Difficulties after leaving TANF: inner-city women talk about reasons for returning to welfare.

Anderson, Steven G. ; Halter, Anthony P. ; Gryzlak, Brian M. 等

As state Temporary Assistance for Needy Families (TANF) programs have implemented five-year time limits and other restrictions on welfare use, the importance of fostering sustainable welfare exits has grown. This task is most daunting in large inner cities, in which the debilitating effects of concentrated poverty often militate against economic mobility (Jargowsky, 1997; Wilson, 1996). Although most states have initiated program evaluations to ascertain basic outcomes for people who leave TANF, less is known about the experiences of inner-city leavers in post-TANF welfare environments. Also, research has not focused on the reasons why one-fifth to one-third of TANF leavers return to welfare within one year (Acs & Loprest, 2001; Loprest, 1999).

Based on focus groups conducted in five high-poverty neighborhoods in Chicago, this article explores the problems faced and strengths used by women who are poor as they attempted to achieve self-sufficiency after leaving TANF. Because our sample included primarily women who left but then returned to TANF, study findings provide unique perspectives on why TANF exits often fail in poor inner-city areas. Such unsuccessful results often have been overlooked, because early studies have focused on average outcomes (Moffitt & Roff, 2000). Study participants' observations about their experiences with public bureaucracies also provide valuable information about how program implementation issues can powerfully affect success after leaving TANF.

Early TANF Results, Welfare Exits, and Inner-City Poverty

Welfare caseloads have declined dramatically since the 1996 Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA) (P.L. 104-193) established TANF programs. Early state studies have found that about 50 percent to 70 percent of TANF leavers are employed immediately after exiting (Acs & Loprest, 2001; Tweedie, Reichert, & O'Connor, 2000). However, average earnings are usually below the poverty level (Acs & Loprest; Parrot, 1998). Furthermore, job instability is common, because leavers typically are employed in temporary or low-skilled jobs (Anderson, Halter, Julnes, & Schuldt, 2000). Studies also have shown that public supports like the earned income tax credit (EITC), transitional Medicaid, and child care are underused (Anderson, Halter, & Schuldt, 2001).

There is little information available concerning the principal reasons that those who leave TANF often return. However, pre-TANF studies found that welfare returns resulted largely from structural employment problems (Harris, 1996; Pavetti, 1993). Edin and Lein (1997) reported that those who left welfare often concluded they were worse off after leaving, because of both work disincentives and the tenuous employment available to poor women with limited education.

The problems faced by those who leave TANF may be especially difficult in large cities, because of the high concentrations of recipients, limited job opportunities, and city fiscal constraints (Kahn & Kamerman, 1998; Quint et al., 1999). Wilson (1996) argued that the decline of manufacturing jobs in large cities, when coupled with the "out migration" of middle-class members of ethnic minority groups and the related disintegration of community institutions, has resulted in dense poverty areas characterized by lack of businesses, few successful role models, and limited job networks. Therefore, Wilson contended that welfare reform must combine education and training, income support, and job creation strategies if it is to improve the well-being of the inner-city poor population.

Others researchers have suggested that long-term poverty results primarily from the dysfunctional behaviors of people who are poor. Welfare recipients are seen as being mired in a "culture of poverty," reinforced by welfare policies that make not working relatively attractive (Mead, 1992). This perspective is reflected in TANF policies such as time limits and forced work searches, which assume people will choose welfare over work unless precluded from doing so.

Qualitative studies preceding TANF cast doubt on the accuracy of these behavioral arguments. For example, several researchers argued that discouragement about work opportunities results from long-term effects of structural constraints, such as unstable jobs, limited education, and weak government support services systems (Jarrett, 1994; Wilson, 1996). In addition, people living in poverty were found to share mainstream aspirations regarding work and to exhibit strengths as they struggled with material deprivation (Edin & Lein, 1997; Seccombe, 1999). Qualitative researchers also found that poor people were dissatisfied with their economic position and were stigmatized by welfare receipt (Dodson, 1998; Jarrett).

Because TANF fundamentally changed the welfare system, it is not clear how these arguments may explain welfare exits and returns in evolving state and local programs. The TANF program placed little emphasis on job development, so the structural problems observed in pre-TANF research studies may continue to undercut successful welfare exits. Likewise, if behavioral problems consistent with arguments about the culture of poverty are prominent, those who leave may have difficulties coping with work expectations and may return to welfare to escape work stresses.

We selected Chicago as an important city in which to re-explore these and other concerns about life after leaving welfare in the new TANF program environment. Earlier research had shown that 33 percent of TANF leavers in Cook County, in which Chicago is located, returned to TANF within one year (Julnes et al., 2000). Cook County leavers also had less education and job experience, had longer earlier spells on welfare, were more likely to be African American or Hispanic, and were more likely to never have been married than other Illinois TANF leavers.

Method

The perspectives of social program participants generally receive minimal attention in assessing social programs (Chapin, 1995). These exclusions have serious implications, because the social distances between analysts and recipients are great, and the frames of reference used to assess programs differ (Tickamyer, Henderson, White, & Tadlock, 2000). For example, although policy analysts apply analytic skills to program descriptions and data, welfare recipients are uniquely positioned to describe direct experiences that reflect on the implementation of services.

Focus groups are one useful method for obtaining unfiltered perspectives from program participants. These groups include a small number of people who share common experiences. Group facilitators encourage the expression of participant ideas, but discussions generally are unstructured (Greenbaum, 1993). It also is assumed that group dynamics result in insights that may be unobtainable through individual interviewing (Lederman, 1990).

The focus groups for this study were arranged with the assistance of tire agencies serving poor people in different Chicago neighborhoods. Staffs from these agencies recruited participants by distributing flyers about the project, as well as through existing contacts with TANF recipients. For four groups, participants were screened to ensure that they had left welfare and then returned. People in these groups were back on TANF when interviewed, so a priority was to learn why their TANF exits had been unsuccessful. The final group consisted of people who had left TANF and remained off at least six months at the time of the group meeting; this group was added to ensure that client strengths after leaving TANF were considered.

We conducted the focus group meetings between August 1999 and September 2000. Each participant in the five focus groups was paid $40 for a two-hour session. The participants in the four focus groups that returned to welfare met in sponsoring community agencies, and the fifth group, those participants who remained off TANF, met in a public housing project. Participants typically lived in high-poverty areas near the focus group sites and had received TANF and related services from a variety of local offices of the Illinois department of human services.

Two members of our research team moderated the focus groups. Because we were from outside the community, staffs from the sponsoring agencies introduced the moderators and expressed their support. Given concerns about participant distrust of TANF agencies and caseworkers, the moderators stressed their independence from the state human services department.

We developed an interview guide designed to capture various aspects of participants' work and welfare experiences, but the content of the sessions was flexible based on participant concerns. The topics explored in all sessions included participants' reasons for TANF exits and returns; experiences with work and support services; assessments about their relative well-being when on and off TANF; and the problems and strengths that either hindered or facilitated successful exits. Discussions also led to broader participant assessments of the welfare system, including perspectives on caseworker performance and the success of welfare reform implementation.

Fifty-one people participated in the focus groups. Forty-three of them had left TANF but then returned, and the remaining eight remained off TANF. Seventy-five percent were African American, 20 percent were Hispanic, and 5 percent were white. All study participants were women, and nearly all were in their 20s and 30s. Participants had an average of 2.8 children, with the number of children ranging from one to nine.

Each focus group session was audiotaped and transcribed. Project staffs then coded participant comments thematically based on analysis of response content (Strauss & Corbin, 1990), and selected quotes that best illustrated various themes. To maintain the integrity of participant responses, these quotes are presented verbatim in the following sections.

Findings

Marginal Employment after Leaving TANF

Focus group participants indicated that their inability to maintain work that paid a living wage most often led to their returns to TANF. All study participants had worked after leaving TANF, and employment had played an important role in most TANF exits. However, the jobs they found paid low wages and often did not last. Generally having little if any savings and often having lost health care benefits, the costs of job loss for these women were high and immediate. Job losses sometimes resulted from health, performance, or support services issues, but participants more often spoke of the structural characteristics of the jobs. Many of the jobs were temporary or seasonal, but the women did not always understand this when they began working. As one respondent said, "They hire you seasonal but they don't tell you that when they hire you. They hire you full-time and then as soon as the season is over you are out the door."

Even when these women did not lose jobs involuntarily, wage levels sometimes resulted in returns to TANF. Low wages led some to compare their economic status before and after leaving public assistance. When these comparisons led participants to conclude that work did not increase economic well-being, disillusionment was evident:

 I went to a job as an assistant teacher, and I
 thought I really was going to make some
 money and do very well. I got off of public
 welfare, because I knew I could not make it
 that way. And so, what happened was, I waited
 five whole weeks on a check. And when I got
 my first check and I began to calculate my
 week's pay against what I was getting per
 month on public aid, including my benefits, I
 was getting $80 a month less. Even though I
 was struggling on public aid, now I was working
 very hard every single day, and I was getting
 $80 a month less by working.

 I had left public aid and went and got a job. I
 was making maybe $5 or something an hour.
 And when the deal came down and I added it
 up, I was doing better with public aid. Because
 they had took my medical card, then cut my
 food stamps so short I had to put cash with
 that to go shopping, and when you're dealing
 with five kids and they're teenagers, well, that
 wasn't enough food. So I said, I might as well
 quit this job and go back to public aid.

Caseworkers were seen as contributing to these problems, because of the jobs to which they referred TANF recipients. As one woman said, "The only trouble is the jobs she [caseworker] sends me on. They're like government funded. One might last six months, one might last nine months ... it really doesn't last." Some participants also indicated that employers took advantage of public incentives for hiring welfare recipients. According to another woman: "Some of these people are getting employed in hospitality and hotels and work a little while. Those places get that money that public aid is paying them and then they let those people go."

Participants recognized that having poor job skills or limited experience contributed to their low pay. Many questioned the "work-first" philosophy of placing recipients in the first available job, because they thought this resulted in job placements that were insufficient to remain off TANF. Several focus group members said TANF instead should provide more extensive skills training and education and develop linkages to better jobs:

 Before they kick them off, be sure they're educated.
 You can't take a person off and put them
 on a job without a high school diploma. It's
 easy to go out and get a $5 job, but you need
 your GED or high school diploma--some type
 of schooling to keep moving forward. Because
 if you do that [send people to low-paying
 jobs], they're gonna be back on aid.

 But the mistake they're doing is limiting our
 education ... we go to work, but what happens
 is we are still falling back into the system
 no matter what. Because we are not educated
 enough, we don't have the proper training,
 and we are not going to advance in our job,
 which makes us fall back where we started
 from--back into the system.

Even those few who obtained better-paying and more stable jobs faced tenuous circumstances. Because of the shortage of jobs in poor neighborhoods, participants often commuted long distances. This created child care difficulties in addition to transportation problems, as child care providers were not always available for women leaving early for, or returning late from, work. The unavailability of child care for irregular shift times resulted in similar problems. For example, one participant quit a well-paying job after being moved to a later shift simply because she could not find child care for that time period.

Difficulties in Obtaining Supportive Services

Problems in obtaining support services also contributed to TANF returns. Participants emphasized the lack of medical insurance offered through the jobs that they found. The jobs that did provide coverage required employees to first serve probationary periods, which could be problematic given the previously mentioned job instability. Employer-based health plans also generally required employee premiums and co-payments, which compromised affordability.

Illinois offers transitional Medicaid coverage for working TANF leavers for up to one year and extended coverage for low-income children. However, many participants were unaware of the availability of this coverage or were confused by program rules. They consequently did not always receive medical benefits for which they were eligible, which contributed to decisions to return to TANF so that Medicaid coverage would be available.

Obtaining child care services presented similar difficulties. Illinois has a program that provides non-time-limited child care for low-income children based on a sliding-fee scale. Although recognizing that this help was intended, participants encountered problems in gaining access to it. Such difficulties led to returns to TANF, which provided unintended signals about policy to recipients:

 I applied [for child care] when I was off and
 they said I made too much money. So when I
 quit my job and went on [TANF] they paid for
 it. [Moderator: Because you're going to
 school?] Exactly ... you all don't want to do
 this when I'm trying to work, O.K., so I'll go
 to school.

Many participants argued that transitional income assistance was needed to help with short-term cash flow problems associated both with employment start-up costs and delays in receiving initial paychecks. In addition, transitional assistance was viewed as a safety net in case of job loss. This concern was elevated by the delays participants had experienced when trying to get back on TANF after losing jobs:

 Then when you do find a job, you've got probation
 periods of 90 days. I think they should
 not cut you off right away, to make sure that
 this is a steady job, and this is a job for you,
 and this is a company that wants you. Because
 if its not and it doesn't work out, then you got
 to go through channels all over again.

 I think in my case what would have made the
 most difference is a transitional period. If I am
 fortunate enough to get a job, let me keep my
 food stamps and my benefits for six months
 ... six months is a relatively good time for me
 to say that people should be able to get on
 their feet ... and don't mind me saving $200
 in the bank.

Interactions with Caseworkers

Focus group participants often made reference to the power caseworkers had in decision making: "This is your life ... and these people got you on a string and you're their puppet." Participants spoke of two principal dimensions when discussing caseworker interactions. First, caseworkers were seen as critical in determining whether TANF leavers could obtain medical care, child care, and other services, and many participants thought that caseworkers had not helped them get benefits to which they were entitled. This was attributed to a lack of competence in some instances, but also to deliberate withholding of benefits: "We have some caseworkers that know information, but they won't give information." Participants also indicated that systemic problems such as high caseloads or caseworker turnover complicated their efforts to learn about services:

 And with the caseworkers, the turnover, the
 turnover, the turnover. You have one caseworker
 for four months, then you get another
 caseworker--it's set up so you can't even build
 a relationship.

A second important concern was the interpersonal dynamics between caseworkers and TANF recipients. Although some talked of good personal relationships with their caseworkers, participants more commonly expressed anger over perceived mistreatment. Caseworkers' tendencies to stereotype and look down on recipients especiaLly offended many group members:

 She [the caseworker] stereotyped every
 woman in there. She told us that, now, the
 days of selling your body, and selling your
 food stamps, and getting high with your
 money and all kinds of other junk, she said
 that it was over.

 To me their job is to downgrade us. They
 don't like us, they don't respect us as human
 beings, they feel that we are the ones that are
 in need--not them--and that they can treat
 us like low people that are in poverty.

Ironically, negative recollections about case-worker interactions appeared to inhibit returns to TANF, because participants often viewed the experience as so demeaning. As one woman said, "I really don't want to go through the hassle anymore [of getting TANF] ." Such caseworker-induced stigma is an important issue for social work.

Strengths after Leaving TANF

Participants described positive experiences that reinforced their efforts. The most notable strength observed was the high value that participants attached to working, even in jobs that policy analysts generally consider marginal. Three factors contributed to such positive attitudes. First, contrary to popular stereotypes, these women paralleled the general population in deriving psychological benefits from employment. Working served both as a source of pride in accomplishment and as a place to form new friendships:

 To get up and go out and make a start, makes
 you feel better about yourself. So I think it's
 [welfare reform] a good idea, and I know a lot
 of people now that were on public aid and
 they're off now. And they have a job, and they
 feel much better about themselves.

 After my first child, I started working at
 Wendy's in Chicago--I worked there for three
 years. I felt very good about myself.

Second, some group members spoke of the improved economic circumstances they enjoyed when working. Although low wages often undercut these rewards, those who obtained higher wage jobs spoke of tangible benefits they could not enjoy when on TANF:

 I'm buying my first house--I'm proud of myself.
 I never had made that kind of step--I
 think it's a good step to make. So I enjoy every
 week--I don't care if it's $200 a week--I enjoy
 seeing that check.

 When I was on public aid, there was times that
 I wanted to take the kids to certain restaurants
 to eat or shows or buy them something special--working,
 I can do that. I found myself. I've got
 a little piggy bank, and every time I get paid, I
 put $10-$15 in that bank. And it's just adding
 up. It's going to be a little Christmas money. I
 never was able to do that. It just feels good.

Third, several participants said that working or going to school sent a positive message to their children. As one woman said, "Since I've been working, my children see me really trying, and they try harder too." Some also indicated that they had better conversations with their children after they began work or training, partly because these activities provided them with experiences to share.

Despite their concerns that TANF provided inadequate job preparation, participants described positive experiences with some job-training programs. Among the benefits noted were receiving basic information about self-presentation and job searches, as well as the camaraderie derived from sharing experiences with people in similar circumstances. These benefits seemed particularly important to women with very limited skills and social networks:

 They taught you how to fill out applications
 correctly--be professional when you go to job
 interviews, and to just feel good about yourself.

 I wound up meeting people in the classroom
 that I still have a rapport with. We had a
 chance to talk about the struggles we went
 through in life. And the program motivated
 me to want to go out there and go to work.

Finally, participants emphasized that families and other informal supports were critical to their success after leaving TANF. Mothers, husbands and former husbands, boyfriends, siblings, and children all were mentioned as sources of strength that provided both tangible and psychological support. Informal help in caring for children was considered especially important. Because most group members had returned to TANF, we cannot tell whether those who remained off generally had greater supports than those who returned. This is an important area for further research, as is experimentation with program models that systematically assess TANF recipients' strengths and that develop case plans accordingly.

Discussion and Implications

In considering the lessons that emerged from these focus groups, one striking feature was how infrequently participants spoke of returning to TANF as a strategic choice. Popular myths and some research have portrayed decisions to go on or to leave welfare as resulting from careful cost-benefit calculations (Edin & Lein, 1997). Yet, participants in these groups more commonly simply returned to TANF after their exit experiences negatively deviated from expectations. The need to return often resulted from an unexpected job loss or a medical emergency. In other instances, it simply reflected learning by people with little previous job experience:

 Once I finished school, I got off public aid and
 got a job. I wasn't anticipating child care--I
 didn't realize it cost so much.

 When you first take the job
 you think "yeah, this is going
 to be something that's going
 to be there for me," but then
 you learn after you're on it for
 a while that its not quite what
 you figured.

As evidenced by the focus group comments on caseworker interactions, returning to TANF was not a welcome decision after a casual "testing" of the job market. Rather, it was viewed as an unpleasant but necessary alternative, borne much more from desperation than calculation.

This experiential emphasis points to distinctions about welfare returns under TANF compared with Aid to Families with Dependent Children (AFDC). The TANF program raised expectations among poor women, because its marketing heavily emphasized that they would be better off by leaving welfare for work. Most focus group members appeared to embrace this rhetoric. However, when actual experience deviated from TANF rhetoric and resulted in returns to welfare, cynicism about the intent of welfare reform often resulted:

 It [welfare reform] was needed, but the way it
 is set up now, it's just a bunch of hype and it's
 just a farce. When I think of welfare reform
 today, it's just something that they put on
 television to show white collar or possibly
 even blue collar people that you don't have to
 pay tax money for these women to have any
 more kids.

 The purpose is to help you find a low-paying
 job and have you live check to check. They
 don't help you. I think reform should be to
 help you find a higher paying job so that you
 will never have to return.

In many jurisdictions, returns to TANF also may be less welcome than under AFDC. These focus group members had managed to return to TANF, but some indicated that caseworkers questioned their need to do so. More research is needed on whether those who leave face obstacles to re-entry when TANF returns become necessary. It also would be useful to determine if concerns about time limits affect considerations about whether to return to TANF.

The recipient perspectives presented here suggest that both TANF-related policy revisions and improvements in the implementation of existing policies could enhance the stability of TANF exits. Consistent with pre-TANF research on welfare returns (Edin & Lein, 1997; Harris, 1996), study participants emphasized that low wages and job instability caused returns to TANF. These employment constraints are more consistent with structural than culture-of-poverty explanations of welfare dependency. The high level of job satisfaction expressed by participants also casts doubt on hypotheses that emphasize behavioral deficiencies among welfare recipients. This suggests that jobs typically viewed as unfulfilling may be rewarding to many people if accompanied by sound income and support services policies.

The continuing difficulty that low-income inner-city women experience in finding stable jobs should not be surprising, because PRWORA neither focused on job creation issues nor on improving job skills. Rather, it promoted immediate labor force attachment, under the assumption that work experience would stimulate sustainable employment. Resulting work-first program models have demonstrated government cost savings and have contributed to caseload reductions. However, research on these programs has shown only marginal short-term effects on economic well-being (Acs & Loprest, 2001; Parrot, 1998), and the long-term effects are unclear.

Job supply issues and the limited education and experience of inner-city TANF recipients cast doubt on the wisdom of relying on work-first program models in poor neighborhoods. Focus group members stressed that skill deficits consigned them to low-wage jobs with little chance of advancement and that additional training was needed to improve prospects for better jobs. This position is consistent with evidence indicating the need for the least-skilled recipients to increase education if they are to compete for stable jobs (Jargowsky, 1997; Wilson, 1996). A more balanced approach to TANF program development would allow education and training as alternatives to work. States such as Illinois, Kentucky, and Maine have moved in this direction by allowing TANF recipients to meet work activity requirements by attending college, junior college, or vocational training (Sweeney et al., 2000).

It also is difficult to envision long-term employment success for inner-city TANF leavers without economic reinvestment strategies to increase the supply of inner-city jobs. For example, enterprise zones are being tested as a mechanism for stimulating private investment. Because of the deterioration of many cities, public jobs programs also are needed. Such strategies may include infrastructure redevelopment, neighborhood clean ups, and human services jobs such as child care aides and playground supervisors (Danziger & Gottschalk, 1995; Wilson, 1996). Although these programs primarily offer jobs of last resort, they can provide stable employment and related job experience if accompanied by reasonable support services.

This raises the question of the adequacy of support services for those who leave TANF for work. Study participants pointed both to deficiencies in the implementation of existing services and to the need for additional supports. For example, Illinois policies stipulate that TANF leavers ate entitled to extended Medicaid and child care benefits, but focus group members often were unaware of or confused about these benefits. They also expressed the opinion that services receipt depended on caseworker attitudes and inclinations.

This seeming discretionary nature of benefit distribution represents a radical departure from intended TANF-related policies and from public administration norms concerning the dispassionate delivery of benefits. However, it is consistent with studies that have interviewed and observed caseworkers (Brodkin, 1997; Iverson, 2000). Additional research is needed to examine how support services implementation varies under differing administrative arrangements, as well as with improved worker training and information dissemination approaches.

The adequacy of existing income and services supports, even if well implemented, is another concern. With limited additional expansion of EITC, full-time workers could be ensured incomes of at least the poverty level without food stamp receipt (Wilson, 1996). In addition, at least 11 states have implemented state EITC supplements, so the precedent for state action to support low-income workers is well established (Sweeney et al., 2000).

The lack of health care coverage for working poor people is perhaps the greatest remaining disincentive to leave TANF. Although incremental expansions have dramatically improved public medical benefits for low-income children, adults who leave TANF generally do not receive health insurance through their jobs and receive only transitional coverage through Medicaid. Consequently, health care coverage tends to be better on TANF than off (Julnes et al., 2000). Although universal health care coverage should remain an important advocacy goal for social workers, more incremental expansions of coverage for low-income workers also merit attention. For example, many states have extended children's health care coverage available through Children's Health Insurance Programs or Medicaid for their parents (Sweeney et al., 2000).

Child care coverage for TANF leavers generally is even weaker than medical coverage. Although available federal child care resources have expanded since PRWORA was enacted, TANF programs are not required to provide even time-limited child care. Ensuring the availability of such subsidized care, therefore, is an important goal for TANF.

As the mix of work and training options, income assistance, and support services evolves, social workers have important roles to play in services delivery and in advocacy. One overarching issue concerns caseworker-recipient interactions involving TANF-related requirements and benefits. There is no evidence that well-designed case planning processes have been established in most TANF programs, and both the findings presented here and in earlier studies suggest suboptimal case planning implementation at best (Brodkin, 1997; Iverson, 2000).

A social work perspective on case planning suggests the need for a mutually negotiated services plan and expectations tailored to individual needs and strengths. At a minimum, TANF case planning should include the careful assessment of recipient employment skills and deficits, clear discussion of TANF work and training requirements and options, and the provision of information about services and benefits. Although public assistance caseworkers have become increasingly deprofessionalized in recent years, the new emphasis on time-limited assistance and employment provides an opportunity to advocate for these principles.

Many opportunities exist for social workers to advocate for policies designed to enhance the success of TANF exits. Because TANF devolved most welfare decision making to the states, effective social work advocacy related to TANF requires the development and maintenance of coalitions at the state and local levels. Schneider and Netting (1999) have developed specific process recommendations for social workers interested in developing TANF advocacy strategies, which are broadly applicable in advocating for various substantive recommendations.

References

Acs, G., & Loprest, P. (2001). Initial synthesis report of the findings from ASPE's "leavers" grants. Washington, DC: U.S. Department of Health and Human Services.

Anderson, S. G., Halter, A. P., Julnes, G., & Schuldt, R. (2000). Job stability and wage progression patterns among early TANF leavers. Journal of Sociology and Social Welfare, 27(4), 39-58.

Anderson, S., Halter, A. P., & Schuldt, R. (2001, Fall). Support service use patterns by early TANF leavers. New Directions for Evaluation, 91, 87-99.

Brodkin, E. Z. (1997). Inside the welfare contract: Discretion and accountability in state welfare administration. Social Service Review, 71, 1-33.

Chapin, R. K. (1995). Social policy development: The strengths perspective. Social Work, 40, 506-514.

Danziger, S. H., & Gottschalk, P. (1995). America unequal. Cambridge, MA: Harvard University Press.

Dodson, L. (1998). Don't call us out of name: The untold lives of women and girls in poor America. Boston: Beacon Press.

Edin, K., & Lein, L. (1997). Making ends meet: How mothers survive welfare and low-wage work. New York: Russell Sage Foundation.

Greenbaum, T. L. (1993). The handbook of focus group research. New York: Lexington Books.

Harris, K. M. (1996). Life after welfare: Women, work, and repeat dependency. American Sociological Review, 61, 407-426.

Iverson, R. R. (2000). TANF policy implementation: The invisible barrier, Journal of Sociology and Social Welfare, 27(2), 139-159.

Jargowsky, P. A. (1997). Poverty and place: Ghettos, barrios, and the American city. New York: Russell Sage Foundation.

Jarrett, R. L. (1994). Living poor: Family life among single parent, African-American women. Social Problems, 41, 30-49.

Julnes, G., Halter, A., Anderson, S., Frost-Kumpf, L., Schuldt, R. L., Staskin, F., & Ferrara, B. (2000). Illinois study of former TANF clients: Final report. Washington, DC: U.S. Department of Health and Human Services.

Kahn, A. J., & Kamerman, S. B. (1998). Big cities in the welfare transition. New York: Columbia University, School of Social Work.

Lederman, L. C. (1990). Assessing educational effectiveness: The focus group interview as a technique for data collection. Communication Education, 38, 117-127.

Loprest, P. (1999). Families who left welfare: Who are they and how are they doing? Assessing the new federalism (Discussion Paper No. 99-02). Washington, DC: Urban Institute.

Mead, L. M. (1992). The nonworking poor in America: The new politics of poverty. New York: Basic Books.

Moffitt, R., & Roff, J. (2000). The diversity of welfare leavers. In Welfare, children, and families: A three city study (Policy Brief 00-2). Baltimore: Johns Hopkins University Press.

Parrot, S. (1998). Welfare recipients who find jobs: What do we know about their employment and earnings? Washington, DC: Center on Budget and Policy Priorities.

Pavetti, L. A. (1993). The dynamics of welfare and work: Exploring the process by which women work their way off welfare. Unpublished doctoral dissertation, Harvard University, Cambridge, MA.

Personal Responsibility and Work Opportunity Reconciliation Act of 1996, P.L. 104-193, 110 Stat. 2105.

Quint, J., Edin, K., Buck, M. L., Fink, B., Padilla, Y. C., Simmons-Hewitt, O., & Valmont, M. E. (1999). Big cities and welfare reform: Early implementation and ethnographic findings from the project on devolution and urban change. New York: Manpower Demonstration Research Corp.

Schneider, R. L., & Netting, F. E. (1999). Influencing social policy in a time of devolution: Upholding social work's great tradition. Social Work, 44, 349-357.

Seccombe, K. (1999). "So you think I drive a Cadillac?"--Welfare recipients' perspectives on the system and on its reform. Boston: Allyn & Bacon.

Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park, CA: Sage Publications.

Sweeney, E., Schott, L., Fremstad, S., Goldberg, H., Guyer, J., Super, D., & Johnson, C. (2000). Windows of opportunity: Strategies to support families receiving welfare and other low-income families in the next stage of welfare reform. Washington, DC: Center on Budget and Policy Priorities.

Tickamyer, A. R., Henderson, D. A., White, J. A., & Tadlock, B. L. (2000). Voices of welfare reform: Bureaucratic rationality versus the perceptions of welfare recipients. Affilia, 15, 173-192.

Tweedie, J., Reichert, D., & O'Connor, M. (2000). Tracking recipients after they leave welfare. Washington, DC: National Conference of State Legislatures.

Wilson, W. J. (1996). When work disappears: The world of the new urban poor. New York: Alfred A. Knopf.

Steven G. Anderson, PhD, is assistant professor, School of Social Work, University of Illinois, 1207 West Oregon, Urbana, IL 61801; e-mail: sandersn@uiuc.edu. Anthony P. Halter, DSW, is associate professor emeritus, School of Social Work, University of Illinois, and Brian M. Gryzlak, MSW, is research assistant, Department of Epidemiology, University of Iowa, Ames. The authors gratefully acknowledge the Joyce Foundation for grant funding that supported this research.

Original manuscript received October 10, 2000

Final revision received December 3, 2001

Accepted August 6, 2002