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

III. Analyzing WtW Program Costs

[Main Page of Report | Contents of Report]

Contents

1. Variation in Total Costs
2. Variation in WtW Resource Allocation
   1. Outreach and Recruitment Were Important WtW Investments for Some Programs
   2. Job Readiness, Placement, and Case Management Are the Core of WtW Services
   3. Emphasis on Work Experience Differentiated Some WtW Programs
   4. WtW Transportation Assistance Helped Fill a Gap in Transitional Support
3. Variation in Costs per Participant
   1. Sources of Cost Variation Go Beyond Program Model, Scale, and Duration of Participation
   2. Costs Differences Reflected Programs' Emphasis on Work Experience and Postemployment Support
4. Variation in Costs per Placement
   1. Cost per Placement Clearly Reflects Job-Matching Strategy
   2. Urbanicity and Prevailing Unemployment Rates Had No Obvious Effect on Costs per Placement
   3. Participant Characteristics May Have Influenced WtW Costs per Placement

Endnotes

Although the WtW programs included in the cost analysis were all part of the same federal initiative, the flexible nature of WtW as a funding stream resulted in programs with unique characteristics. Some programs operated in large cities, while others served rural areas. Some were independent, self-standing programs; others were part of larger-scale WtW initiatives. The programs offered and emphasized different combinations of services to their target populations. These diverse characteristics all influenced the programs' costs.

In this chapter, we explore the rich variation in characteristics of WtW programs to glean insights into their cost implications. Section A describes the variation in total costs of WtW programs. Section B discusses important similarities and differences in the allocation of total costs across WtW program components or services. Sections C and D explore variation in average costs per participant and average costs per placement, respectively.

A. Variation in Total Costs

Estimates of total costs help convey the overall scale and scope of a program's operations. In comparisons of different programs, exploring scale is important because it can reveal important differences in their aims or operations.

Our cost estimates reflect the significant amount of resources that was invested in the WtW initiative. On an aggregate basis, operating for one year the 18 WtW programs included in the analysis cost an estimated $22.6 million. Nevertheless, total costs per program for one year of WtW operations varied widely, ranging from just over $200,000 to more than $7 million (for Philadelphia-TWC, not shown in Figure III.1). The WtW programs also varied in their total

Figure III.1
Total Wtw Program Costs, by Total Participation

participation, ranging from just under 100 to more than 2,000 (for Philadelphia-TWC, not shown in Figure III.1) individuals ever active during the programs' cost analysis years.

Much of this variation was by design and reflected the diverse organizational context of the WtW programs. For example, the WtW programs operated by Maximus, E&ES, and Catholic Charities in Chicago, Philadelphia-TWC, and Phoenix-EARN all served large metropolitan areas and were developed specifically to help large numbers of WtW-eligible individuals move into employment. In contrast, the smaller WtW programs--for example, those in Boston and Fort Worth--tended to be part of larger-scale WtW initiatives that involved multiple providers, each offering tailored services to a relatively small number of WtW-eligible participants.

[Go To Contents]

B. Variation in WtW Resource Allocation

The WtW programs allocated resources to specific services and activities. On an aggregate level, examining this cross-component allocation can reveal important characteristics of the WtW initiative as a whole. Across individual programs, examining the variation in allocation of costs across services or components helps us understand important differences in their emphasis or operations.

1. Outreach and Recruitment Were Important WtW Investments for Some Programs

Outreach and recruitment costs represented only five percent of the costs for the average WtW program (Figure III.2), as only eight programs incurred costs for such activities. Although devoting resources to outreach and recruitment was not always planned, these eight programs spent from $27,426 to more than $274,000 to try to identify and recruit eligible WtW participants (TableIII.1). Their outreach and recruitment costs translated into $92 (Philadelphia-TWC) to $1,302 (Phoenix-EARN) per enrollee during the cost analysis year.

Figure III.2
Average Allocation of Total Costs Across WtW Components

Most of the outreach and recruitment costs these programs incurred represented the time some WtW staff had to devote explicitly to participant recruitment. Phoenix-EARN and Philadelphia-TWC, the programs with the largest outreach and recruitment costs, had dedicated outreach staff whose principal responsibility was to help identify and recruit prospective WtW participants. In the other programs, directors, case managers, and other WtW staff members reported spending a fair amount of their time visiting welfare offices and other potential referral sources to try to increase referrals by "talking up" their programs.

The remaining 10 programs could, in general, rely on well-established relationships with the local TANF agencies to get enough WtW referrals. While the outreach and recruitment costs for these programs are reported as zero, staff members from many of these programs reported investing small but indeterminate amounts of time in general outreach activities. For example, program directors, case managers, or other WtW staff in Chicago or Fort Worth periodically visited welfare offices and other potential referral sources to maintain these important relationships. Thus, even programs with minimal outreach and recruitment costs considered these efforts important.

Unanticipated recruitment difficulties arose from the restrictiveness of the initial WtW criteria, falling TANF caseloads, and the availability of alternative WtW or other programs in some areas.⁽¹⁾ In spite of their outreach efforts, some of the WtW programs included in our analysis may have had the capacity to serve additional individuals at a relatively low marginal cost, which could have reduced their average costs somewhat.

2. Job Readiness, Placement, and Case Management Are the Core of WtW Services

Core WtW services included such activities as job readiness classes; intake, assessment, and general preemployment case management; job development and placement services; and postplacement followup. This subset of services was present in all of the WtW programs examined and accounted for between 34 percent (Philadelphia-TWC) and 84 percent (Fort Worth-Women's Center) of the total costs for the programs (Figure III.3).⁽²⁾ In the average WtW program, almost two-thirds of total costs were associated with activities aimed at engaging eligible participants, helping them prepare for and secure unsubsidized employment, and supporting them after such placements (Figure III.2).

3. Emphasis on Work Experience Differentiated Some WtW Programs

Both the availability and extent of use of paid work experience were characteristics that differentiated WtW programs. While paid work experience accounted for 16 percent of total

Figure III.3
Percent of Total Costs Spent on Core Versus Other Wtw Services

costs in the average WtW program (Figure III.2), only half of the programs in the cost analysis offered their participants paid work experience. Moreover, the programs that offered paid work experience emphasized such placements to varying extents. For five programs (Chicago-Easter Seals, Chicago-Pyramid, Philadelphia-TWC, West Virginia-HRD, and Yakima-FWC) costs related to paid work experience represented the largest proportion of WtW costs, ranging from 37 to 41 percent (Table II.2). Other programs devoted fewer resources to work experience or did not offer such placements.

4. WtW Transportation Assistance Helped Fill a Gap in Transitional Support

Consistent with the authorizing legislation, WtW resources were used for support services primarily as a complementary strategy. In the average program, support services accounted for only six percent of total WtW costs (Figure III.2). However, 14 of the 18 programs examined incurred some WtW support service costs. Among these programs, the support service most commonly provided was transportation assistance for the period following placement in unsubsidized employment.

The main reason for WtW's limited emphasis on support services was that most participants had access to TANF-funded support services to help ease their transition from public assistance to employment.⁽³⁾ These transitional benefits usually included assistance with child care, health care, and transportation. However, many programs were limited in their ability to use WtW funds to provide support services to participants, because they had ceilings on support service expenditures or restrictions on allowable expenditures.⁽⁴⁾ Therefore, the fact that, on average, only six percent of total WtW costs were associated with support services does not mean that WtW participants only needed such assistance to this extent.

The programs that provided WtW-funded transportation assistance reported that TANF's transportation benefits typically ended while participants were still engaged in WtW activities or shortly after job placement. Transportation difficulties could jeopardize participants' employment, so WtW-funded transportation assistance helped fill an important gap in transitional support.

[Go To Contents]

C. Variation in Costs Per Participant

As we just saw, the WtW programs in the cost analysis offered different mixes of services. Since specific services require different combinations of resources (for example, more or less staff time, space, materials, or equipment), differences in the programs' service mix could contribute to differences in their costs. Even when examining WtW programs that offered a similar mix of services to their clients, expenditure patterns can reveal important differences in where programs truly concentrated their efforts or in how services were delivered.

Unit measures of cost allow us to explore these differences most clearly, because they factor out the influences of program scale. In particular, estimates of cost per participant describe the amount of resources that programs spent, on average, to serve a single participant. These estimates make it possible to compare the magnitude of WtW expenditures on individual participants, overall or for particular services, across programs.

The average WtW program spent $3,607 to serve each participant. As with other estimates, however, there was substantial variation in costs per participant across the programs. The least costly program spent $1,887 per participant (Fort Worth-Women's Center) while the most costly (Philadelphia-TWC) spent $6,641.

1. Sources of Cost Variation Go Beyond Program Model, Scale, and Duration of Participation

The WtW process evaluation identified three basic program models, representing distinct employment philosophies and approaches to providing WtW services.⁽⁵⁾ As a starting point for our analyses, we classified the WtW programs into these same model categories (Table III.2):

· Enhanced Direct Employment (EDE). These WtW programs focused on moving WtW participants into unsubsidized employment as soon as possible. They were "enhanced" because, unlike traditional "rapid attachment" interventions, they complemented placement services with pre-employment job readiness activities, individualized counseling and support, and extended followup after employment. Among the programs included in the cost analysis, another distinguishing characteristic of enhanced direct-employment programs was that they helped their WtW participants gain access to a wide range of locally available employment opportunities.

· Transitional Employment (TE). These programs tried to gradually and systematically improve their participants' employability. Their objective was for participants to ultimately--not immediately--obtain unsubsidized employment. The WtW programs in the cost analysis used three distinct approaches to this systematic building of employability skills:

1. Paid Work Experience. In some transitional programs, WtW clients were generally expected to participate in work experience as an intermediate step before unsubsidized employment.
2. Employer-Tailored Programs. Some transitional programs were tailored to the needs or expectations of particular employers. These programs aimed to help participants develop the skills necessary for jobs with a given employer or a group of employers with similar characteristics.
3. Small Steps. Only one of the WtW programs in the cost analysis used this approach, which was distinct from other transitional programs. The Nashville-Pathways program did not have participants follow a particular sequence of activities, nor did it offer an explicit menu of WtW services. Instead, case managers and participants together identified a highly individualized set of employment-related objectives and activities for the participant to pursue. Importantly, "small steps," such as time spent arranging child care or resolving housing issues, were counted as WtW activities and helped participants meet their TANF work activity requirements.

. Postemployment Skills Development (PSD). The two JHU programs in the cost analysis emphasized supporting WtW participants after they had secured an unsubsidized job. They focused on providing services and assistance to help participants retain unsubsidized employment and advance to better jobs, in the hope of improving their prospects for long-term self-sufficiency.

On average, differences in the cost per participant of WtW programs were in the direction their model classifications would suggest. PSD programs aimed to serve primarily people who had already found jobs, and thus focused exclusively on postemployment services. They cost less per participant, on average, than EDE and TE programs, which provided both pre- and postemployment services (Table III.2). Similarly, EDE programs, which emphasized a quick entry into unsubsidized employment, were less costly, on average, than TE programs, which emphasized more gradual, systematic acquisition of employability skills.

Costs varied considerably, however, among programs in any given model. As a result, some EDE programs could cost less, as much, or more than some TE programs (Table III.2). Similarly, PSD programs were not cheaper than all EDE or TE programs. For programs belonging to the same WtW model, average costs per participant were not necessarily lower if they operated on a larger scale or had lower average durations. This suggested that factors other than program model, scale of operations, or overall duration of participation contributed to the cost differences across WtW programs.

2. Costs Differences Reflected Programs' Emphasis on Work Experience and Postemployment Support

The programs in each WtW model shared similar priorities and, in general, offered a similar mix of services, but they emphasized different individual services or program components. This variation in emphasis led to notable differences in average costs per participant, even for WtW programs in the same model category. In particular, differences in the costs per participant of WtW programs reflected their relative emphasis on, and the structure of, work experience and postemployment components.

a. Costs Were Higher, but Still Varied, in Programs That Offered Paid Work Experience

Programs that included paid work experience generally had higher costs per participant than those that did not (Figure III.4). However, the costs per participant of WtW programs with such a component still varied widely, reflecting specific features of these components. In particular, costs varied based on (1) the prevalence of participation in paid work experience among WtW participants, (2) wage payments and other costs (for example, payroll taxes) incurred on behalf of work experience participants, and (3) the overall intensity of work experience activities, as defined by participation duration and the extent of job readiness services offered with work experience (Table III.3).

Figure III.4
Allocations of Costs per Participant to Main Wtw Program Components

Table III.3
Factors Influencing the Per-participant Costs of WtW Work Experience

Programs with a Work Experience Component (Average Cost per Participant of Work Experience)(a)	Prevalence of Participation	Wage Payments and Other Costs Incurred on Behalf of Participants	Intensity of Work Experience Activities
			Duration of Participation	Integration of Job Readiness Services
Philadelphia-TWC ($3,101)	Most. Participants completing 2-week orientation/job readiness component	High. $5.15 per hour, for up to 25 hours weekly; $50 monthly payment to mentors	Long. Until assessed work ready, up to 6 months	High. Career development training for 10 hours weekly while in transitional work; weekly worksite visits to assess performance and work readiness
Chicago-Pyramid ($2,244)	Most. Participants completing 4-week job readiness workshop	High. OJT positions paid between $6.25 and $8.25 per hour; 30 hours per week	Short. 4 to 6 weeks	Limited. No structured job readiness concurrent with work experience
Yakima-FWC ($1,824)	As needed. Only participants failing to secure employment after 12 weeks of structured job search	High. $6.72 per hour (minimum wage in Washington), up to 20 hours weekly	Long. Until participant secured unsubsidized employment, up to 9 months	High. Individualized work readiness and job search support during work experience
West Virginia-HRD ($1,430)	All. Participants completing 4-week job readiness workshop	Modest. - Mostly OE: $1.60 per hour stipend for 25 to 35 hours/week (as determined by TANF benefits/minimum wage) - Some OJTs: WtW paid up to half of entry-level wages (no range available)	Modest. Duration varied according to participants' work readiness: - Less work ready: OEs at nonprofits for up to 6 months - More work ready: OEs at for-profits for up to 1 month or OJTs for up to 6 months	Limited. No structured job readiness concurrent with work experience
Chicago-Easter Seals ($1,266)	As needed. Only participants completing 4-week job readiness/search workshop without employment	High. Industrial workshop participants paid on piecemeal basis, $6 per hour average, 15 hours weekly	Short. Until placed in subsidized or unsubsidized employment, up to 4 weeks	High. Job readiness training for 15 hours per week while in industrial workshop
Chicago-Catholic Charities ($1,084)	As needed. Only participants completing 2-week job readiness class plus 3 weeks of structured job search without unsubsidized job	High. $5.15 per hour, up to 30 hours weekly	Long. Until participant secured unsubsidized job, up to 6 months	Limited. No structured job readiness concurrent with work experience
Yakima-OIC ($937)	As needed. Only participants failing to secure employment after 12 weeks of structured job search	High. $6.72 per hour (minimum wage in Washington), up to 20 hours weekly	Long. Until participant secured unsubsidized employment, up to 9 months	High. Individualized work readiness and job search support during work experience
Yakima-PFP ($883)	As needed. Only participants failing to secure employment after 12 weeks of structured job search	High. $6.72 per hour (minimum wage in Washington), up to 20 hours weekly	Long. Until participant secured unsubsidized employment, up to 9 months	High. Individualized work readiness and job search support during work experience
Nashville-Pathways ($243)	Few. Only participants needing work experience to fulfill 40-hour/week TANF work requirement	High. $5.25 per hour, up to 20 hours weekly	Modest. Up to 3 months	Limited. No structured job readiness concurrent with work experience
a. Average per participant costs of work experience were computed by applying the program's work experience allocation of total costs for the cost analysis year to the estimate of average cost per participant. Thus, these estimates should not be interpreted as the average cost of work experience per participant involved in such activities. OE = occupational exploration; OJT = on-the-job training; TANF = Temporary Assistance for Needy Families.

Among the study site programs, Philadelphia-TWC had the highest cost per participant ($6,641), for all the reasons mentioned above. Most TWC clients participated in the program's work experience component, which was relatively long. Because the program explicitly targeted individuals with limited or no work experience, all TWC clients who completed the program's initial job readiness activities were placed in transitional work activities.⁽⁶⁾ TWC participants were expected to spend 25 hours each week in transitional work and were paid $5.15 per hour. They remained in the positions until they were considered ready for work, up to a maximum of six months. Another distinctive feature of the TWC program was that further job readiness and skills upgrading were integrated into the program's work experience component. While in transitional work, TWC participants were required to attend 10 hours of career development training each week.

High participation rates in work experience and high wage costs also appeared to contribute to the high per-participant costs of the Chicago-Pyramid program ($5,826). In contrast to the employer-tailored programs in Boston, which integrated unpaid job shadowing into participants' job readiness activities, Pyramid featured a four- to six-week paid work experience placement for all its participants. Moreover, while most WtW work experience placements in other programs paid minimum wage, Pyramid placed its WtW participants in on-the-job training positions, which generally paid higher wages.

Conversely, the cost per participant for the West Virginia-HRD program was lower ($3,771), mainly due to modest wage costs. Although all participants were expected to receive some work experience, they did not always receive direct wages from the WtW program. Most participants "worked off their TANF grants" and received only a small stipend ($1.60 per hour) to supplement their cash assistance.

Among WtW programs that offered work experience, costs were lower where such placements were based on need. For example, the Yakima programs (with average costs ranging from $3,530 to $4,912) used paid work experience only for participants who had completed 12 weeks of structured job search without securing unsubsidized employment.⁽⁷⁾ The initial sequence of job readiness and job search activities at Chicago-Catholic Charities and Chicago-Easter Seals helped WtW staff identify relatively "work ready" WtW participants, contributing to lower per-participant costs ($3,310 and $3,087, respectively).⁽⁸⁾

Paid work experience was a minor component of average costs in the Nashville-Pathways program ($1,964). While Pathways offered paid work experience for up to 12 weeks, few participants were placed in such positions. Only 69 of the 870 individuals ever active in Pathways during the cost analysis year (about eight percent) participated in work experience during that year.

b. Differences in Preemployment Costs Reflected the Structure of Job Readiness and the Use of Work Experience

The costs of WtW job readiness and preemployment case management varied widely, largely because of differences in the nature of job readiness activities and the duration of case management. Average costs ranged from $695 (Nashville-Pathways) to $2,548 (Boston-Partners) per participant (Table III.4).

Reflecting an overall commitment to building "human capital," the most costly WtW programs placed participants in structured job readiness components (lasting up to seven weeks) and lengthy work experience with associated case management (that could last up to nine months). Boston-Partners, with the highest costs for job readiness and preemployment case management, put participants in seven weeks of classroom-based work preparation activities, and Partners case managers tracked participants closely until they were placed in unsubsidized jobs. Similarly, the Yakima WtW programs had relatively high average costs for job readiness and preemployment case management ($1,822 to $1,420). Although these programs did not include a lengthy routine job readiness component, many participants were placed in work experience for up to nine months and received individualized work readiness and job search support during that period.

Costs were lowest where participants pursued more independent job readiness activities and individualized objectives, since WtW staff did not always have to be intensely involved in such activities. In Nashville, for example, the main activity for all Pathways participants was a monthly meeting, in which they shared their accomplishments and challenges with their case managers and each other. Job readiness costs averaged $695 per participant.

c. Variation in Postemployment Support Costs Reflects the Labor Intensity of Services and Participant Payments

Helping WtW participants not just secure employment but also retain and, if possible, move on to better jobs was an important goal for all of the programs we examined. In most WtW programs, postplacement support consisted of brief staff contacts with participants (and sometimes their employers), and costs were therefore modest. In 11 programs, these contacts typically lasted through the first six months of employment, with decreasing frequency and intensity, at a cost ranging from $241 (Fort Worth-ANS) to $419 (Chicago-Pyramid). The remaining seven programs provided more intensive postemployment services, wage supplements, or retention incentives to participants, and their costs for postplacement services ranged from $473 (Philadelphia-TWC) to $1,520 (Phoenix-EARN) per participant.

Postemployment costs were relatively high in the two JHU programs--in Florida and Maryland--which provided their participants with intensive postemployment support. These Career Transcript System programs worked mostly with newly employed individuals (and their supervisors), and their objective was to help participants retain their jobs, identify advancement opportunities, and move up a career ladder. By design, then, WtW spending focused on postemployment case management. At costs of $1,309 per participant in JHU-Maryland and $1,198 in JHU-Florida, postemployment services represented 60 and 55 percent, respectively, of overall average costs for these programs.

Postemployment services were most costly, however, in Phoenix-EARN, which complemented follow-up case management with structured mentoring. All EARN participants placed in unsubsidized jobs received regular follow-up visits by professional counselors for six months after placement. These visits were in addition to brief follow-up contacts by EARN case managers (similar to those made by most other WtW programs). This additional attention brought postemployment costs to $1,520. The Yakima-OIC program also featured postplacement mentoring services, which contributed to higher-than-average costs for postemployment services ($504 per participant).

Wage supplements and other direct payments to WtW participants could also raise postemployment costs. WtW participants in West Virginia-HRD who worked 30 or more hours per week and earned less than $7.75 per hour received such supplements for their first 24 weeks of employment.⁽⁹⁾ Philadelphia-TWC participants were eligible for up to $800 in retention bonuses: $400 after completing their first month of unsubsidized employment, $200 after three months of continuous employment, and another $200 after six months of continuous employment.⁽¹⁰⁾ In these programs, postemployment costs per participant were $870 (West Virginia-HRD) and $473 (Philadelphia-TWC).

[Go To Contents]

D. Variation in Costs per Placement

The goal of WtW programs was to place participants in unsubsidized employment and, thus, help them make strides toward self-sufficiency. The cost of achieving this objective can be summarized as the cost per placement--the resources programs had to invest, on average, to have one participant reach unsubsidized job placement. Our estimates of cost per placement for WtW programs cover a wide range--from $3,501 for Boston-Marriott to $13,778 for Philadelphia-TWC.

These estimates should not be interpreted as measures of programs' efficiency or effectiveness, for several reasons. Differences in cost per placement partly reflect differences in the mix and intensity of services that WtW programs offered. For programs that offered similar services, differences in cost per placement may reflect important differences in the populations served and, therefore, the relative ease or difficulty with which programs could achieve placements. Differences in cost per placement also do not take into account potentially important differences in the quality or long-term success of programs' placements. Thus, differences in costs per placement probably bear no relation to differences in outcomes or impacts across programs.

For any individual WtW program, cost per placement is principally a function of its cost per participant and its placement rate. Different factors can influence a program's cost per participant, placement rate, or both--and, through them, its cost per placement. In this section, we examine how program characteristics, contextual factors, and participant characteristics relate to the variation in costs per placement observed across WtW programs.⁽¹¹⁾ In general, we found that job-matching strategies and participant characteristics were the most important factors in understanding differences in WtW costs per placement.

1. Cost per Placement Clearly Reflects Job-Matching Strategy

While all the WtW programs provided placement support to their participants (Nightingale 2002), they still varied in how their staff worked with participants searching for employment. Program staff could be more or less active in identifying job opportunities appropriate for their clients and in letting them know about these opportunities. Similarly, staff could work more or less actively with participants to help them identify and pursue job opportunities that may be appropriate for, or appealing to, them as individuals.

Program efforts to help WtW participants secure unsubsidized employment seemed to follow three general approaches:

1. Self-Directed Matching. Under this approach, participants generally identified their own matches to available job listings, with some guidance and support from WtW program staff. While participants may have attended job clubs or even received some job leads from WtW staff, they were principally responsible for identifying appropriate opportunities and pursuing them. Nashville-Pathways, West Virginia-HRD, and Chicago-Catholic Charities used self-directed matching.
2. Staff-Assisted Matching. Programs that used this approach had dedicated staff who worked actively to identify job openings suitable for their WtW clientele and to match participants individually to them. In several programs with staff-assisted placement--Chicago-Maximus, Chicago-E&ES, Philadelphia-TWC, Phoenix-EARN, and the Yakima programs-WtW staff also had well-established placement relationships with some local employers. Since the programs continuously made placements with this set of employers, ensuring good matches was considered important to keep these employer clients satisfied.
3. Employer-Focused Matching. Under this approach, program staff worked exclusively with selected employers and industries that had job opportunities available and helped WtW participants gain the skills and qualifications needed for such positions. Examples of this approach are the three employer-tailored WtW programs: Boston-Marriott, Boston-Partners, and Chicago-Partners. In these programs, job placement was guaranteed to participants who completed the employer- or industry-specific training.

Overall job-matching strategy helps explain differences in placement rates and costs per placement (Table III.5). On average, programs that used employer-focused or staff-assisted matching approaches achieved higher placement rates. This outcome is to be expected, as these programs took a more active role in facilitating placement and ensuring good job matches for the generally hard-to-place WtW population.

These programs also had, on average, higher overall costs per participant. While their more active approach to placement required additional resources for job development and placement activities, differences in the costs of these activities did not fully account for their higher overall per-participant costs. Thus, the higher average costs for these programs suggest that they worked more intensively with WtW participants in other areas besides placement.

Other characteristics that one may associate with WtW programs that worked more intensively with participants were not as important in understanding cross-program differences in placement rates and, therefore, costs per placement. For example, TE programs, which worked with WtW participants to gradually and systematically upgrade their employability skills, achieved the same placement rates, on average, as EDE programs, which tried to place WtW participants in unsubsidized employment as soon as possible (Table III.6). On average, the costs per placement for TE programs were higher, due mainly to the programs' higher average costs per participant.

The TE programs and others that offered paid work experience also had high average costs per placement (Table III.7). However, presence of a paid work experience component was not synonymous with marked differences in WtW placement rates. Rather, programs with a work experience component had higher overall costs per participant (as discussed previously), and this was the principal factor contributing to their higher costs per placement.

Differences in placement rates and, therefore, the costs per placement of WtW programs could also reflect differences in how selective WtW programs were in the types of jobs into which they placed participants. For example, some programs may have only placed participants in jobs offering a minimum wage rate or number of hours, benefits, or strong advancement opportunities. However, the starting wages appeared to be similar for the jobs that WtW programs helped their participants secure. In general, WtW programs placed participants in jobs with starting wages that ranged from about $6 to $7 per hour. The only notable exception was employer-focused programs, whose participants, on average, secured jobs with higher starting hourly wages (Table III.5). As we discuss below, these programs may have targeted WtW-eligible people who were relatively more job ready, which could have contributed to the higher wage rates for their initial unsubsidized placements.

2. Urbanicity and Prevailing Unemployment Rates Had No Obvious Effect on Costs per Placement

The local economy can influence many aspects of a program, such as the choice of service strategies, operating expenses, and its ability to place participants in employment. However, differences in urbanicity and unemployment rates for the areas the WtW programs served were not strongly associated with differences in costs per placement, placement rates, or costs per participant across WtW programs (Table III.8).

The cost analysis periods examined were generally periods of relative prosperity, strong economic growth, and strong labor markets. Most of the WtW programs in the cost analysis operated in urban areas that, in 2000, had low unemployment, averaging 3.4 percent.⁽¹²⁾ As expected, higher average unemployment rates for the WtW programs operating in rural areas suggested that these programs faced somewhat less prosperous economic conditions. However, these did not appear to lead to higher costs per placement for these programs. Despite serving a rural area with relatively high unemployment (10.6 percent), the Yakima programs achieved relatively high placement rates (50 to 73 percent) and thus, in general, more modest costs per placement ($4,829 to $8,762).

Table III.8
Wtw Costs per Placement by Urbanicity of Service Area

WtW Programs, by Urbanicity of Service Delivery Area	Cost per Placement (in Dollars)	Placement Rate (Percent)	Cost per Participant (in Dollars)	Unemployment Rate (Percent) (a)	Average Starting Hourly Wage (in Dollars)
Rural
West Virginia-HRD	6,182	61	3,771	8.0 (b)	5.84
Yakima-OIC	8,762	51	4,433	10.6	7.40
Yakima-FWC	8,065	61	4,912	10.6	7.07
Yakima-PFP	4,829	73	3,530	10.6	7.75
Averages	6,960	62	4,162	9.9	7.02
Urban
Fort Worth-ANS	7,725	33	2,365	3.2	NA
Chicago-Catholic Charities	8,339	40	3,310	4.2	7.00
Philadelphia-TWC	13,778	48	6,641	4.0	7.26
Chicago-Easter Seals	5,758	54	3,086	4.2	6.64
Boston-Partners	8,192	66	5,407	2.2	8.98
Boston-Marriott	3,251	71	2,308	2.2	9.67
Chicago-Pyramid	8,037	73	5,827	4.2	7.43
Fort Worth-WC	5,241	42	1,887	3.2	NA
Nashville-Pathways	3,685	53	1,964	2.8	7.19
Chicago-E&ES	5,453	62	3,392	4.2	7.05
Phoenix-EARN	6,301	66	4,133	2.7	7.46
Chicago-Maximus	4,622	78	3,605	4.2	6.98
Averages	6,699	57	3,661	3.4	7.57
Note:NA = not available. a. Average unemployment rate in 2000, by Metropolitan Statistical Area. b. Average across 29 service counties, weighted by proportion of WtW participants from each county.

3. Participant Characteristics May Have Influenced WtW Costs per Placement

To the extent that WtW programs worked with relatively more or less hard-to-serve individuals, their costs per placement may vary.⁽¹³⁾ WtW programs that targeted WtW-eligible participants with the most severe barriers to employment, on average, had lower placement rates than programs that targeted the general WtW-eligible population or WtW-eligible participants who met at least a minimum threshold for employability (Table III.9).

Table III.9
Wtw Costs per Placement by Target Population

WtW Programs, by Target Population	Cost per Placement (in Dollars)	Placement Rate (Percent)	Cost per Participant (in Dollars)	Average Starting Hourly Wage (in Dollars)
Hardest-to-Serve Among WtW Eligible
Fort Worth-ANS	7,725	33	2,365	NA
Chicago-Catholic Charities	8,339	40	3,310	7.00
Philadelphia-TWC	13,778	48	6,641	7.26
Chicago-Easter Seals	5,758	54	3,086	6.64
Averages	8,900	44	3,851	6.97
Relatively Job-Ready Among WtW Eligible
Boston-Partners	8,192	66	5,407	8.98
Boston-Marriott	3,251	71	2,308	9.67
Chicago-Pyramid	8,037	73	5,827	7.43
Averages	6,493	70	4,514	8.69
General WtW Eligible
Fort Worth-WC	5,241	42	1,887	NA
West Virginia-HRD	6,182	61	3,771	5.84
Nashville-Pathways	3,685	53	1,964	7.19
Yakima-OIC	8,762	51	4,433	7.40
Yakima-FWC	8,065	61	4,912	7.07
Chicago-E&ES	5,453	62	3,392	7.05
Phoenix-EARN	6,301	66	4,133	7.46
Yakima-PFP	4,829	73	3,530	7.75
Chicago-Maximus	4,622	78	3,605	6.98
Averages	5,904	61	3,514	7.09
Note:  NA = not available.

Some WtW programs--most notably Philadelphia-TWC and Fort Worth-ANS--explicitly targeted WtW-eligible clients with little or no work experience, low literacy or numeric skills, and other substantial barriers to employment (such as homelessness). These programs aimed to serve some of the most disadvantaged WtW-eligible people, and differences in their placement rates are consistent with such targeting strategies.

Conversely, those WtW programs that targeted the relatively job-ready among WtW-eligible people achieved higher placement rates. For example, the Boston employer partnership programs screened applicants for interest and skill. Participants had to apply for the programs and meet minimum literacy requirements. Thus, because of the nature of these programs, their participants may have been somewhat less disadvantaged and more motivated than other WtW-eligible people.

Differences in placement rates or costs per placement should not be equated with program success or effectiveness, however.Previous research has shown that achieving one placement of a harder-to-serve participant may create greater impacts than one placement of a relatively job-ready participant.⁽¹⁴⁾ Thus, despite the lower average placement rates for programs targeting the hardest-to serve WtW-eligible participants, the greatest benefits of WtW may lie in achieving success with this population.

[Go To Contents]

Endnotes

1.  Factors that affected enrollment are discussed in detail in other evaluation reports: Fender et al. 2001; Nightingale et al. 2002; Nightingale 2001; Perez-Johnson et al. 2000; and Perez-Johnson and Hershey 1999. Congress modified the WtW legislation in 1999 as part of the Fiscal Year 2000 Appropriations legislation for the Departments of Labor, Health and Human Services, Education, and related agencies, to expand WtW eligibility.

2.  In Table II.2, both JHU programs appear not to offer job readiness and pre-employment case management services, as their allocation of total costs to these activities is shown as zero. Most JHU program services were similar to those that other WtW programs offered. The main distinction of the JHU programs was that the staff generally (but not always) began working with WtW participants after they secured unsubsidized employment. Thus, most program costs were allocated to postemployment followup.

3.  An important exception was WtW participants who are noncustodial parents of children on TANF. Since these individuals usually are not TANF recipients themselves, they do not qualify for TANF-funded transitional benefits. The potential cost implications of serving more noncustodial parents can be seen somewhat in the differences in WtW support service costs across the Yakima WtW programs (Table II.2). Support services accounted for eight percent of total costs for Yakima-PFP, which targeted noncustodial parents. In contrast, support services accounted for half that amount in the two other Yakima programs, which did not serve, or served far fewer, noncustodial parents.

4.  For example, Nashville-Pathways could provide bus passes or gas reimbursement for 3 months out of a 12-month period for those in job readiness or work experience, among various other supportive services or payments.

5.  See Nightingale et al. 2002.

6.  Seventy-six percent of those individuals ever active during TWC's cost analysis year participated in transitional work during this same period.

7.  Other participants received retention services only, but they still figure in the calculation of average costs for the programs.

8.  The per-participant costs for these two WtW programs also could have been lower for other reasons. As in Philadelphia-TWC, the Chicago-Easter Seals work experience component integrated job readiness and case management activities. It was much shorter, however (four weeks maximum instead of six months maximum). Reportedly, the Chicago-Catholic Charities staff did not interact as often with WtW participants in work experience, who were only required to attend a weekly job club while in paid work experience activities.

9.  During the cost analysis year, the West Virginia-HRD program placed 228 participants in unsubsidized jobs and spent about $110,000 on wage supplements, for an average wage supplement cost of $480 per participant placed in unsubsidized employment. Supplements were reduced over three eight-week intervals--bringing the wage up to $7.75 per hour for the first eight weeks, $6.80 for the second eight weeks, and $5.80 for the third eight weeks.

10.  During its cost analysis year, TWC paid $258,200 in retention bonuses to 498 participants. Program records indicated that 69 percent of the TWC participants who had been placed in unsubsidized jobs by the end of the cost analysis year had received a first bonus, 48 percent had received a second bonus, and 32 percent had received a third bonus.

11.  The JHU programs (Florida and Maryland) are excluded from this discussion because of their postemployment focus. JHU staff helped WtW participants secure unsubsidized jobs if they lost their original positions or were ready to advance to better positions. Initial placement in unsubsidized employment was not a primary program objective, however.

12.  Based on the Bureau of Labor Statistics' average yearly rates for 2000 in each program's Metropolitan Statistical Area (www.bls.gov/lau/laumatch.htm).

13.  Participants across the WtW evaluation sites were similar in prevalence of recognized barriers to employment, such as low levels of education and physical or mental disabilities (Nightingale 2002). In any given evaluation site, however, individual WtW programs could have reached out to relatively more or less disadvantaged participants.

14.  See, for example, Scrivener et al. 1998; Maxfield 1990; and O'Neill 1990.

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)