Texas Families in Transition/
Surviving without TANF:

An Analysis of Families Diverted From or Leaving TANF


A joint project of

Deanna Schexnayder

Daniel Schroeder

Ray Marshall Center
for the Study of Human Resources

Lyndon B. Johnson School of Public Affairs

The University of Texas at Austin

Laura Lein

David Dominguez

Center for Social Work Research

School of Social Work

The University of Texas at Austin

Karen Douglas

Freddie Richards

Center for Innovative Projects
for Economic Development

Prairie View A&M University

January 2002

Executive Summary


In the post-welfare-reform era, many states have begun conducting research to determine how the new policies affect the families they serve.  In particular, states need to understand if former welfare recipients are employed or receiving other types of economic supports, how many have returned to welfare, and reasons for families’ success or failure. 

In 1999, the state of Texas received federal funding from the U. S. Department of Health and Human Services (HHS) to study outcomes for families diverted from TANF at application.  The Texas Department of Human Services (DHS), which was already sponsoring research on families who had been redirected from or left TANF after welfare reform, combined the resources and research approaches of the HHS grant with those of the state-funded research already in progress.  The resulting project, Texas Families in Transition/Surviving Without TANF, provides the most comprehensive look to date at Texas families who have left or been diverted from TANF.  While no one approach can fully assess the effects of welfare reform on poor families, the multiple approaches used in this project provide a more complete picture of how low-income families in Texas are responding to changes enacted as a result of various welfare reform initiatives.

DHS contracted with the Ray Marshall Center for the Study of Human Resources (UTRMC) and the Center for Social Work Research (UTCSWR) at The University of Texas at Austin to conduct this combined research.  UTCSWR sub-contracted with the Center for Innovative Projects for Economic Development at Prairie View A&M University, another partner in this project.

Policy Context

In 1995, the Texas legislature passed its major welfare reform legislation, HB 1863, which established time limits, modified eligibility requirements, and enacted a personal responsibility agreement for recipients of Aid to Families with Dependent Children (AFDC).  The bill also authorized the receipt of a one-time lump sum payment in lieu of receiving monthly welfare cash grants.  As required by federal law at that time, Texas received a waiver from existing federal regulations in order to implement HB 1863.  This waiver, known as Achieving Change for Texans (ACT), remains in effect through March 2002.  HB 1863 also consolidated a number of workforce programs under a new agency, the Texas Workforce Commission (TWC) and devolved the responsibility for management of these programs to 28 local workforce boards.

The Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), passed by Congress in 1996, replaced the AFDC program with a new cash assistance program, Temporary Assistance for Needy Families (TANF).  This legislation includes mandates regarding the TANF program, including a lifetime limit of 60 months on TANF.  States also are required to meet higher work participation rates than were previously required, and fewer exemptions may be granted.  Because Texas already had federal approval to implement ACT before PRWORA went in effect, some federal welfare provisions will not apply to Texas until the state’s waiver expires in March 2002. 

Since the implementation of PRWORA, welfare reform in Texas has continued to evolve.  The following key initiatives, enacted either by the Texas Legislature or agency policy, are relevant to the time period covered by the current research:

·         Texas Works, implemented by DHS on November 1, 1997, which communicates to TANF applicants and recipients that employment is both the goal and the expectation for families receiving TANF, and includes informal diversion (redirection) activities prior to TANF application.

·         Work First, implemented by TWC and DHS on December 1, 1997.  As part of this program, most TANF applicants must attend a workforce orientation session prior to approval of their TANF application.

·         Expansion of the Earned Income Disregard for TANF recipients, enacted by the Texas Legislature in its 1999 session.

·         Restriction of the ‘age of child’ exemption to ACT waiver workforce services participation requirements (and also to Texas time limits) for TANF caretakers with young children, also enacted by the Texas Legislature in 1999.

Research Questions and Populations Being Studied

The research addresses the following research questions:

·         What are the characteristics of families who left or were diverted from TANF?

·         To what extent are these families participating in other government programs, especially Medicaid and food stamps?

·         To what extent are these families employed and/or receiving other economic supports, such as child support and child care subsidies?

·         Over time, how do these families manage and what hardships do they face?

·         How do potential applicants view the diversion/application process?

·         Are there particular points after leaving TANF at which people are the most vulnerable to returning?

·         Which factors are associated with leaving TANF, being employed, or returning to TANF?

This report examines these research questions for two populations of low-income families: those diverted from TANF prior to enrollment and those who have left TANF.  Among ‘diverted’ families, three types are being studied: families redirected prior to TANF application, those denied TANF for non-financial reasons, and approved TANF applicants opting to receive a one-time payment in lieu of TANF benefits. 

‘TANF leavers’ include both single-parent and two-parent families whose TANF cash grant has ended for the entire family and who have not returned to TANF for at least two months. TANF leavers are drawn from two different time periods for this study. Subgroups of both TANF divertees and leavers are defined further in Table ES-1.[1]

Table ES-1:
Populations Included in the Administrative Data Analysis

Population Type

Subgroup Name

Share of Total




n = 43,476


Potential applicants who were redirected (informally diverted) from TANF and did not apply for cash benefits in the same month from April 1998 through June 1999.

Non-financial denials

n= 85,854


Applicants who completed an application for TANF from April 1998 through June 1999, but whose application was denied for a reason unrelated to earnings or assets, such as failure to attend appointments.

One-time recipients

n= 1,791


TANF applicants who opted to receive a $1,000 one-time payment from April 1998 through June 1999 in lieu of TANF for the following year.


Cohort 1

(April 1998-June 1999)

n= 143,491


Families composed of 2-parent caretakers and children or single-parent caretakers and children whose TANF cases were closed in April 1998 through June 1999, and who remained off the rolls for at least 2 months.

Cohort 2

(July 2000-Sept. 2000)

n= 23,113


Families composed of 2-parent caretakers and children or single-parent caretakers and children whose TANF cases were closed in July 2000 through September 2000, and who remained off the rolls for at least 2 months.

Source:  Individual-level agency administrative data records from DHS.

Research Methods

This research project includes monitoring, qualitative, and econometric research methods to answer the research questions listed above.  The methods used for each population and time period are summarized in Table ES-2 and described briefly below. The table also displays the geographic coverage for each of these methods.

Monitoring/Descriptive Approaches

Monitoring and descriptive research approaches allow researchers to track certain populations over time or summarize statistical information about such populations at various points in time.  This study uses two descriptive approaches:

·         Demographic and longitudinal analyses of divertees and TANF leavers using individual-level administrative program data maintained by various Texas state agencies, and

·         A telephone/mail survey of a random statewide sample of TANF leavers (statewide survey) conducted approximately six months following TANF exit.

Table ES-2:
Summary of Research Approaches and Time Periods Used in This Study


Time Period of Cohort Entry

Research Approaches / Data Sources Used

Geographic Coverage

Diverted Applicants

April 1998-June 1999 (Cohort 1)

·         Descriptive analysis using administrative data 2 years prior through 18 months after event

·         Qualitative analysis using intensive interviews occurring sometime in the 15 months after event

Statewide universe

Sub-state sample

TANF Leavers

April 1998- June 1999 (Cohort 1)

·         Descriptive analysis using administrative data 2 years prior through 18 months after event

·         Qualitative analysis using intensive interviews occurring sometime in the 15 months after event

·         Econometric analysis using administrative data only


Statewide universe

Sub-state sample

Statewide universe

TANF Leavers

July – September 2000 (Cohort 2)

·         Descriptive analyses

a.   administrative data 2 years prior through 6 months after exit

b.   telephone/mail survey within 6 months after exit

·         Econometric analysis

a.   administrative data only

b.   combination of administrative and survey data

Statewide universe

Statewide sample

Statewide universe

Statewide sample

Demographic and Longitudinal Analyses.  A number of individual-level administrative data files from programs that serve Texas low-income families were linked to determine the demographic characteristics of families within each population and to follow families’ program participation and economic well-being over time.  The statewide universe of each population being studied is included in this analysis.

Statewide Survey of TANF Leavers.  A statewide, randomly selected sample of 1,596 families leaving TANF in July through September 2000 and remaining off of TANF for at least two months was surveyed approximately six months after leaving TANF.  Of this group, 45 percent of the original sample (723) and 70 percent of those actually receiving the survey responded to it.  Researchers documented that 581 of the original 1,596 families had moved and verified from administrative data that respondents closely resembled the universe from which they were drawn.

Qualitative Analysis/Intensive Interviews

UTCSWR conducted intensive in-person interviews with samples of leavers, redirects, applicants denied for non-financial reasons and recipients of one-time payments.  In-depth interviews were conducted with 461 persons in eight different research sites at some point during the fifteen months after their diversion or TANF exit.[2] These interviews provide examples of families’ experiences that more fully explain the results from the statewide statistical analyses conducted from other data sources.

Econometric Analysis

To determine the factors associated with leaving TANF, being employed and returning to TANF, several regressions were developed for two cohorts of TANF leavers to measure factors influencing the probability of TANF exit, the probability of employment, and the probability of returning to TANF.  Independent variables included a number of demographic, program-specific, employment, and county-level economic variables. 

Summary of Research Findings

Demographic Characteristics of the Research Populations

Demographic characteristics of divertees and leavers were determined from administrative data files.  These files were also used to check the comparability of the samples of survey and interview respondents to the larger populations from which the samples were drawn.

Divertees.  From April 1998 through June 1999, 131,121 families were diverted from TANF (25-40 percent of all applications received).  One third were redirected from TANF prior to filing an application.  Sixty-five percent applied for TANF but were denied for non-financial reasons, and only 1.4 percent were approved for TANF but opted to receive a lump-sum payment.  (See Table ES-1 for definitions of each group.)  All TANF divertees averaged 30-32 years old with two children. Members of each subgroup differed in their prior employment experience and the size of the counties in which they lived.  Two-parent families comprised over half of the one-time recipients (compared to eight percent of TANF entrants), a trend that continued throughout the study period.

Leavers.  From April 1998 through June 1999 (Cohort 1), 143,500 caretaker-headed TANF cases were closed, while 23,113 similar families left TANF from July through September 2000 (Cohort 2).  Ninety-four percent of Texas families leaving TANF were headed by single females.  Nearly 40 percent of these parents were less than 25 years old, while another third were between 26 and 34 years old.  Nearly half (45 percent) were Hispanic, with the remainder divided somewhat evenly between Black and White caretakers.  Most families had two children, with the youngest child being less than five years old.  Half of the caretakers in families leaving TANF had completed high school.  The characteristics of TANF leavers did not vary much between the two cohorts identified for this study. 

Participation in Government Programs

Administrative and survey data were used to analyze the degree to which divertees and leavers used TANF, Medicaid, and food stamps, while examples drawn from the interviews more fully described the nature of families’ experiences.

Use of TANF after Diversion or Departure from TANF

Divertees.  Twelve percent of redirected families and 26 percent of those denied for non-financial reasons received TANF at some time during the year following diversion.  The rates of TANF usage after the diversion activity quickly returned to prior levels of usage for these two groups.  As set by policy, persons receiving the one-time payment did not re-enter TANF for twelve months following the payment.  By 18 months after the payment, ten percent of those families were receiving TANF, a figure confirmed from both administrative data and qualitative interviews.

Leavers.  After leaving TANF, 63 percent of Cohort 1 families remained off TANF for the entire year after exit.  Most returns occurred within the first six months.  Although Cohort 2 leavers could only be tracked for six months after exit from administrative data, patterns of return to TANF were similar. 

Use of Medicaid

Divertees.  The use of Medicaid varied among different groups of divertees.  Only one fourth of redirected caretakers received Medicaid at some time in the year following diversion, compared to 38 percent for families denied for non-financial reasons, and 54 percent of one-time recipients.  Approximately ten percent of redirected adults were enrolled in Medicaid in all months of the observed period.  For the other two groups, Medicaid usage dipped at the time of the diversion, then increased to levels as high or higher than before the diversion.  Data limitations prevented the computation of Medicaid usage for children of diverted populations.

Leavers.  Sixty percent of adult caretakers and 77 percent of children received Medicaid at some time in the year after leaving TANF.  While only 20 percent of adults and children continued their Medicaid receipt when their TANF case was closed, 62 percent of children and 46 percent of caretakers received Medicaid in the following quarter.  Qualitative interviews found that some prior recipients still believed that their Medicaid coverage was linked to TANF and assumed that they and, to a lesser extent, their children were no longer eligible for Medicaid after departing TANF.

Use of Food Stamps

Divertees.  Sixty-three percent of redirected families, 56 percent of those denied for non-financial reasons, and 87 percent of families accepting one-time payments received food stamps at some point in the year following TANF diversion.  All families diverted from TANF increased Food Stamp participation in the period immediately following the diversion.  A year after the diversion, all groups of families had settled at rates of use similar to those prior to the diversion. 

Leavers.  Food Stamp participation dropped substantially for families after they left TANF.  Most of these families still appeared to be financially eligible for food stamps, and 68 percent of them used food stamps at some time in the year following their exit from TANF, a rate confirmed by the statewide survey.  However, while three fourths of these families received food stamps two months prior to TANF exit, that figure dropped to 37 percent at TANF exit and continued to decline to 30 percent by the end of the 18-month follow-up period. Qualitative interviews indicated that some leavers had difficulty understanding and responding to the eligibility criteria for the Food Stamp and the Food Stamp Employment and Training program.

Employment and Other Economic Supports

Employment status and access to other economic supports – including child support, child care subsidies, and the Earned Income Tax Credit – among TANF divertees and leavers were examined using all data sources.

Employment after Diversion or Departure from TANF

Divertees.  Among divertees, redirected applicants had the highest rates of employment (55 percent) and earnings ($2,096) in the quarter of diversion.  Those denied for non-financial reasons and recipients of the one-time payments had employment rates (28 - 31 percent) and quarterly earnings ($1,255 - $1,490) lower than or similar to those of approved TANF applicants.  Earnings for all of those employed, across all three groups, dipped in the quarter prior to their application for TANF benefits, and then rebounded in the 18 months following application.

Qualitative interviews indicated that recipients of one-time payments were more likely to engage in seasonal work than other diverted groups and requested the one-time payment when facing a period of unemployment.  These seasonal occupations included farm labor, work for school districts, and work in the tourist industry.

Leavers.  The employment and earnings of TANF leavers increased in the months immediately prior to TANF exit, with approximately 55 - 57 percent employed in the quarter of exit.  Seventy percent of leavers were employed at some time in the year following their exit but only one-third held jobs in all four quarters after leaving TANF.  Eighteen months after leaving TANF, half of Cohort 1 leavers were employed.  Earnings increased during the six quarters following exit, with employed Cohort 1 leavers reaching average quarterly earnings of $2,500 by the end of the study period. 

The statewide survey provided additional information about the nature of the jobs held by TANF leavers.  At the time of the survey completion (approximately six months after TANF exit), 46 percent of all respondents reported employment with an average hourly wage of $7.20.  Approximately 29 percent of those employed received some employee benefits connected with their jobs.  Among those employed, 25 percent considered themselves in temporary or seasonal work.  Respondents were heavily represented in clerical work, retail work, food services, and health care.

Respondents to qualitative interviews who entered higher paid occupations such as teaching and nursing explained their success in part by access to education.  Overall, respondents valued employment and described themselves as employed even if they were underemployed or planning to start a new job.

Child Support

Divertees.  Only between three to eight percent of caretakers diverted from TANF received formal child support payment in the month of the diversion activity, with payments averaging $227-$314 per month.  Although these percentages increased over time, only ten percent were receiving child support 18 months following diversion.

Leavers.  Families leaving TANF received child support more often than divertees, but the average monthly payments were smaller.  Nine percent of Cohort 1 TANF leavers (12.5 percent for Cohort 2) received child support at exit, with monthly payments averaging $207 and $232 for the two time periods.  Both the rates and average amounts of child support increased steadily over time, with 21 percent of Cohort 1 leavers receiving at least one payment in the year following exit.  While the statewide survey of TANF leavers reports 19 percent of respondents receiving child support in the four to eight month period after leaving TANF, some of this reported support was undoubtedly informal assistance obtained outside the formal child support system.  Such informal payments could be sizeable, but were often irregular.

Subsidized Child Care

Divertees.  Fewer than five percent of persons diverted from TANF (and five to nine percent of employed divertees) received subsidized child care prior to diversion, a figure that remained relatively constant throughout the study period.  The qualitative interviews found that only a small proportion of diverted families used formal child care arrangements, typically the same families that received subsidies.  One-time payment recipients, who were more likely to be members of two-parent families, used the other parent as a source of child care more often than other groups did.

Leavers.  Eleven percent of Cohort 1 leavers (17 percent of employed leavers) received subsidized child care in the month of exit.  Cohort 2 leavers were more likely to receive child care subsidies, with 16 percent of all caretakers (24 percent of employed caretakers) receiving such care at TANF exit.  Both administrative and survey data found lower rates of subsidized child care usage six months after TANF exit.  As with divertees, the proportion of those receiving subsidies was almost identical to those using formal day care centers and registered family homes for care.

Qualitative interviews found that both leavers and divertees rely heavily on informal child care arrangements.  According to the statewide survey, about ten percent of households (usually those with older children) depended on children caring for themselves.

Total Family Income

Most families in this study relied on a combination of earnings by adult family members and/or TANF payments as their primary sources of income.  While many families are eligible for child care subsidies and formal child support to supplement their income, less than one in five families received income from these sources at any time during the period of study.  Although the 1999 National Survey of America’s Families (NSAF) estimates that 33 percent of current and 69 percent of past Texas TANF recipients had ever received the Earned Income Tax Credit (EITC), only 3.3 percent of respondents in the statewide survey of recent TANF leavers reported using the EITC in the recent past.[3]  (The timing of the survey may have influenced this finding.)

Hardships and How Families Manage

Data from the statewide survey of TANF leavers and the qualitative interviews with both leavers and divertees provided elaboration on the hardships and barriers to employment faced by families and their strategies for meeting these problems.  Strategies used were similar for TANF divertees and leavers, with any differences identified.

Barriers and Hardships

Unemployed survey participants cited an average of 2.3 different barriers to their employment, with 33 percent reporting multiple (three or more) barriers.  Qualitative interviews revealed that one barrier may make it more difficult to respond to another barrier.

The most prominent barriers mentioned were:

Child Care.  Thirty-three percent of all survey respondents reported problems with child care in the preceding six months.  Thirty-one percent of unemployed respondents reported child care as a contributing factor.  According to the respondents of qualitative interviews, formal sector child care was unaffordable by most families without child care subsidies.  Informal child care was often unreliable.

Transportation.  Twenty-three percent of all survey respondents and 26 percent of unemployed respondents reported employment-related problems caused by poor transportation in the preceding six months.  Qualitative interviews indicated that difficulties with transportation not only affected employment, but also access to other services and supports, including food stamps and Medicaid.

Health Problems.  Unemployed survey respondents reported health problems as a cause both of loss of employment (18 percent) and of return to TANF (15 percent).  Twenty percent of all respondents reported a health problem or injury that interfered with usual activities, and 18 percent reported health problems among other family members.  Qualitative interviews indicated that health problems also affected their employment options.

Although difficulties with housing and food were not included in the survey questions on specific barriers to employment, many respondents also reported having problems in those areas.  Thirty-eight percent of survey respondents reported that, at some time in the previous six months, they could not afford housing costs.  Thirty-seven percent also reported that they had had at least one occasion in the past six months when they needed food but could not afford to purchase it.  Qualitative interviews indicated that housing and food problems affected employment options.  Without stable sources of housing and food, respondents could not become or remain reliable employees.

How Families Manage

Strategies used by families to gather resources and sustain employment included:  1) gaining assistance from family and friends for help with transportation, child care, food, housing, and other items; 2) finding irregular jobs or second jobs; and 3) gaining assistance from local agencies.

Contrary to some early predictions that welfare reform would result in the increased use of child protective services, little evidence was found to support this view for Texas TANF leavers.  Substantiated investigations of child abuse or neglect were reported for less than two percent of families, and children were placed in foster care for less than one percent of families leaving TANF.  Similar measures could not be calculated for diverted populations due to data limitations.

Experiences in the Welfare Office and Views of the Application Process

Data from the statewide survey and accounts from the qualitative interviews gave further insight into the nature of people’s experiences in the welfare office.

Views about Work Requirements.  Clients were eager to work, and valued services and supports they perceived as helpful to them in entering the labor force.  In particular, they cited case management activities that provided access to both training and job search assistance.

Difficulties with TANF.  Clients distinguished between difficulties in dealing with caseworkers and difficulties with TANF policies, exhibiting a range of attitudes and responses.  Many expressed support for TANF policies that encouraged employment and accepted the eligibility requirements and the need to re-certify for benefits periodically.  Complaints of difficulties emphasized poor office management and the actions of selected caseworkers.

Points at Which Families are Most Vulnerable to TANF Recidivism

A longitudinal analysis of administrative data found that families are most likely to return to TANF in the first few months after leaving the program, with 28 percent returning to the rolls in the first six months after exit.  Over the 18-month follow-up period, 41 percent of families returned to TANF at some time, although many left again during the observed period.  Families returning to TANF were disproportionately Black, young, or with several children.  Short-term leavers (i.e., those returning to TANF within six months after exit) used more government services than others.  They also had less income and less reliable employment than other leavers.

Factors Associated with Leaving TANF, Employment, and
Returning to TANF

Regression analyses using both administrative and survey data explored the factors associated with leaving TANF, employment, and returning to TANF.  (Readers should note that statistical association does not always equal causality.)  Detailed case studies drawn from the qualitative interviews were used to better illustrate the statistical findings.

Factors Associated with Leaving TANF.  About 16 percent of TANF families left TANF each month during the study period.  The major factors associated with increased probability of exit were:  current employment of caretaker, a caretaker’s refusal to register for employment services, or being in a two-parent family. Respondents to the statewide survey also cited employment as the primary reason for departure from TANF and also reported receipt of income from other sources and failure to comply with TANF regulations as common reasons for exit.

To a lesser degree, these other factors were also associated with TANF exits: receipt of any TANF penalty, exemption due to caring for a young child, employment for a larger percent of time, or larger average monthly child support payments. 

Factors Associated with Employment for TANF Leavers.  Several regressions were run for this measure (see Table ES-2) to take advantage of the total universe of administrative data and additional variables available for persons sampled in the statewide survey.  From the regressions using only administrative data, the greatest predictor of current employment was the percent of time a caretaker had been employed in the prior two years.  Other major factors associated with current employment included: receipt of subsidized child care after leaving TANF, participation in the Choices program during the most recent period on TANF, receipt of children’s Medicaid after leaving TANF, or being a Black caretaker.  Regressions that included survey data revealed that having a youngest child cared for by non-relative, having looked for a job in the past six months, or access to reliable transportation also were associated with higher employment rates.

In the administrative data regressions, the strongest predictor of lower employment rates was being a caretaker in a two-parent family. The statewide survey revealed that, in these families, one parent cared for their children while the other worked.  Other factors associated with lower employment included: an eighth grade education or less, receipt of a non-workforce penalty, refusal to register for employment services, or receipt of larger amounts of child support.  Regressions using the additional variables from the statewide survey found that widows, persons who had entered TANF after a divorce, and respondents who cared for their youngest child themselves, relied on family or friends to help with transportation in prior six months, lived with family or friends in the past six months, or experienced multiple barriers to employment were also less likely to be employed than other TANF leavers.

Factors Associated with TANF Reentry.  Regressions using both administrative and survey data found that caretakers were less likely to return to TANF if they were currently employed, had higher wages at TANF exit, left TANF because of marriage, received financial contributions from another adult, collected larger amounts of child support, or had stable transportation.  Those who needed assistance with food or housing costs, received child care through the Choices program while on TANF, or had their youngest child cared for by a non-relative were more likely to return to TANF.  Persons with a history of prior employment also were more likely to return to TANF, which suggests that the employment of some TANF leavers was not very stable.

Conclusions and Policy Implications

Conclusions.  Families diverted from TANF quickly resumed their prior levels of TANF, Medicaid, and Food Stamp participation and their prior level of earnings.  This suggests that Texas’ diversion policies have few long-term effects on diverted families.  The experiences of Texas TANF leavers in the post-welfare-reform era fall with the ranges found in other states in their use of government benefits, employment and earnings.  However, Texas families are somewhat more likely to return to TANF than those in other states, which could be caused either by fewer post-TANF economic supports or less restrictive Texas time limits and diversion policies.

Within Texas, far fewer families are using TANF than used AFDC in the pre-welfare reform period, some due to increased employment and other economic supports but others because the welfare rules seem onerous or confusing to them.  There are fewer long-term cash welfare recipients now than in the early 1990s and a somewhat larger share of families who cycle between work and welfare.  Except for those families leaving TANF due to welfare reform requirements, reasons for exits from and returns to welfare (employment and change in family structure) have not changed, nor have the factors influencing the success or failure of welfare exits.  A clearer view of the lives of non-employed TANF leavers can be gleaned from the current study than was possible from the pre-welfare reform studies that used only administrative data.

Policy Implications.  Some existing TANF and related policies aid families’ employment and overall economic well-being while others merely confuse them or add to the challenges of managing their work and family responsibilities. In general, subsidized child care, post-TANF Medicaid benefits and Choices services enhanced families' income or employability and were viewed positively by recipients.  However, the limited availability of such services (due to funding) and program restrictions favoring TANF families mean that many former TANF families could not access these services. 

Both the statistical and qualitative analyses identified other policies that hamper families’ efforts to both work and care for their families.  In particular, the paperwork and confusing requirements of TANF eligibility and re-certification procedures, the personal responsibility agreement, and associated programs (e.g., Food Stamp E&T) should be reviewed and simplified wherever possible.  Families also cited eligibility restrictions for family members who co-signed for car loans and limited options for the development of stable transportation as other impediments to their employment.

Some existing policies that can help families – such as children's Medicaid and food stamps for non-TANF recipients and the Earned Income Tax Credit – were not used by many of the families who appeared to be eligible for these services because many families did not realize that they were eligible for them.  Other new initiatives – including the expanded earned income disregard and publicity efforts to inform families about availability of children's Medicaid and family Food Stamp eligibility – could not be properly assessed during the time period of this study.

[1] Some families who left TANF during the study period and attempted to re-apply for benefits may be counted as both leavers and divertees.

[2] Research sites included Bexar County (San Antonio), Harris County (Houston), Jasper County (Jasper), McLennan County (Waco), Hale County (Plainview), the two-county Valley area comprised of Cameron and Hidalgo counties, and single offices in Austin and El Paso.  These sites were selected to provide variation in urbanization, racial and ethnic demographics, and labor markets.

[3] Unpublished tabulations from the 1999 National Survey of America's Families performed by The Urban Institute (2001).