[ Main Page of Report | Contents of Report ]
The WtW grants program is intended to help the least employable and most disadvantaged welfare recipients and noncustodial parents (NCPs) make the transition from welfare to work. The legislation placed particular emphasis on serving individuals facing the most difficult barriers to employment, including persons who have dropped out of high school, have low reading or math skills, have limited work experience, have been dependent on welfare for long periods, or have substance abuse and mental health problems. Targeting and enrollment were perhaps the most difficult early issues programs encountered in implementing WtW. The strict eligibility criteria included in the legislation required grantees to devote considerable effort to identify eligible persons and verify eligibility. When enrollment proceeded more slowly than expected in all the study sites, each program adopted strategies to increase their participation levels. Despite the enrollment challenges, however, programs in the study sites eventually approached their planned participation levels(12) and targeted populations generally considered to have serious barriers to employment.
WtW grantees were instructed by Congress to serve welfare recipients and other low-income parents who are the most disadvantaged. Grantees in the study sites generally met this challenge by targeting the hard-to-employ within the TANF population, although several programs focused on specific WtW-eligible subgroups within TANF. While the 1999 amendments allow WtW programs to also serve non-TANF low-income parents, study grantees rarely did. Although some administrators and staff expressed interest in expanding their population, they generally noted that they had to focus more on improving their enrollment of TANF recipients, especially those who met the 70 percent criteria. If they could obtain a high enough level of "70 percent" participants, then they could focus on other eligible groups. Similarly, the federal law allows programs to enroll noncustodial parents, and most of the grantees indicated they intended to serve this group, but aside from the Milwaukee NOW program very few NCPs were enrolled in the study sites.
In the study sites, there is generally no specific screening or targeting to decide which TANF recipients might enroll in a WtW grant-funded program. Instead, TANF recipients usually enter WtW programs in less formal ways. TANF staff typically have discretion to refer recipients to various programs, one of which might be funded by a WtW grant, and many participants enroll in this way. Many TANF clients also enter WtW programs as a result of direct outreach and recruitment efforts undertaken by the program staff. In both situations, staff at either the TANF office or the WtW program screen potential enrollees to determine if they meet the WtW eligibility criteria and whether they qualify under the 30 percent spending category or the 70 percent spending category, but no other special targeting is done.
Several WtW grantees, nonetheless, used indirect targeting strategies to focus on particular subgroups of WtW-eligible TANF recipients. A common indirect targeting method is the selection of service delivery contractors who have special experience. Even though services are open to all eligible persons, some study grantees (e.g., Chicago, Fort Worth, Nashville) in effect target special groups because they select service providers who specialize in serving certain groups such as homeless families, persons with mental or physical disabilities, individuals with limited education or English-speaking skills, persons who reside in certain neighborhoods, or persons from particular ethnic groups.(13)
A different targeting strategy involves focusing WtW resources on those TANF recipients who do not obtain employment through the regular TANF work program, presumably the harder-to-employ. If only those recipients who have first participated in a TANF work program can get into a WtW program, those in WtW are more likely than not to have barriers to employment. Philadelphia's TWC program, Nashville's Pathways, and the Yakima WtW program, for example, serve individuals after they have already participated in the official TANF work program but were not able to find a job.
Grantees were to develop employment-related programs for persons with serious barriers and who met the congressionally established eligibility criteria. Especially initially, programs had limited flexibility over deciding who could receive WtW-funded services. Most WtW grantees expected that referral from TANF offices would be the primary way participants entered their programs, and that TANF agencies would help verify eligibility. To meet these needs, grantees or programs funded by grantees established procedural agreements with the TANF agency. In many study sites there were problems with the referral process and WtW program administrators devoted considerable time and effort to developing eligibility determination, verification, and intake procedures to document the eligibility of each participant. Enrollment was slower and enrollment levels were lower than expected in all study sites, motivating grantees to undertake their own outreach and recruitment.
One of the most difficult aspects of WtW implementation related to low enrollment levels. A combination of factors contributed to enrollment problems in addition to the strict eligibility and spending requirements in the legislation.(14) For example, individuals with relatively serious personal and employment problems proved particularly difficult to enroll, given the other challenges in their lives.
In addition, the primary means by which WtW programs initially expected to obtain eligible individuals was through direct referrals from other local agencies, especially the TANF agency. All study programs, with the exception of the Milwaukee-NOW program (which targeted NCPs under probation or parole) intended to rely upon the TANF office to identify and refer WtW-eligible clients. For a variety of reasons, most of the study programs had difficulty getting enough referrals from TANF. In some states and localities, TANF policies and practices affected enrollment into WtW programs, sometimes inadvertently limiting the number of WtW participants. For instance, in some of the study sites, the WtW grantee and TANF agency had agreed that only individuals who had already participated in the TANF work program and had not been able to obtain a job would be eligible for WtW. In other sites, there was a de facto agreement that only those TANF recipients who were subject to work requirements were eligible for WtW. And in some places where TANF workers had discretion to refer clients to any of a number of employment-related programs in the community, workers were more inclined to refer to long-established programs with which they were more familiar rather than to a fairly new WtW program. Finally, in sites where the WtW grantee depended on TANF agencies to verify an individual's eligibility, confirmation sometimes took several weeks during which time some individuals lost interest. One implication of the various enrollment procedures followed is that the characteristics of the WtW-eligible individuals served vary across programs, as discussed below.
Over time, as the number of referrals was slower and lower than expected, WtW-funded programs adopted various methods to increase their participation levels. It quickly became evident that relying on referrals from TANF would not allow programs to reach their planned number of participants. Most of the study grantees, therefore, pursued active outreach and direct recruitment, which generally increased participation levels.
Recruitment approaches used included:
Since most of the study programs do some direct outreach, individuals enter programs in multiple ways. A large proportion are referred by TANF agencies, but according to local staff, perhaps half of WtW participants in many programs are recruited directly by the programs. In Fort Worth, for instance, about half of participants are recruited directly by WtW program contractors, who screen individuals to determine if they are likely to be eligible, and "reverse refer" those people back to the group orientation session for TANF recipients required by the TANF agency.
Program outreach is a common component of many programs operated by CBOs and other employment and training providers, so active outreach by WtW programs, many of which are operated by CBOs, is a logical response. It is, however, a procedural and operational change from what was originally planned, since most programs assumed that their participants would be referred to them by TANF agencies.
Given the enrollment difficulties, it took some programs considerably more time than expected to enroll the number of participants they had planned to serve. The extension of time given to grantees by Congress in the 1999 amendments provided grantees with an opportunity to increase their enrollments. After two years of operations, some of the study sites were still operating below their planned levels although nearly all expected that they would reach their original enrollment levels within the five-year period. A few programs had made a conscious planning decision to extend the period of their program given the flexibility they obtained with the 1999 amendments. And three of the programs had reached and exceeded their goals after two years.
By design, WtW grant-funded programs are relatively small in scale. The WtW-funded programs tend to serve a few hundred participants a year. However, both the enrollment goals and actual numbers of individuals served vary substantially across the sites. In part, the relatively small scale of the programs reflects the fact that many of the programs are operated by CBOs and other nongovernment providers, rather than by large government agencies.
Across the 11 study sites, slightly over 20,000 persons were expected to participate over the three-year period originally allowed. The plans ranged from a low of 510 over three years for the West Virginia-HRDF grantee (serving 29 primarily rural counties) to a high of 9,000 for the Chicago grantee (across the 19 programs) (Chart III.1). The largest single program in the study sites is the one operated in Philadelphia by TWC, which planned to serve 3,000 persons. The average (mean) participation goal across the 11 sites for the original three-year period of WtW was 1,853 individuals.(15) Under the terms of the WtW legislation, grantees are allowed five years in which to spend their grant funds.(16)
As of April 2001, none of the study sites had yet completed the five-year period of their grants. From the start of the programs (generally in late 1998 or early 1999) through April 2001, a total of 18,175 individuals had enrolled in WtW programs across the 11 sites. The numbers served (through April 2001) ranged from over 9,000 in Chicago to a low of 250 in the Milwaukee NOW program. The average number of participants served per site across the 11 sites was 1,652 individuals. With the exception of the two largest sites (Chicago and Philadelphia-TWC), each of the other study sites had served fewer than 1,000 individuals. Chicago and Philadelphia-TWC account for well over half of the enrollees across the 11 sites.
About two years after beginning operations, WtW programs were approaching their planned enrollment levels. By April 2001, three of the 11 study grantees had reached or exceeded their planned goals for participation--West Virginia-HRDF, Philadelphia-TWC, and Chicago. Two other grantees--Yakima and Indiana-RVR--were nearing their overall goal (94 percent and 88 percent of their goals, respectively), and Fort Worth and Phoenix had reached about 70 percent of their goals. Three of the remaining study sites had reached about half of
Chart III.1
Participation Goals Versus Actual Participation, by Study Site
Source: Management Information System data from the local programs.
their overall enrollment goals--Boston, Nashville, and the JHU sites in Baltimore and St. Lucie--all of which had consciously decided to extend their timeframe. The final site, the Milwaukee NOW program, had reached only about one-fourth of the site's original participation goal and was continuing to experience very slow enrollment.(17) The Milwaukee program has had ongoing difficulty enrolling and retaining noncustodial parents (NCPs), a population that has proven to be difficult to recruit in many programs across the country.(18)
The WtW grantees in the study sites designed programs intended to serve hard-to-employ welfare recipients and NCPs. Baseline information collected on samples of enrollees in nine of the 11 study sites for the evaluation suggests that the study programs have concentrated on TANF recipients with characteristics often associated with employment problems. High proportions of participants across the sites are African-Americans and Hispanics; and in some of the sites relatively large numbers have less than 12 years of education and self-reported disabilities. A forthcoming report from the evaluation will focus in detail on the characteristics of participants in comparison to all TANF recipients, as well as their employment outcomes.(19)
It is useful here to simply describe the characteristics of participants in the study programs, since they provide additional insight into the types of programs operating in the study sites and the variation across programs--the topic of the following two chapters.
Similar to TANF recipients nationally, most WtW participants are between 18 and 44 years of age, and the vast majority are women. The median age of participants when they enrolled in a program in study sites where baseline information was collected was between 28 and 34. Across the study sites, Phoenix serves a slightly younger group and JHU-Baltimore and West Virginia-HRDF serve a somewhat older group of participants. In terms of gender, although staff in nearly every site indicated they had hoped to serve NCPs, in only three of the study sites do males account for more than 15 percent of participants--Milwaukee-NOW (95 percent male), Yakima (25 percent), and West Virginia-HRDF (15 percent).
Many WtW participants in the study sites have characteristics often associated with disadvantages in the labor market--minority status, limited education, and mental and physical disabilities. With the exception of West Virginia's HRDF program, most participants in the study sites where baseline information was collected are nonwhite, suggesting they may face special challenges in the job market to the extent that racial discrimination exists. Fairly high proportions of participants are Hispanic in four of the study sites--in Phoenix (49 percent), Yakima (36 percent), Boston (36 percent) and Fort Worth (27 percent). Over 80 percent of the participants in Philadelphia-TWC, Nashville, Chicago, and Milwaukee-NOW are African-American (Chart III.2). Some sites, such as Boston, though, specifically developed programs to target certain groups by, for example, contracting with service providers that specialize in serving Hispanics or have offices or programs in neighborhoods with high minority populations.
Many WtW participants also have weak educational backgrounds. Across all sites in which baseline information was collected, about 46 percent of all participants have less than 12 years of education, similar to the proportion of TANF adults nationwide according to HHS data. Some WtW programs, however, are serving considerably more persons with less than a high school education (Chart III.3). In Philadelphia-TWC, Phoenix, and Yakima, over 55 percent of participants lack a high school diploma or GED. As discussed in the following chapter, some of the study programs developed strategies to address the low skills of their participants by
Source: Program Baseline Information Forms. Yakima: Program Management
Information System data.
National data: "TANF Report To Congress," Department of Health and
Human Services, 2000.
Note: Administrative data unavailable for Indiana-RVR and JHU Long
Beach sites.
Source: Program Baseline Information Forms. Yakima: Program Management
Information System data.
National data: "TANF Report To Congress," Department of Health and
Human Services, 2000.
Note: Administrative data unavailable for Indiana-RVR and JHU Long
Beach Sites.
implementing pre-employment components that include basic skills remedial instruction, often in conjunction with employment activities.
There is also some evidence that many WtW participants have disabilities that might interfere with work. Between 25 and 46 percent of participants across the sites reported at the time of enrollment that they had medical, physical, emotional, substance abuse, and/or other disabilities (Chart III.4). In Fort Worth, Nashville, and West Virginia-HRDF, over 40 percent report having a disability. Some programs specifically target those with disabilities; others incorporate services for those with special diagnostic or treatment needs into their overall program options. Some grantees, for example in Chicago and Fort Worth, contract with service
Source: Program Management Information Systems.
Note: Administrative data unavailable for Yakima, Indiana-RVR, and
JHU Long Beach sites.
providers that specialize in serving individuals with particular disabilities or substance abuse problems.
The WtW-funded programs in the study sites thus appear to be targeting individuals with potentially serious employment problems. Future reports from the evaluation will address whether programs are serving individuals similar to the general TANF population or whether they have targeted certain groups. What seems clear from the basic program data presented here is that many participants in the study sites appear to have employment problems and that some of the programs in these sites specifically deal with individuals with certain problems, as discussed in more detail in the following two chapters.
12. Surveys of all grantees nationwide in 1999 also found that in during the first year, enrollment levels were below planned levels and that a year later the pace of enrollment had increased. See Irma Perez-Johnson and Alan Hershey, Early Implementation of the Welfare-to-Work Grants Program: Report to Congress, Princeton, NJ: Mathematica Policy Research, Inc, March 1999; and Irma Perez-Johnson, Alan Hershey, and Jeanne Bellotti, Further Progress, Persistent Constraints: Findings from a Second Survey of the Welfare-to-Work Grants Program, Princeton, NJ: Mathematica Policy Research, Inc., April 2000.
13. One possible implication of such specialization could be that other eligibles might be less likely to be serviced. More detailed analysis of the participants in the eleven evaluation study sites will appear in a separate report from the evaluation by Fraker, et al., forthcoming 2003.
14. A more detailed description of outreach and recruitment efforts is in Lynne Fender, Demetra Nightingale, and Alan Hershey, "Welfare-to-Work Grant Programs Tackle Recruitment Challenges," Program Brief, Princeton, NJ: Mathematica Policy Research, Inc. 2000.
15. Chicago and Philadelphia-TWC account for over half the planned participants across the sites. Excluding them, mean planned participation per site was about 930.
16. The 1999 legislative amendments extended the time from three years to five years.
17. Enrollment continued to be very slow in the JHU-Long Beach site as well, but that site is not included in Chart III.2 because that site was not included in the administrative data collection by the evaluation.
18. As discussed in greater detail in another report prepared by Mathematica and The Urban Institute focusing on WtW programs serving NCPs, there are a number of special challenges to recruitment of NCPs. See Karin Martinson, John Trutko, and Debra Strong, Serving Noncustodial Parents: A Descriptive Study of WtW Programs. A Report from the Evaluation of the Welfare-to-Work Grants Program conducted by Mathematica Policy Research, Inc. Washington DC: The Urban Institute 2000.
19. A summary of selected characteristics of participants by study site appears in Appendix B.
Main Page of Report | Contents of Report
Home Pages:
Human Services Policy (HSP)
Assistant Secretary for Planning and Evaluation
(ASPE)
U.S. Department of Health and Human Services
(HHS)