Understanding the Costs of the DOL Welfare-to-Work Grants Program

IV. Welfare-to-Work Costs in Context

[ Main Page of Report | Contents of Report ]

Contents

1. WtW Costs Compared to Earlier Programs
   1. WtW Costs Reflect Targeting of Hard-to-Employ Individuals
   2. WtW Favors Work but Still Emphasizes Skill Building
   3. Focus on Employment for the WtW Population Has Required More Intensive Services
2. Implications for Future Programs

Endnotes

Over the past 40 years, welfare policies have increasingly emphasized work. Several generations of programs to help welfare recipients prepare for and enter employment have been implemented, reformed, rethought, and replaced. Programs funded under the WtW grants program are another step in this evolution. Comparison of program costs for WtW and earlier programs can suggest useful implications about the costs of future programs as they might emerge. In this final chapter, we address two questions:

1. How do WtW costs compare to those for similar interventions, and what are the reasons for differences?

2. What are possible implications of our findings for future programs?

A. WtW Costs Compared to Earlier Programs

The costs of WtW programs can be best understood against the backdrop of costs in programs of the past three decades. We compared estimated cost for the WtW programs, presented in earlier chapters, to cost estimates for 10 previously evaluated efforts to promote employment among welfare recipients and populations at risk of becoming welfare recipients. These earlier programs were part of distinct welfare policy regimes or important demonstration efforts: (1) four Work Incentives (WIN) models, (2) four Job Opportunities and Basic Skills (JOBS) models, (3) the Supported Work demonstration, and (4) the Minority Female Single Parent (MFSP) demonstration.⁽¹⁾

The costs of WtW programs appear to fall in the middle of the range of these programs' costs. As with WtW programs, these earlier models represented a variety of program strategies and a range of costs per participant (Table IV.1). Compared to these efforts, WtW programs were neither the least costly nor the most expensive (Figure IV.1).

Three characteristics of WtW programs help explain where they fall in the historical range of program costs. First, WtW programs targeted hard-to-employ individuals--earlier interventions did not always focus on, or even serve, such individuals. Second, WtW programs favored work over education but still emphasized human capital development in a different form. Third, WtW programs needed more intensive case management and related services to maintain this simultaneous focus on employment and human capital development with a hard-to-employ population.

1. WtW Costs Reflect Targeting of Hard-to-Employ Individuals

WtW programs as a whole represent a targeted effort to meet the needs of hard-to-employ individuals. After passage of PRWORA, Congress expected that some individuals would need intensive assistance to secure employment and make strides toward self-sufficiency. Hence, WtW funds were made available to supplement the work-first efforts of welfare agencies.

Some earlier employment programs intentionally omitted, or at least did not focus on, the hard to employ, and this difference accounts in part for their lower costs. For example, WIN programs--which, on average, cost $2,147 per participant, compared to $3,607 for WtW--targeted the most employable AFDC recipients (Figure IV.1). Even when program registration mandates applied more broadly, many individuals were deferred from participation due to transportation, health, or other problems (Levitan et al. 1971, Pacific Consultants 1976; and O'Neill 1990). Moreover, when program openings were scarce, priority was often given to the most employable.

Figure IV.1
Range of Average Costs per Participant for Wtw and Comparable Programs

Expansion of target populations has been reflected in increased costs. Over time, programs to increase employment among welfare recipients have broadened the populations they serve. Instead of focusing on the immediately employable only, groups were included that had less work experience or more potential barriers to employment (such as limited education and longer welfare spells). For example, JOBS programs were specifically required to focus on long-term AFDC recipients, not just on the most employable recipients. Costs were higher than for the earlier WIN programs, averaging $3,327, even when, in practice, many hard-to-employ individuals were still exempted from participation.⁽²⁾

Differences in target population, however, do not fully explain program cost differences. WtW programs explicitly targeted the most disadvantaged TANF recipients, including many who might have been exempted from JOBS participation requirements, yet average WtW costs were similar to the average costs of JOBS programs (Figure IV.1). Moreover, WtW costs were substantially lower than those for programs operated as part of the National Supported Work Demonstration ($11,572), which also targeted hard-to-employ individuals. The reason for these apparent discrepancies is that the differences between WtW and both JOBS and Supported Work also reflect important differences in the services offered or emphasized.

2. WtW Favors Work but Still Emphasizes Skill Building

Programs that emphasize quick entry into employment and provide mostly job search assistance and placement services consistently have had relatively low costs, regardless of their target population. With mean costs of $502 and $2,131 per participant, respectively, the WIN and JOBS "labor force attachment" programs were the least costly among related efforts (Table IV.1).Similarly, the "enhanced direct employment" WtW programs ($3,559) were less costly, on average, than the $4,077 average across all "transitional employment" WtW programs (Table III.6).

Programs serving harder-to-employ participants have generally provided more than basic employment services. Programs have offered education and training, work experience, and other intensive service components aimed at increasing skills and overall human capital to individuals for whom job search and placement assistance alone did not, or were not expected to, result in rapid or adequate employment. Not surprisingly, programs have had higher costs when they offered expanded services such as adult basic education (ABE), English as a Second Language (ESL) instruction, and occupational training. For example, WIN I ($3,384) and JOBS human capital ($3,934) programs emphasized such education alternatives for their less work-ready participants, and cost more, on average, than WIN ($502) and JOBS labor force attachment programs ($2,131), their rapid-attachment counterparts (Table IV.1).

WtW programs have not emphasized traditional education and training activities, yet their costs are similar to programs that did. Instead of emphasizing education and training, WtW programs sought to build participants' foundation for employment through direct work experience and other activities more directly related to employment. WtW programs emphasized structured job readiness classes, work experience, and skills upgrade activities wrapped around work hours, as a way to enhance participant human capital. Thus, the costs of WtW programs overall ($3,607) were roughly similar to those for education-based programs (see above) because both emphasized skill building for their hard-to-employ participants.

However, WtW programs that offered work experience (at an average cost of $5,098 per participant--Table III.2) were, on average, more costly than all WIN and JOBS programs, even though WIN and JOBS also typically offered work experience as a preemployment option (Table IV.1). This was because WtW programs generally paid or subsidized participant wages and often provided work experience to a substantial portion of their participants, while WIN and JOBS programs did not pay participant wages and enrolled few participants in work experience.⁽³⁾ Furthermore, work experience in WtW lasted longer--six months or more in seven of the nine WtW programs that offered work experience, compared to a three-month limit under both WIN and JOBS.

None of the WtW work experience programs, however, were as costly as the National Supported Work Demonstration programs, which provided highly structured and closely supervised work experience to their hard-to-employ participants. Supported Work was particularly costly because program operators had to set up and maintain business enterprises that could provide appropriate work experience opportunities to groups of participants at any given time (Hollister et al. 1984). In contrast, the WtW programs that offered work experience relied primarily on placements in the public or nonprofit sectors. Hence, even the most structured WtW work experience program, Philadelphia-TWC, had an average cost of $6,641 per participant, a full 42 percent less than the average Supported Work program ($11,572).⁽⁴⁾

3. Focus on Employment for the WtW Population Has Required More Intensive Services

Because they are dealing with the WtW target population and still focusing on employment, most WtW programs needed more intensive case management and more specialized staff than earlier programs. Both of these shifts have contributed to higher costs. Under WIN, welfare agencies primarily relied on eligibility workers to assess, then refer, AFDC recipients for placement services, which were provided mostly through local employment service offices (Levitan et al. 1972; and Mitchell et al. 1979). Under JOBS, welfare offices expanded case management to include more assessment, closer case supervision, and employment counseling, sometimes provided by specialized workers (Riccio et al. 1994).

Compared to these earlier programs, the core services offered by WtW represent a further intensification of individualized case management and other specialized services, for several reasons. First, all the WtW programs we examined offered postplacement services to help participants retain their jobs and potentially advance to better positions. Earlier programs usually did not provide such follow-up services. Second, some WtW programs had job developers and other placement staff who managed comprehensive or systematic placement components, and some programs even formed partnerships with local businesses or industries to fill positions for specific employers. Finally, some WtW programs were explicitly designed to be consistent with TANF work activity requirements. Hence, job readiness, case management, and employment activities were usually more structured and closely managed by staff, so that programs could insure that participants spent as many as 35 hours weekly in allowable work activities.

[Go To Contents]

B. Implications for Future Programs

There are no current plans to reauthorize the WtW grants program, and whether states will continue to provide similar supplemental services to hard-to-employ TANF recipients is unknown. However, recent developments suggest that states may have strong incentives to continue to focus on hard-to-employ individuals or even to intensify past efforts. Earlier steep declines in TANF caseloads, which enabled many states to meet aggregate work requirements, have leveled off, and some states have experienced slight increases--possibly as a result of the economic slowdown of late 2001. Furthermore, the President's welfare reform proposals made in early 2002 urge Congress to require substantially more welfare recipients to get jobs and work longer hours.⁽⁵⁾

States' future ability to meet existing or expanded work mandates may depend on their ability to engage not only the most work-ready, but also a substantial proportion of hard-to-employ individuals on their caseloads, in approved work activities. Depending on the strength of the economy and other factors, states may find that a high proportion of the individuals on their caseloads need assistance meeting expanded work requirements. As our review has shown, doing "what it takes" to help these individuals is likely to have cost implications.

While it is impossible to predict the direction of future efforts, we see at least two adjustments that the operators of programs that target the hard-to-employ may decide to pursue. First, program operators may further intensify structured services and case management to help hard-to-employ individuals prepare for, secure, and succeed in employment, while simultaneously meeting work activity requirements. Future programs that target the hard-to-employ and use this approach may cost as much as, or more than, WtW programs. Second, state calls for increased flexibility in program design may allow programs to place more emphasis on education and training, while still retaining a work-first focus. The cost implications of this second potential shift are more uncertain.

Costs of education and training are more uncertain in part because WtW programs did not emphasize them. Long-term education and training were excluded as approved work activities under TANF, and short-term education and training were limited, at least initially, by the WtW regulations. Even after these restrictions were relaxed, and some programs made education and training services available, WtW participants did not use them extensively.⁽⁶⁾ This was especially true when education and training activities had to be pursued concurrent with employment, instead of as part of the structured services offered by WtW programs.

Thus, the WtW process and cost analyses leave considerable uncertainty about how more emphasis on basic or occupational training would affect costs. Integrating education and training into structured services could increase participation in such activities and, therefore, the costs of programs that target hard-to-employ individuals.⁽⁷⁾ However, to the extent that new policies require participants to pursue education and training activities concurrent with employment, participation may continue to be limited (as it has been in WtW programs) and cost increases therefore less pronounced.

[Go To Contents]

Endnotes

1.  Appendix B provides details on the services offered by these programs and their costs.

2.  JOBS participation was mandatory for those without children under age 3, but individuals could be deferred for illness, remoteness from the program, lack of child care, or other acceptable reasons. This was similar to WIN deferral policies, except that the age of the youngest child under WIN was age 6, rather than age 3 (U.S. General Accounting Office 1999).

3.  WtW payments of wages or stipends were partly offset by decreases in TANF cash assistance, but usually not on a dollar-for-dollar basis. Thus, compared to earlier unpaid work experience programs, WtW programs offering paid work experience would still be more costly from a societal perspective. WtW regulations required WtW programs to compensate participants for the hours spent in work experience. To be compensated, participants could work up to the number of hours calculated by dividing their TANF grant by the minimum wage (the same strategy as many of the WIN and JOBS programs that we have characterized as unpaid), but the payment of wages was explicitly preferred. Yet, the greater use of subsidized work experience under WtW may also reflect a shift toward providing more work experience and other opportunities for transitional employment (such as OJT) in the private rather than the public or nonprofit sectors. For example, WtW programs like Chicago-Pyramid or West Virginia-HRD had to pay or, at least, partially subsidize participants' wages to secure such placements.

4.  As demonstration programs, both Supported Work and MFSP programs may also have incurred some additional costs compared to WIN, JOBS and WtW, for several reasons. First, because they were not explicitly linked to welfare, the programs had to build their own referral linkages to and from other community resources. Second, both demonstrations also included many participants who were not receiving AFDC, so programs had to make a wide range of services available to them.

5.  Key components of the President's welfare reform agenda are (1) increasing minimum work requirements so that, by 2007, 70 percent of welfare families are required to participate in work and other activities designed to help them achieve self-sufficiency; and (2) requiring welfare recipients to work 40 hours per week, either at a job or in programs designed to help them achieve independence (www.whitehouse.gov/news/releases/2002/02/20020226.html).

6.  Nightingale, et al. 2002.

7.  The cost estimates for GAIN programs, which emphasized education and training, can be used to estimate the magnitude of this potential increase in costs. The mean cost per participant for GAIN program services other than education and training (namely, orientation, assessment, appraisal, and job search) was $1,610. Program costs increase to $2,996 when the average costs of basic education (ABE and ESL) are added and to $4,190 when the costs of postsecondary and vocational education are also added (Riccio et al. 1994). Thus, adding education and training activities, if pursued by a high proportion of participants, could add $3,000 or more to the per-participant costs of programs.

Where to?

Top of Page | Contents
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)