rd quarter after exit: 18.9%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Universe: Sample of single and two-
Earnings:
Health assistance:
parent cases who left Jobs First for at
Average hourly wage:
Respondent covered by Medicaid in month
least 2 consecutive months prior to the
Less than $6.00: 12.7%;
prior to interview: 59.6%
18 month interview.
$6.00 - $7.49: 13.3%;
$7.50 - $11.49: 47.9%;
Data type: Survey conducted 18
$12.00 or more: 22.3%
months after entry into Jobs First.
Employment rate:
Response rate: Not reported.
At time of survey: 69.6%
District of Columbia: The Status of TANF Leavers in the District of Columbia - Final Report, January 2001
Universe: Sample of single and two-
Earnings (Median):
Health assistance:
parent cases closed in the 4th quarter
1st quarter after exit: $3,416;
Participating in Medicaid in month 12 after
of 1997 that remained closed for at
4th quarter after exit: $3,934
exit: 37.6%
least 2 consecutive months.
Employment rate:
Food assistance:
Data type: Administrative data.
1st quarter after exit: 62.2%;
Participating in food stamps in month 12 after
4th quarter after exit: 59.8%
exit: 34.2%
Follow-up: 12 months.
Recidivism:
Ever returned to TANF within 12 months of
exit: 18.9%
Universe: Sample of single and two-
Earnings:
Health assistance:
parent cases closed in the 4th quarter
Average hourly wage: $8.74
Participating in Medicaid at the time of the
of 1998 that remained closed for at
survey (adults/children): 53.8%/60.4%;
least 2 consecutive months.
Hours of employment:
Participating in health insurance through
Average hours per week (all jobs): 35.5
employer (adults/children): 19.1%/11.9%
Data type: Survey conducted 12
months after exit.
Employment rate:
Child care assistance:
At time of survey: 60.3%
Type of arrangement among leavers with
Response rate: 61%.
children 3 and younger:
Recidivism:
Child in school: 31.7%;
Received TANF since exit: 24.6%
Licensed Registered Provider: 21.7%
Paying for child care among leavers with
children 3 and younger
Received any source of child care assistance:
30.0%;
Received child care assistance from welfare
office: 6.6%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Food assistance:
Received food stamps since exit: 55.2%
Received WIC since exit: 16.3%

Florida: Tracking the Outcomes of Welfare Reform in Florida for Three Groups of People - Leaver Sample, October 2000
Universe: Single-parent cases closed
Earnings:
Health assistance:
in the 2nd quarter of 1997 that
Average annual earnings July 1997 - June 1998:
Average percent receiving Medicaid in each
remained closed for at least 2
$6,893
month after exit (respondent/children):
consecutive months.
45.5%/52.2%
Employment rate:
Data type: Administrative data.
Average quarterly employment rate July 1997 -
Food assistance:
March 1999: 53.3%;
Average percent receiving food stamps in each
Follow-up: 21 months.
Ever employed July 1997 - June 1998: 70.4%
month after exit (respondent/children):
35.8%/41.3%
Recidivism:
Ever using cash assistance within 12 months of
exit: 26.1%;
Average percent receiving cash assistance in
each month after exit: 10.1%
Universe: Sample of single-parent
Hours of employment:
Health assistance:
cases closed in the 2nd quarter of 1997
Average hours per week:
Receiving Medicaid at time of survey: 56.9%
that remained closed for at least 2
Less then 25: 20.6%;
consecutive months.
25 - 40: 29.0%;
Child care assistance:
40 or more: 50.0%
Receiving child care assistance at time of
Data type: Survey conducted between
survey: 27.0%
May and December 1999,
Employment rate:
approximately 21 months after exit.
At time of survey: 56.7%
Food assistance:
Receiving food stamps at time of survey:
Response rate: 23.0%
48.1%
Georgia: Amended Quarterly Progress Report: Outcomes for Single Parent Leavers by Cohort Quarter for January-March 1999 –
1st Quarter 1997 Leavers, April 1999
Universe: Single-parent cases closed
Earnings:
Not reported.
in 1997 that remained closed to cash
1st quarter after exit: $2,193;
aid for at least 2 consecutive months.
4th quarter after exit: $2,389
Data Type: Administrative data.
Employment rate:
1st quarter after exit: 64.2%;
Follow-up: 12 months.
4th quarter after exit: 53.3%;
Ever employed since exit: 73.9%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Recidivism:
Returned in 4th quarter after exit: 13.4%
Georgia: Welfare Leavers Study Initial Results, December 1999
Universe: Sample of single-parent
Earnings:
Health assistance:
cases closed beginning July 1999 that
Self-reported monthly income:
Children covered by insurance at time of
remained closed to cash aid for at
Less than $500: 12%;
survey: 78%
least 2 consecutive months.
$500-$999: 58%;
$1,000-$1,499: 24%;
Data type: Survey conducted monthly
Greater than $1,499: 5%
for period July 1999 through mid-
November 1999.
Employment rate:
At time of survey: 59.0%
Response rate: 53.7%.
Recidivism:
Returned in September of 1999: 12%
Idaho: Project Self-Reliance: TAFI Participant Closure Study (II), March 1998
Universe: Sample of all cases closed
Earnings:
Health assistance:
July 1997 to December 1997.
Less than $5.25: 27%;
Receiving Medicaid at time of survey: 80%
$5.25 - $7.00: 51%;
Data type: Survey conducted in
$7.01 or more: 21%
Child care assistance:
January 1998 between 1 and 5 months
Receiving help from ICCP (government) at
after exit from cash assistance.
Hours worked per week:
time of survey: 15.2%
40 or more: 32.7%
Response rate: 16.5%.
Food assistance:
Employment rate:
Receiving food stamps at time of survey: 77%;
At time of survey: 52.9%
Receiving WIC: 41%
Illinois: Study of Former TANF Clients - Final Report, August 2000
Universe: Single and two-parent cases
Earnings:
Health assistance:
closed once for at least 2 consecutive
1st quarter after exit: $2,659;
Among cases that remained closed for 12
months during study period from July
4th quarter after exit: $2.961
months, percent who ever received Medicaid
1997 through December 1998.
within 7 months after exit: 54.8%
Employment rate:
Data type: Administrative data.
1st quarter after exit: 53.6%;
Child care assistance:
4th quarter after exit: 54.1%;
Among cases with child under the age of 13 at
Follow-up: Up to 24 months (varies).
Ever employed since exit: 69.1%
first exit, percent who received a child care
subsidy in the 1st quarter after exit: 20.3%
Recidivism:
Ever returned within 7 months of exit: 24.0%
Food assistance:

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Among cases that remained closed for 12
months, percent who ever received food stamps
within 7 months after exit: 35.3%
Among cases with child under the age of 6 at
first exit, percent receiving WIC in 1st quarter
after exit: 38.2%
Universe: Sample of single and two-
Earnings:
Health assistance:
parent cases closed in December 1998
Average hourly wage in current main job:
Some medical coverage since exit
that remained closed for at least 2
$7.89;
(respondent/children): 63.8%/71.7%
consecutive months.
Average weekly take home pay in current main
job: $274
Child care assistance:
Data type: Survey conducted in Mid-
Pay for at least one child care arrangement at
June 1999 through late August 1999.
Hours of employment:
time of survey: 43.5%
Average hours worked per week on current
Response rate: 51.3%.
main job: 36.4
Food assistance:
Received food stamps since exit: 44.1%;
Employment rate:
Receiving food stamps at time of survey:
At time of survey: 63.2%;
32.9%
Employed continually since exit: 36.6%
Received WIC after exit: 19.5%
Recidivism:
Returned to welfare at least once in 6 to 8
months since exit: 18.5%
Indiana: Who is on and who is off? Comparing Characteristics and Outcomes for Current and Former TANF Recipients, September
1997
Universe: Families who had received
Earnings:
Health assistance:
AFDC sometime between May 1995
Less than $6 hour: 39.4%;
Receiving Medicaid (families): 52.9%;
and May 1996.
$6 -7.99 hour: 40.7%;
Covered by health insurance (adults/children):
$8 or more: 19.9%
46.1% / 65.1%
Data type: Survey conducted early in
1997, 12 to 18 months after
Hours of employment:
Child care assistance:
individuals enrolled in the state’s
Full-time employment (35 or more hours per
Received child care assistance from a
welfare demonstration program.
week): 61.7%
community organization since exit: 6.4%
Response rate: Not reported.
Employment rate:
Food assistance:
At time of survey: 64.3%;
Receiving food stamps at time of survey:
Ever employed since exit: 84.3%
37.9%;
Receiving WIC at time of survey: 24.6%;
Receiving school lunch at time of survey:
46.2%;
Receiving school breakfast at time of survey:
34.7%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Iowa: Iowa’s Limited Benefit Plan (LBP), May 1997
Universe: Individuals in active
Earnings:
Health assistance:
welfare cases who, in November or
Average weekly earnings: $170.20
Receiving Medicaid at time of survey: 66.4%;
December of 1995, or January of 1996
Health insurance available from most recent
entered month seven of their initial
Hours of employment:
job held: 36.4%
assignment to the LBP.
Average number of hours worked: 31.27;
Percentage working 40 or more hours per week:
Child care assistance:
Data type: Survey conducted between
38.1%;
Child care is a type of support received from
February and April of 1996 (during
family, friends and neighbors since FIP
months 7 and 12 of the Limited
Employment rate:
benefits were terminated.
Benefit Plan).
Ever employed since exit: 52.6%
Food assistance:
Response rate: Not reported.
Received food stamps: 63.5%;
Received WIC: 29.9%;
Kansas: Statistical Summary of Leaver Survey: Twelve Month Survey Results, June 1999
Universe: Sample of all cases closed
Earnings:
Health care:
December 1997 to November 1998.
Less than $5.00 per hour: 10%; $5.01 to $7.00:
Receiving Medical Assistance (families) at
51%;
time of survey: 60%
Data type: Survey conducted 12
$7.01 or more: 33%
months after exit.
Child care assistance:
Hours of employment:
Receiving child care assistance at time of
Response rate: 51.8%.
Full-time employment (more than 37.5
survey: 6%
hours/week): 65%
Food assistance:
Employment rate:
Receiving food stamps at time of survey: 30%;
At time of survey: 36%
Recidivism:
TAF cases reopened: 18%;
Average months off cash assistance: 7
Kentucky: From Welfare to Work: Welfare Reform in Kentucky (No. 2) Second Year Survey of Discontinued K-TAP Recipients,
January 1999
Universe: Sample of all cases closed
Earnings:
Health assistance:
January to November 1997 that were
$5.15 or less per hour: 13.1%;
Receiving Medicaid at time of survey (adult):
still closed at the time of the survey.
$5.16 - $6.99: 46.1%;
27.8%;
$7.00 or more: 37.4%
Medical card or passport at time of survey
Data Type: Survey conducted October
(children): 60.4%;
1998.
Hours of employment:
Receiving other medical insurance at time of
40 or more hours per week: 59.5%
survey (adults/children): 27.0%/18.6%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Response rate: 42%.
Employment rate:
Child care assistance:
Left program for employment: 48.3%
Respondents using some type of child care at
time of survey: 37.5%;
Recidivism:
Respondents using and paying for child care at
Would return to K-TAP of possible: 18.2%
time of survey: 69.9%;
Respondents using and paying for child care
who receive payment assistance at time of
survey: 28.1%
Food assistance:
Receiving food stamps at time of survey:
45.6%;
Receiving WIC at time of survey: 24.7%
Louisiana: Exiting Welfare: The Experiences of Families in Metro New Orleans, June 1998
Universe: Sample of all cases closed
Earnings:
Health assistance:
January to March of 1998.
Average monthly income - working: $802.59;
Receiving Medicaid at time of survey: 39.8%
Average monthly income - not working:
Data type: Survey conducted in April
$403.65;
Child care assistance:
and May 1998.
Average total Monthly Household Income:
Receiving child care assistance at time of
$533.41;
survey: 3.2%
Response rate: 17.5%.
Hours of employment:
Food assistance:
Full-time (40 or more hours per week):18.7%
Receiving food stamps at time of survey:
Part-time: 14.9%
49.0%;
Receiving WIC at time of survey: 23.5%;
Employment rate:
At time of survey: 33.6%
Maryland: Life After Welfare: Fifth Report, October 2000
Universe: Sample of single and two-
Earnings:
Health assistance:
parent cases closed for any period of
1st quarter after exit: $2485;
After exit, anyone in unit received Medical
time from October 1996 to March
4th quarter after exit: $2753;
Assistance/S-CHIP in months 1-3: 50.0%;
2000.
12th quarter after exit: $3441
After exit, anyone in unit received Medical
Assistance/S-CHIP in months 25-36: 55.5%;
Data type: Administrative data.
Employment rate
1st quarter after exit: 49.2%;
Food assistance:
Follow-up: Up to 36 months (varies).
4th quarter after exit: 48.4%;
After exit, received food stamps in months 1-3:
12th quarter after exit: 51.9%
78.4%;
After exit, received food stamps in months 25-
Recidivism:
36: 42.7%
36 months after exit (excludes those who return
within 30 days): 36.0%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Massachusetts: After Time Limits: A Study of Households Leaving Welfare Between December 1998 and April 1999, November
2000
Universe: Sample of cases closed for
Earnings:
Health assistance:
reaching the time limit between
Average hourly wage: $8.21
Medicaid coverage at time of survey
December 15, 1998 and April 30,
(respondent/children): 83.7%/87.0%
1999.
Hours of employment:
Average hours per week: 31
Child care assistance:
Data type: Survey conducted October
Children less than 13 years of age in child care
1999 through March 2000.
Employment rate:
while respondent at work or school: 54%;
At time of survey: 72.6%;
Child care arrangement and receiving federal
Response rate: 75.5%.
Ever employed since exit: 90.2%
or state subsidy: 38.7%
Food assistance:
Receiving food stamps at time of survey:
52.4%;
Received food stamps ever since exit: 67.0%
Universe: Sample of cases closed for
Earnings:
Health assistance:
non-time limit reasons between
Average hourly wage: $8.62
Medicaid coverage at time of survey
December 15, 1998 and March 14,
(respondent/children): 80.0%/80.5%
1999.
Hours of employment:
Average hours per week: 34
Child care assistance:
Data type: Survey conducted October
Children less than 13 years of age in child care
1999 through March 2000.
Employment rate:
while respondent at work or school: 59.8%;
At time of survey: 70.5%;
Child care arrangement and receiving federal
Response rate: 74.4%.
Ever employed since exit: 90.5%
or state subsidy: 49.2%
Food assistance:
Receiving food stamps at time of survey:
29.0%;
Received food stamps ever since exit: 44.3%
Michigan: Longitudinal Study of Family Independence Program (FIP) Recipients: 1998 Closed FIP Cases, 1998
Universe: Cases that were receiving
Earnings:
Not reported.
cash assistance (FIP) in 1/97, but were
Average hourly wage: $7.02
closed at time of survey.
Hours of employment:
Data type: Survey conducted 6/98 -
Average hours worked per week: 31.7
9/98.
Employment rate:
Response rate: 44.5%.
At time of survey: 71.1%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Recidivism:
All of the cases included in the analysis were
closed at the time of the survey.
Mississippi: Tracking of TANF Clients: First Report of a Longitudinal Study, January 1999
Universe: Sample of all cases closed
Earnings:
Health assistance:
January to April 1998 for any length
Average hourly wage: $5.77 ($5.97 urban,
Children receiving Medicaid at time of survey:
of time.
$5.57 rural)
44%;
Children receiving private insurance at time of
Data type: Survey conducted July to
Hours of employment:
survey: 10%
September 1998.
Average Hours per week: 35;
Full-time (35 or more hours per week): 67%
Child care assistance:
Response rate: 86.7%.
(75% urban; 59% rural);
Respondents using child care in the last 6
Part-time: 33% (25% urban ; 41% rural)
months who received state sponsored
payments: 56%;
Employment rate:
At time of survey: 35% (27% rural, 46% urban); Food assistance:
Ever Employed in last 6 months: 47% (52%
Receiving food stamps at time of survey: 58%
urban; 43% rural)
(47% urban, 67% rural);
Participate in school lunch program at time of
Recidivism:
survey (of families with school age children):
Received TANF, Medicaid, and food stamp
82%
benefits at least one month since originally
leaving the rolls: 47% (37% urban; 73% rural)
Missouri: Former AFDC Recipients in Missouri - Chapters 1-3, June 2000
Universe: Sample of single and two-
Earnings:
Health assistance:
parent AFDC/TANF cases (those
Average hourly wage: $6.00
Receiving Medicaid at time of survey
receiving a cash payment) who were
(adult/children): 38%/65%
active during September, October,
Employment rate:
and/or November 1996, and who left
At time of survey: 65%;
Child care assistance:
AFDC during the fourth quarter of
Not employed at survey, but had worked since
Percent of eligible population who received a
1996 and remained off the rolls for at
exit: 25%
subsidy at time of survey (administrative data):
least 2 consecutive months.
28%;
Hours of Employment:
Percent of eligible population who received a
Data type: Survey conducted between
Average hours worked per week: 39
subsidy since leaving cash assistance
March 15 and August 31, 1999
(administrative data): 55%
(Supplemented with administrative
Recidivism:
data records where noted),
Received cash assistance since exit: 31%
Food assistance:
approximately 30 months after exit.
(survey); 44% (administrative data); 50%
Received food stamps in last month: 47%;
(either source).
Received food stamps any time since exit
Response rate: 74.5%.
Receiving cash assistance at time of survey:
(administrative data): 80%;

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
14%
Received food stamps any time since exit
(either data source): 83%
Received WIC when surveyed: 23%
Missouri: Preliminary Outcomes for 1996 Fourth Quarter AFDC Leavers, September 1999
Universe: All single and two-parent
Earnings:
Health assistance:
AFDC/TANF cases (those receiving a
1st quarter after exit: $2,192;
Receiving Medicaid at any time during the 8
cash payment) who were active during
8th quarter after exit: $3,055
quarter follow-up period (self/children):
September, October, and/or November
48.0%/59.2%
1996, and who left AFDC during the
Employment rate:
fourth quarter of 1996 and remained
1st quarter after exit: 58.1%
Food assistance:
off the rolls for at least 2 consecutive
8th quarter after exit: 56.7%
Receiving food stamps at any time during the 8
months.
Ever employed since exit: 79.5%
quarter follow-up period: 74.9%
Data type: Administrative data.
Recidivism:
Returned to cash assistance at any time during
Follow-up: 2 years (8 quarters).
the 8 quarter follow-up period: 36.2%
Montana: Montana’s Welfare Reform Project: FAIM February 1998 Update, February 1998
Universe: All cases closed that
Earnings:
Not reported.
remained closed from March 1996 to
Over the 18-month period: $5,908
September 1997.
Employment rate:
Data Type: Administrative data.
Reported employment earnings: 57%
Follow-up: 18 months.
Montana: Montana’s Welfare Reform Project: FAIM February 1998 Update, February 1998
Universe: Current and former
Earnings:
Health assistance:
recipients of welfare at time of survey.
Percent earning less than minimum wage: 11%
Receiving Medicaid at time of survey 74.5%;
Data type: Survey conducted
Hours of employment:
Child care assistance:
December 1997 to January 1998.
More than 20 hours per week: 47%
29 survey respondents were enrolled in Job
Supplement Program (JSP), with 6 utilizing
Response rate: 208 respondents
Employment rate:
publicly funded child care at time of survey
(response rate not provided).
At time of survey: 66%
Food assistance:
Receiving food stamps at time of survey: 63%;
Receiving School Lunch at time of survey:
55.3%;
Receiving WIC at time of survey: 47.6%;

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
New Jersey: WFNJ Evaluation: Current and Former WFNJ Clients: How are they Faring 30 Months Later: Leaver Sample,
November 2000
Universe: Sample of all cases who
Earnings:
Health assistance:
entered the program from July 1997 to
Average Total Household Monthly Income:
Public health insurance at time of survey
December 1998, a subsample of which
$1,429;
(respondent/children): 46%/59%;
have left (the leaver sample).
Average Total Household Monthly Income
Private health insurance at time of survey
among employed leavers: $1,832
(respondent/children): 17%/14%
Data type: Survey conducted in spring
Employment rate:
Child care:
2000.
Among those who had left 24 months after
Percent of employed former clients were
WFNJ entry : 59.7%
receiving government child care subsidies:
Response rate: 80%.
27%
Food Assistance:
Eligible for and receiving food stamps: 50%
New Mexico: Survey of Current and Former Welfare Recipients, 2000
Universe: Sample of all cases that
Earnings:
Health assistance:
were receiving welfare benefits in
Less than $5 per hour: 9.1%;
Received Medicaid for self or children at time
New Mexico at some time between
$5.00 - $6.99 per hour: 48.7%;
of survey: 58.4%
July 1998 and June 1999 (results
At least $7.00 per hour: 42.2%
reported here are for those who left
Child care assistance:
welfare during this period).
Hours of employment:
Pay for child care and receive help from county
Less than 20 hours per week: 5.7%;
at time of survey : 33%
Data type: Survey began in November
21-39 hours per week: 40.5%;
1999.
At least 40 hours per week: 53.9%
Food assistance:
Received food stamps at time of survey: 38.5%;
Response rate: 72.5%.
Employment rate:
Received WIC at time of survey: 23.1%
At time of survey: 62.5%;
Not currently working, but had worked since
leaving welfare: 51%
Recidivism:
Previously left welfare but went back: 57.0%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
New York: After Welfare: A Study of Work and Benefit Use After Case Closing, December 1999
Universe: Single and two-parent cases
Earnings:
Health assistance:
closed January to March 1997 that
1st quarter after exit: $3,868;
Receiving Medicaid in the 4th quarter after case
were closed for at least 2 months.
4th quarter after exit: $4,230
closing (all/adults/children): 45%/32%/35%
Data type: Administrative data.
Employment rate:
Food assistance:
1st quarter after exit: 55%;
Receiving food stamps in the 4th quarter after
Follow-up: 12 months.
4th quarter after exit: 53%;
case closing (all/adults/children):
Ever employed since exit: 66%
29%/24%/29%
Recidivism:
Ever returned to welfare by month 12 of follow-
up: 21.1%
New York: Leaving Welfare: Findings from a Survey of Former New York City Welfare Recipients, September 1998
Universe: Sample of TANF cases
Earnings:
Health assistance:
closed in November 1997.
Hourly wage rates below $5.75: 10%;
Employed and receiving insurance through
$5.75 - $7.00: 30%;
Medicaid at time of survey: 8%
Data type: Survey conducted April
Over $7.00: 49%
Employed and receiving insurance through
1998 through July 1998,
work and Medicaid at time of survey: 6%
approximately 6 months after exit.
Hours of employment:
Less than 30 hours worked per week: 39%;
Response rate: 22%; However rate
30 - 40: 26%;
increases to 60% among those with
40 or more: 29%
current and accurate telephone
numbers.
Employment rate:
Ever employed since exit: 67%
Recidivism:
Returned to welfare since exit: 15%
North Carolina: Evaluation of the North Carolina Work First Program: Study of Families Leaving Work First in Selected
Counties, May 2000
Universe: Sample of single and two-
Earnings:
Health assistance:
parent cases who had left Work First
Hourly wage rates at primary job:
Receiving Medicaid at the time of the survey
for any reason between December
Less than $7.00: 38.7%;
(overall/children): 66.7%/70.8%
1998 and April 1999 for at least one
$7.00 - $7.99: 23.0%;
month in eight selected counties.
$8.00 or more: 38.3%
Child care assistance:
Respondents with children under 18 who
Data type: Survey conducted between
Employment rate:
reported that they used paid or unpaid child
February 1999 and June 2000.
Employed at time of survey: 64.1%
care at time of survey: 59.7%;

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Response rate: 70.1%.
Recidivism:
Respondents who use paid child care who
Returned to welfare the time of the survey:
receive payment assistance from the county at
17.3%
time of survey: 52.8%
Food assistance:
Receiving food stamps at the time of the
survey: 45.0%;
Receiving WIC at the time of the survey:
23.9%
North Carolina: Evaluation of the North Carolina Work First Program: Analysis of Administrative Data on Families Receiving
Welfare and Diversion Assistance - Fourth Quarterly Report, May 2000
AFDC exit cohort -
Earnings:
Food assistance:
Universe: All families who
1st quarter after exit: $2,075;
Receiving food stamps:
received a cash assistance
4th quarter after exit: $2,211;
12 months after exit: 33.2%;
payment in January 1995 but
12th quarter after exit: $2,524;
24 months after exit: 29.4%;
did not receive a check in
18th quarter after exit: $3,210
36 months after exit: 27.9%
February 1995.
Employment rate:
Data type: Administrative data.
1st quarter after exit: 65.7%;
4th quarter after exit: 61.6%;
Follow-up: 39 months.
12th quarter after exit: 62.4%;
18th quarter after exit: 59.8%
Recidivism:
Receiving welfare in month 6 after exit: 21.2%;
Receiving welfare in month 18 after exit:
18.8%;
Returned to welfare in the first 18 months after
exit: 37.6%
September 1996 exit cohort -
Earnings:
Food assistance:
Universe: All families who
1st quarter after exit: $2,251;
Receiving food stamps:
received a cash assistance
4th quarter after exit: $2,275;
12 months after exit: 32.6%;
payment in August 1996 but did
12th quarter after exit: $2,843
24 months after exit: 29.0%;
not receive a check in
36 months after exit: 31.1%
September 1996.
Employment rate:
1st quarter after exit: 68.1%;
Data type: Administrative data.
4th quarter after exit: 65.2%;
12th quarter after exit: 59.8%
Follow-up: 39 months.
Recidivism:
Receiving welfare in month 6 after exit: 18.0%;

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Receiving welfare in month 18 after exit:
14.5%;
Returned to welfare in the first 18 months after
exit: 33.1%
June 1997 exit cohort -
Earnings:
Food assistance:
Universe: All families who
1st quarter after exit: $2,135;
Receiving food stamps:
received a cash assistance
4th quarter after exit: $2,374;
12 months after exit: 38.2%;
payment in May 1997 but did
8h quarter after exit: $2,568
24 months after exit: 31.7%
not receive a check in June
1997.
Employment rate:
1st quarter after exit: 70.8%;
Data type: Administrative data.
4th quarter after exit: 65.6%;
8th quarter after exit: 62.2%
Follow-up: 30 months.
Recidivism:
Receiving welfare in month 6 after exit: 18.7%;
Receiving welfare in month 18 after exit:
12.3%;
Returned to welfare in the first 18 months after
exit: 32.7%
June 1998 exit cohort -
Earnings:
Food assistance:
Universe: All families who
1st quarter after exit: $1,999;
Receiving food stamps:
received a cash assistance
4th quarter after exit: $2,119
12 months after exit: 39.7%
payment in May 1998 but did
not receive a check in June
Employment rate:
1998.
1st quarter after exit: 69.7%;
4th quarter after exit: 62.7%
Data type: Administrative.
Recidivism:
Follow-up: 18 months.
Receiving welfare in month 6 after exit: 15.5%;
Receiving welfare in month 18 after exit:
11.2%;
Returned to welfare in the first 18 months after
exit: 30.0%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
North Carolina: Evaluation of the North Carolina Work First Program: Study of Families Leaving Work First due to Time Limits -
Results from the Second Follow-up Surveys, February 2000
Universe: All cases closed in August
Earnings:
Health care:
1998 because of a 24-month state time
$5.00 - $6.99 per hour: 55.3%;
Receiving Medicaid at time of survey: 75.6%
limit (92% had received cash welfare
$7.00 - $8.99: 19.3%;
for at least 40 of the 43 past months.
$9.00 or more: 25.4%
Child care assistance:
Receiving a child care subsidy at time of
Data type: Survey conducted between
Hours of employment:
survey: 7%
September and December of 1999
20 to 39 hours per week: 36.2%;
(supplemented with administrative
40 or more: 53.2%
Food assistance:
data), approximately 13 to 16 months
Receiving food stamps at time of survey:
after reaching the time limit.
Employment rate:
58.8%;
At time of survey: 65.6%
Receiving WIC at time of survey: 5.4%
Response rate: 70.2% of original
sample (89.5% of respondents to first
Recidivism:
survey).
All cases closed because of the time limit are
ineligible for TANF cash benefits for 3 years.
Ohio: How Are They Managing? A Six Month Retrospective of Cuyahoga County Families Leaving Welfare, 1999
Universe: Sample of all families who
Earnings:
Health assistance:
exit cash assistance in the 4th quarter
Total Family Income (including cash value of
Receiving Medicaid at time of survey
of 1998 and 1st quarter of 1999 for at
food stamps) greater than or equal to 100% of
(adults/children): 59% / 69%;
least 2 consecutive months.
1998 Poverty Level: 55%
Receiving Private Health Insurance at time of
survey(adults/children): 17% / 15%;
Data type: Survey conducted 6
months after exit.
Employment rate:
Child care assistance:
At time of survey: 67%;
Response rate: 70%.
Ever employed since exit: 87%
Receiving child care assistance at time of
survey: 22%
Recidivism:
Returned at least once: 24%
Food assistance:
Receiving food stamps at time of survey: 53%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Ohio: Employment and Returns to Public Assistance Among Single, Female-Headed Families Leaving AFDC in Third Quarter,
1996 in Cuyahoga County, May 1999 (revised September 2000)
Universe: Single, female headed
Earnings:
Health assistance:
families who exited cash assistance in
1st quarter after exit: $2,756;
Received Medicaid:
the 3rd quarter of 1996 for at least 2
4th quarter after exit: $2,952
1st quarter after exit: 41.4%;
consecutive months.
4th quarter after exit: 37.7%
Employment rate:
Data type: Administrative data.
1st quarter after exit: 59.3%;
Food assistance:
4th quarter after exit: 56.8%
Received food stamps:
Follow-up: 12 months.
1st quarter after exit: 42.5%;
Recidivism:
4th quarter after exit: 39.4%
Returned to cash assistance within 12 months
after exit (adults/children): 35%/37%
Oklahoma: Family Health and Well-Being in Oklahoma: An Exploratory Analysis of TANF Cases Closed and Denied October
1996-November 1997: Sample of Adult and Children, September 1998
Universe: Sample of all cases that
Earnings:
Health assistance:
were closed or denied for any reason
Average hourly wage: $6.51
Receiving Medicaid only at time of survey
from October 1996 to November 1997
(adult/children): 60.6%/62.6%;
regardless of whether they were
Hours of employment:
Receiving Private (work-related) & Medicaid
receiving TANF at the time of the
Average hours worked per week: 34
at time of survey (adult/children): 9.2%/18.5%;
survey.
Full-time employment (40 or more hours per
week): 50.3%;
Child care assistance:
Data type: Survey conducted January
Eligible households receiving child care
1998 through April 1998.
Employment rate:
assistance: 33.9%
At time of survey: 50.3%;
Response rate: 53.3%.
Respondents who live in households with at
Food assistance:
least one employed individual: 68.5%
Receiving food stamps only: 8.9%;
Receiving WIC/other (including Food Banks),
Recidivism:
only: 25.7%
Received TANF at least once since October
Receiving food stamps and WIC: 40.1%;
1996: 33.6%
Oregon: Oregon Families Who Left Temporary Assistance to Needy Families (TANF) or Food Stamps: A Study of Economic and
Family Well-being from 1998 to 2000, Volume 1, January 2001
Universe: Sample of families who left
Earnings:
Not reported.
in the 1st quarter of 1998 who didn’t
Average monthly earnings: $1,016.32
return fo two consecutive months.
Hours of employment:
Data type: Survey conducted October
Hours per week:
1999 through December 1999, 18 to
Less than 20: 9.7%;

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
21 months after exit, supplemented
20 - 39: 36.5%;
with administrative data.
40 or more: 53.8%
Response rate: 75.8%.
Employment rate:
This response rate refers to the second
At time of survey: 71%
survey, for which results are reported.
The response rate for the first survey
was 80.9%, with 93.7% of these
respondents completing a second
survey.
Pennsylvania: Welfare Reform After Two Years: Technical Report on Former Welfare Recipients in Pennsylvania (March 1997 -
March 1999), November 1999
Universe: Single and two-parent cases
Earnings:
Health assistance:
that were closed March 1997 to
Annualized income:
Received Medicaid for some period of time
December 1998 that remained closed.
4th quarter after exit: $11,711;
since exit: 76%
Food stamp and Medicaid
8th quarter after exit: $13,759
Participation data is reported from
Food assistance:
March 1997 to April 1999.
Employment rate:
Received food stamps for some period of time
Ever employed since exit: 68%
since exit: 65%
Data type: Administrative Data.
Recidivism: Only cases that remained closed
Follow-up: Up to 8 quarters (varies).
were included in the sample.
Pennsylvania: Welfare Reform After Two Years: Technical Report on Former Welfare Recipients in Pennsylvania (March 1997 -
March 1999), November 1999
Universe: Single and two-parent
Employment rate:
Health assistance:
cases that were closed February to
At time of survey: 58.4%;
Receiving Medicaid at time of survey
November of 1998 that remained
Ever employed since exit: 80.8%
(households): 81.6%;
closed for 6 consecutive months.
Receiving Private insurance at time of survey:
Recidivism:
27.2%
Data type: Survey.
All of the cases in the sample were still closed at
the time of the survey.
Child care assistance:
Response rate: 29%.
Respondents with extended/after-school care
for children who receive state child care
payment assistance at time of survey: 42.1%
Respondents with preschool children and a
caretaker who receive state child care payment
assistance at time of survey: 64.4%
Food assistance:
Receiving food stamps at time of survey:

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
51.2%;
Receiving WIC at time of survey: 26.4%;
Receiving School lunch program at time of
survey: 47.2%;
Receiving summer feeding program at time of
survey: 25.6%
Rhode Island: Where Are They Now? A Post-Cash Administrative Data Study, May 1999
Universe: All cases closed in April
Recidivism:
Not reported.
1997 and April 1998 that remained
Cases Re-opened Within the Year:
closed for at least 31 days.
April 1997: 29.9%;
April 1998: 30.5%
Data type: Administrative data.
Follow-up: 12 months.
South Carolina: Survey of Former Family Independence Program Clients: Cases Closed During January through March, 1998,
June 1999
Universe: Sample of all cases with at
Earnings:
Health assistance:
least one adult subject to FIP work
Average hourly wage: $7.00
At least one household member covered by
requirements closed January to March
Medicaid at time of survey: 77.6%;
of 1998 that remained closed up to the
Hours of employment:
Receiving private insurance at time of survey
time of the survey.
Average hours worked per week: 35.1; Work
(adults/children): 7.2%; 7.4%
30 or more hours per week: 81.4%
Data type: Survey conducted January
Child care assistance:
1999 through April 1999, 12 to 15
Employment rate:
Receiving assistance at time of survey (all
months after exit from cash
At time of survey: 54.7%;
households): 9.9%;
assistance.
Ever employed since exit: 80.7%
Among those using care at time of survey, 30%
receive subsidy for preschool; 27.4% for
Response rate: 75.0%.
Recidivism:
school-aged
Pretty sure they will not need welfare again:
51.0%
Food assistance:
Receiving food stamps at time of survey:
52.6%;
Receiving school lunch program at time of
survey: 52.1%;
Receiving WIC at time of survey: 27.1%;
Receiving Summer feeding program at time of
survey: 4.9%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Tennessee: Summary of Surveys of Welfare Recipients Employed or Sanctioned for Non-Compliance, March 1998
Universe: Sample of cases
Earnings:
Health assistance:
sanctioned/closed for noncompliance
Average hourly wage: $5.50
Receiving TennCare at time of survey: 88%
with Families First program
requirements since January 1997.
Employment rate:
Child care assistance:
At time of survey: 39%
Receive child care assistance at time of survey:
Data type: Survey with data collected
16%
through October 1997.
Recidivism:
Think they will return to Families First: 47%;
Response Rate: 56%.
10% report receiving other benefit checks to
help pay the bills - listed source of benefit
checks is AFDC, so some of these 10% may be
back on public assistance.
Universe: Sample of cases
Hours of employment:
Health assistance:
sanctioned/closed fo their refusal to
Full-time: 25%;
Receiving TennCare at time of survey: 91%
sign a Personal Responsibility Plan
Part-time: 17%
(PRP) since January 1997.
Child care assistance:
Employment rate:
Receive child care assistance at time of survey
Data type: Survey with data collected
At time of survey: 42%
March 1, 2001: 21%
through October 1997.
Recidivism:
Response rate: 59%.
Think they will return to Families First: 25%
Texas: The Impacts of Welfare Reform Changes in Texas: Early Findings, December 1998
Universe: Sample of cases closed in
Earnings:
Health assistance:
November 1997 that remained off
Average hourly wage: $6.28 ($6.15 for
Received Medicaid since exit
TANF for 6 months, as well as a
Redirects; $6.35 for Leavers)
(leavers/redirected): 74%/ 66%
sample of cases redirected from TANF
through Texas Works in December of
Hours of employment:
Food assistance:
1997.
Average hours per week: 34 (34.1 hours/week
Received WIC since exit (leavers/redirected):
for leavers & 34.7 for redirects)
33% / 42%;
Data type: Survey, supplemented with
administrative data.
Employment rate:
At time of survey: 55% (59% of leavers & 50%
Response rate: 51%.
of those redirected);
Ever employed since exit: 68% (76% of leavers
& 62% of those redirected).

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Texas: Why People Leave Welfare II: An Expanded Follow-up Study of the Effects of Welfare Reform, December 1998
Universe: Sample of single and two-
Earnings:
Health assistance:
parent cases closed November 1997 to
Average monthly salary: $771
Receiving Medicaid at time of survey: 62%
February 1998.
Employment rate:
Child care:
Data type: Survey.
Percent who left cash assistance for
Receiving child care assistance at time of
employment: 48%
survey: 12%
Response rate: Ranged from 54% to
66%.
Hours of employment (among the 48% who left
Food assistance:
cash assistance for employment):
Receiving food stamps at time of survey: 56%
Percent employed full-time: 75%
Utah: Multiple Impacts of Welfare Reform in Utah: Experiences of Former Long-term Welfare Recipients, June 2000
Universe: Sample of all cases who
Earnings:
Health assistance:
had received cash assistance for at
Average monthly income: $1,092
Receiving Medicaid at time of survey: 83%;
least 36 months whose cases had been
Received Medicaid since exit: 88%
closed for at least 2 consecutive
Hours of employment:
months whose reason for case closing
Average hours per week: 35
Child care:
was “income.”
Receiving child care assistance at time of
Employment rate:
survey: 34%;
Data type: Survey conducted from
At time of survey: 77%
Received child care assistance since exit: 37%
July 1999 to January 2000.
Food assistance:
Response rate: 90%
Receiving food stamps at time of survey: 56%;
Received food stamps since exit: 64%
Universe: Cases who had received
Earnings:
Health assistance:
cash assistance for at least 36 months
Average monthly income: $750
Receiving Medicaid at time of survey: 80%;
whose cases had been closed for at
Received Medicaid since exit: 81%
least 2 consecutive months whose
Hours of employment:
reason for case closing was “time
Average hours per week: 27
Child care:
limit.”
Receiving child care assistance at time of
Employment rate:
survey: 13%;
Data type: Survey conducted from
At time of survey: 43%
Received Child care assistance since exit: 14%
February 2000 to May 2000.
Food assistance:
Response rate: 69%.
Receiving food stamps at time of survey: 85%;
Received food stamps since exit: 85%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Universe: Cases who had received
Earnings:
Health assistance:
cash assistance for at least 36 months
Average monthly income: $814
Receiving Medicaid at time of survey: 64%;
whose cases had been closed for at
Received Medicaid since exit: 66%
least 2 consecutive months whose
Hours of employment:
reason for case closing was “other.”
Average hours per week: 30
Child care:
Receiving child care assistance at time of
Data type: Survey conducted from
Employment rate:
survey: 11%;
July 1999 to January 2000.
At time of survey: 42%
Received Child care assistance since exit: 13%
Response rate: 61%.
Food assistance:
Receiving food stamps at time of survey: 62%;
Received food stamps since exit: 63%
Virginia: Experiences of Virginia Time-Limit Families in the Six Months After Case Closure: Results from an Early Cohort, Final
Report, November 1999
Universe: All cases closed February
Earnings:
Health assistance:
through June 1999 due to the state
Average hourly wage: $5.99
Among working, percent who enrolled in
time-limit.
employer-provided health benefits: 28.8%
Hours of employment:
Data type: Survey conducted 6 to 12
Average hours per week: 33.5
Child care:
months after exit from cash
Among working or in training, percent who
assistance.
Employment Rate:
receive at least some government assistance:
At time of survey: 71%;
28.7%;
Response rate: 78%.
Ever employed since exit: 85.5%
Among working or in training, percent who
were aware of government assistance but chose
Recidivism:
not to participate: 34.7%
Percent receiving income from TANF in month
before interview: 2.4%
Food assistance:
Percent receiving income from food stamp
program in month before interview: 76.1%
Washington: A Study of Washington State’s TANF Leavers and TANF Recipients - Welfare Reform and Findings from
Administrative Data Final Report: February 2000
Cohort I:
Universe: Single and two-
Earnings:
Child care assistance:
parent cases closed in 4th
1st quarter after exit: $2,778;
Receiving state subsidized child care:
quarter 1996 that remained
4th quarter after exit: $3,340;
Quarter of exit: 11%
closed to cash aid for at least
8th quarter after exit: $3,701
1st quarter after exit: 10%;
2 consecutive months.
4th quarter after exit: 8%;
Employment rate:
8th quarter after exit: 7%
Data type: Administrative
1st quarter after exit: 51%;
data.
4th quarter after exit: 56%;
8th quarter after exit: 57%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Follow-up: 24 months.
Recidivism:
Ever returned in the four quarters after exit:
Note: Health and Food
28.8%
Assistance data is not
available for Cohort I.
Cohort II:
Universe: Single and two-
Earnings:
Health assistance:
parent cases closed in 4th
1st quarter after exit: $2,714;
Received Medical Assistance:
quarter 1997 that remained
4th quarter after exit: $3,351
Quarter of exit: 99%;
closed to cash aid for at least
1st quarter after exit: 53%;
2 consecutive months.
Employment rate:
4th quarter after exit: 43%
1st quarter after exit: 57%;
Data type: Administrative
4th quarter after exit: 58%
Child care assistance:
data.
Receiving state subsidized child care:
Recidivism:
Quarter of exit: 14%;
Follow-up: 18 months.
Ever returned in the four quarters after exit:
1st quarter after exit: 11%;
23.2%
4th quarter after exit: 9%
Food assistance:
Received food stamps:
Quarter of exit: 91%;
1st quarter of exit: 46%;
4th quarter of exit: 36%
Cohort III:
Universe: Single -parent
Earnings:
Health assistance:
cases closed in 4th quarter
1st quarter after exit: $2,387
Received Medical Assistance:
1998 that remained closed to
Quarter of exit: 98%;
cash aid for at least 2
Employment rate:
1st quarter after exit: 60%
consecutive months.
1st quarter after exit: 62%
Child care assistance:
Data type: Administrative
Recidivism:
Received state subsidized child care:
data.
Ever returned in the two quarters after exit:
Quarter of exit: 18%;
17.1%
1st quarter after exit: 14%
Follow-up: 6 months
Food assistance:
Received food stamps:
Quarter of exit: 90%;
1st quarter of exit: 47%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Washington: A Study of Washington State’s TANF Leavers and TANF Recipients - Findings from the April-June 1999 Telephone
Survey Final Report: March 2000
Universe: Sample of single and two-
Earnings:
Health assistance:
parent cases closed in October 1998
Average hourly wage: $7.70
Receiving Medicaid, medical coupons, healthy
that remained closed for at least 2
options at time of survey (adults/children):
consecutive months.
Hours of employment:
53%/77%
Average number of hours worked per week: 36;
Data type: Survey conducted April
Percent working full-time (35 or more hours)
Child care assistance:
through June 1999.
per week: 69%
State-subsidized child care among those
parents who had childcare arrangements at
Response rate: 72%.
Employment rate:
time of survey: 39%
At time of survey: 59%
Food assistance:
Recidivism:
Received food stamps within the last six
Returned to TANF at the time of the survey:
months: 53%
19%
Wisconsin: Post-Exit Earnings and Benefit Receipt among Those Who Left AFDC in Wisconsin, January 1999
Cohort I:
Universe: Single mothers
Earnings:
Food assistance:
who left welfare in the last
1st quarter after exit: $2,545;
Ever received food stamps in year after exit:
quarter of 1995 for at least 2
4th quarter after exit: $2,940
57.4%;
consecutive months followed
Ever received food stamps in 3 years after exit:
through December 1998.
Employment rate:
65.8%
1st quarter after exit: 69.0%;
Data type: Administrative
4th quarter after exit: 68.7%;
Data.
Ever employed in first year after exit: 81.1%;
Ever employed over 3 years after exit: 87.8%
Follow-up: 36 months (3
years).
Recidivism:
Ever received TANF in year after exit: 29.0%;
Ever received TANF in 3 years after exit: 34.4%
Cohort II:
Universe: Single mothers
Earnings:
Food assistance:
who left welfare in the last
1st quarter after exit: $2,081;
Ever received food stamps in year after exit:
quarter of 1997 for at least 2
4th quarter after exit: $2,744
80.6%
consecutive months followed
through December 1998.
Employment rate:
1st quarter after exit: 69.6%;
Data type: Administrative
4th quarter after exit: 68.1%;
Data.
Ever employed in first year after exit: 83.9%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Follow-up: 12 months.
Recidivism:
Ever received TANF in year after exit: 24.5%
Wisconsin: Post-Exit Earnings and Benefit Receipt among Those Who Left AFDC in Wisconsin, January 1999
Universe: Cases closed for at least 2
Earnings:
Health assistance (5th quarter after exit):
consecutive months between August
1st quarter after exit: $2,440;
Receiving Medicaid only: 23.4%;
1995 and July 1996.
5th quarter after exit: $2,751
Receiving Medicaid and food stamps: 26.0%
Data type: Administrative data.
Employment rate:
Food assistance (5th quarter after exit):
1st quarter after exit: 72.4%;
Receiving food stamps only: 2.3%;
Follow-up: 15 months.
5th quarter after exit: 75.8%
Receiving Medicaid and food stamps: 26.0%
Ever employed in year following exit: 81.7%
Recidivism:
Returned to AFDC within 15 months of leaving:
30%
Wisconsin: Survey of Those Leaving AFDC or W-2 January to March 1998: Preliminary Report, January 1999
Universe: Cases closed between
Earnings:
Health assistance:
January and March 1998, that
Employed at time of interview - hourly wage
Receiving Medicaid or Healthy Start Received
remained closed at the time of the
$7.42;
at time of survey: 87%;
survey.
Ever employed - hourly wage: $7.28
Private Insurance at time of survey: 26%
Data type: Survey conducted August
Hours of employment:
Child care assistance:
to November of 1998, 6 to 9 months
Mean hours per week at best job for those
Receiving child care assistance at time of
after exit from cash assistance.
currently employed: 36;
survey: 17%
Mean hours per week at best job for those who
Response rate: 69%.
had worked since their exit but are currently
Food assistance:
unemployed: 33
Receiving food stamps at time of survey: 49%;
Receiving WIC at time of survey: 38%;
Employment rate:
Receiving School lunch program at time of
At time of survey: 62%
survey: 47%;
Ever employed since exit: 83%
Receiving summer feeding program for
children at time of survey: 5%
Recidivism: Agree or strongly agree that they
are pretty sure they will not need to be on
welfare again: 60%

---

Study
Earnings/Recidivism
Health Care/ Child Care/ Food
Wyoming: A Survey of Former POWER Recipients, May 1998
Universe: Sample of all cases closed
Earnings: (all respondents):
Health assistance:
January 1997 to March 1998.
Less than $5/hour: 3.5%;
Receiving Medicaid at time of survey
5.01 to $7.50: 47%;
(families): 82.5%
Data type: Survey conducted in
$7.51 or more: 9.0%
March 1998.
Child care assistance:
Hours of employment:
Receive public assistance day care at time of
Response rate: 31.8%.
40 or more hours: 47%
survey: 28%
Employment rate:
Food assistance:
At time of survey: 61%
Receiving food stamps at time of survey: 61%
Recidivism:
Back on POWER at time of survey: 2.5%
Source: Table prepared by the Congressional Research Service (CRS) based on information compiled from the state report evaluations and Isaacs
and Lyon, 2000.

---