The Parents’ Fair Share (PFS) Demonstration:
Matching Opportunities to Obligations

Chapter 7:
The Employment and Training Component

[ Main Page of Report | Contents of Report ]

Contents

1. Defining the Employment and Training Service Approach
   1. The Role of Employment and Training in PFS
   2. Promising Employment and Training Service Approaches
   3. Reconciling Competing Priorities
   4. The Implementation Challenge
2. Employment and Training in the Sites
   1. Services, Staffing, and Organizational Approaches
   2. Early Participation Patterns
   3. Explanations for the Participation Patterns
3. The Evolution of the Employment and Training Component

   A. MDRC's Efforts to Promote Change
   

This chapter examines the employment and training services that have been provided to PFS participants during the demonstration's pilot phase.  The chapter begins by describing the role of employment and training in PFS, the challenges MDRC faced in designing guidelines for this component, and the approach that was eventually adopted.  The second section describes how the employment and training component was implemented in the pilot sites and examines the patterns of participation in employment and training activities.  The third section describes how the employment and training component has evolved during the pilot phase as sites have worked with MDRC to identify operational problems and develop strategies to strengthen the component.; The fourth section presents three case studies of particularly innovative employment and training strategies that were implemented in the pilot sites and the final section offers some conclusions.

MDRC did not prescribe a specific employment and training service model for the PFS pilot phase.  Instead, the sites were encouraged to offer a wide range of employment and training options to serve a diverse population and were required to place a special emphasis on on-the-job training (OJT), which was seen as a vehicle for providing both training and income-producing work.  Overall, the pilot phase experience has confirmed the feasibility of exposing noncustodial parents to employment and training activities and tracking their participation.  Moreover, most of the pilot sites have developed at least some strong employment and training services, and many innovative approaches have been tested.

At the same time, there has been considerable variation in the quality, intensity, and nature of the employment and training services participants have received.  A variety of institutional issues and conscious program design choices made it difficult for many of the sites to deliver a broad menu of employment and training options.  Data on the activities of noncustodial parents who entered PFS by the end of February 1993 show that more than half of those who participated in employment and training activities received only relatively short-term job search or job readiness services.  Although more than 40 percent of the employment and training participants attended some type of education or training activity, this mostly consisted of adult basic education classes; the pace of OJT placements was slower than expected and relatively few participants entered classroom skills training.  There was concern that this narrow mix of services was appropriate for only a subset of the PFS population.

MDRC worked closely with sites to identify the reasons for these early patterns of participation.  Several kinds of technical assistance were provided and most sites made substantial changes in their employment and training strategies during the pilot.  These changes did not begin to yield dividends until after the period covered by the data available for this report, but they should allow the sites to offer a richer and more intensive mix of employment and training services in the future.

I. Defining the Employment and Training Service Approach

PFS is not designed to test a particular type of employment and training service such as a job club or a skills training course.  Rather, the demonstration is examining the feasibility and, later, the impact of delivering an integrated package of employment and training and other services to a large, diverse group of noncustodial parents in the context of an enhanced child support enforcement system.  Given these broad objectives, it seems clear that no single employment and training service would be appropriate for all sites and participants.

Despite the lack of a single fixed model, it was important for MDRC to provide some guidance to sites before the project began about the overall objectives of the employment component and the types of employment and training services that seemed most likely to be effective for the PFS population.  This was a difficult task, both because existing knowledge about the PFS target group was limited and because the history of employment and training programs for men has not yielded many positive results.  Nevertheless, in an effort to develop guidelines for the states applying for admission to PFS, MDRC examined past research on the likely characteristics of the PFS target population, visited existing PFS-like programs that had emerged in several jurisdictions to learn more about the men served and their needs, reviewed the results of past employment programs for men, and consulted with experts on employment and training.   This section describes the results of this effort and the challenges that faced the pilot sites as they designed employment and training strategies for PFS.

A.The Role of Employment and Training in PFS

Although all four components are closely linked to the project's diverse goals, employment and training services lie at the core of PFS.  The PFS approach is based on the hypothesis, grounded in research findings, that higher incomes among noncustodial parents are associated with greater and more regular child support payments.^[1]  Peer support and mediation are intended to reenforce the importance of parental responsibility and provide a forum for discussing and resolving issues related to support, and enhanced child support enforcement aims to tighten the link between income and child support.  But the employment component is the key to making greater support a reality for many of the parents in PFS.  The baseline data described in Chapter 3 appear to confirm that many noncustodial parents lack the income to pay child support. 

At the same time, PFS employment and training services must reflect the fact that the normal patterns of work and child support for the program's target population are probably both dynamic and diverse.  As in any employment and training program, some PFS participants would begin to work and pay child support even if the program did not exist.  Others would fare quite poorly without the program.  Thus, in order to make a difference, PFS must help participants find better jobs than they could obtain on their own, or to get jobs faster and/or keep them longer.  If PFS succeeds in this goal, the short-term investment (of foregone earnings and support payments plus program resources) will be more than repaid by the later increase in earnings and support above the levels that otherwise would have occurred.  The random assignment evaluation planned for Phase II of the demonstration will assess whether this happens.

B. Promising Employment and Training Service Approaches

Ultimately, MDRC's background research yielded two broad and, to some extent, contradictory conclusions about the types of employment and training services that were likely to be needed in PFS:

• An array of employment and training services -- including opportunities for participants to build their skills and earning capacity -- would be needed to serve a diverse group of noncustodial parents.
• Both institutional realities and the likely circumstances of noncustodial parents suggested that intense time pressure would play a major role in service delivery for this group; many noncustodial parents would want jobs and income quickly. Each of these is discussed in more detail below.

1. The Need for Skill-Building Services.  Most of the earlier programs linking CSE with employment and training (described in Chapter 1) relied on relatively short-term job search assistance activities.  This emphasis was not surprising given the limited funding that was typically available for these efforts and the fact that they emerged from a child support enforcement system that focuses on maximizing short-term support collections.  Put simply, the programs were designed to get noncustodial parents into the workforce and paying child support as quickly as possible.

Although no systematic studies were undertaken to assess the effectiveness of these programs, MDRC concluded that a broader employment and training strategy was needed for PFS.  This conclusion did not imply that job search activities were inappropriate or ineffective.  On the contrary, studies have shown that programs relying primarily on job search can generate modest earnings and employment impacts, particularly for women on welfare.^[2]  Moreover, other research has shown that minority men -- who were likely to account for a large proportion of the PFS population -- often have difficulty in the labor market because they do not have access to the same kinds of informal networks that many white men use to find jobs.^[3]  Programs that teach participants formal job search techniques and help them identify and connect with job openings may help minority men gain access to jobs they could not otherwise find.

However, while recognizing the value of job search assistance, MDRC concluded that these activities would only be sufficient for a subset of the PFS population.  Studies have shown that the jobs people obtain through programs stressing job search usually pay about the same, on average, as the jobs people obtain on their own.  When these programs produce earnings impacts, they usually do so by increasing the proportion of people who work at all, or by helping people find jobs more quickly than they otherwise would.  The limited available data suggested that the PFS population was likely to be diverse, including men with a variety of skill and education levels and work histories.  Although many would face daunting barriers to employment, most would have had at least some work experience.  Some fraction would have worked regularly in the past, but many would have exhibited patterns of unstable or sporadic employment in low-wage or "off the books" jobs.  Job search assistance seemed unlikely to move many of the men in this last group into better, more stable jobs that could help them break the cycle of intermittent employment and earn enough to both support themselves and meet their child support obligations.  This suggested that PFS should include, in addition to job search assistance, opportunities for participants to enter skill-building education and training activities that might lead to better jobs.

Two recent studies have shown that broad coverage, mandatory programs using this kind of mixed approach have produced some positive results for disadvantaged men.  The first, the Saturation Work Initiative Model (SWIM), operated in San Diego County from 1985 to 1987.  Serving both male and female AFDC recipients, SWIM placed participants in a fixed sequence of activities beginning with a 2-week job search workshop.  Those who failed to find employment during this period were usually placed in a three-month unpaid work experience position.  Those who were still unemployed after completing work experience could be referred to education and training programs in the community.  As might be expected given the sequence of activities, SWIM participants were more likely to receive job search services than education and training.  Nevertheless, the program's emphasis on skill-building activities was much stronger than in most previous welfare-to-work programs.  SWIM produced statistically significant increases in the employment rates of mostly-male recipients of AFDC-U^[4] during a five year follow-up period.  The program also increased the earnings of AFDC-U clients during the first year of follow-up, although these impacts evaporated in later years.^[5]

More recently, California's statewide Greater Avenues for Independence (GAIN) welfare-to-work program has tested a model relying even more heavily on education and training.  Under GAIN rules, individuals who do not have a high school diploma or who fail a basic skills test are required to enroll in education activities unless they opt to try job search first.  In six counties participating in MDRC's evaluation of GAIN, from 49 to 91 percent of AFDC-U clients who started any GAIN activity participated in education or training during the first 11 months after attending a program orientation.^[6] During the first two years of follow-up, GAIN produced statistically significant impacts on the employment rates of AFDC-U clients in 4 of the 6 counties and impacts on earnings in 3 counties.^[7]

Although these results are encouraging, it is important to note that the SWIM and GAIN target populations differed from the PFS population in several respects.  First, AFDC-U clients are by definition public assistance recipients and they generally live with their families.^[8]  Second, individuals usually must have a recent connection to the labor force in order to be eligible for AFDC-U.^[9]  Third, California's AFDC-U caseload includes many Asian immigrants who are not present in large numbers in other areas of the country.

2. The Need for Immediate Income.   Unlike the GAIN and SWIM participants, it was assumed that most of the noncustodial parents in PFS would not be public assistance recipients and, therefore, would be in need of immediate income.^[10]  Long classroom training or education programs, although possibly useful in building some participants' earning power, would delay their ability to earn money in the short-term.  Thus, participants might not be willing to make the initial investment required to achieve longer-term impacts, or would need a way to combine training and work to cover their living expenses.  (Neither JOBS nor JTPA could provide income support for these parents while they participated in program services.)  In addition, it was assumed that many of the participants would have had negative school experiences and would resist making a long-term commitment to classroom-based skills training or basic education programs. 

The institutional imperative of the child support enforcement system reenforced the sense of urgency in service delivery.  Although child support enforcement agencies participating in the demonstration consciously agreed to make some short-term investment in services and foregone support in hopes of longer-term payoffs, many had not agreed to an approach that could delay support payments for long periods.

C. Reconciling Competing Priorities

MDRC sought to balance these two objectives in developing specific guidelines for the states applying for admission to PFS.  The resulting guidelines for the employment and training component emphasized two main points:

• The need for a menu of service options.  It was assumed that without countervailing pressure the "default" in most PFS sites would be to emphasize short-term job search activities for the reasons cited above.  Thus, states applying for admission to the demonstration were strongly encouraged to develop a range of employment and training options for PFS participants including both job search assistance and education and training opportunities.  The guidelines urged program planners to think about service strategies for three broad groups of participants:  1) those with fairly consistent work histories, 2) those with histories of unstable employment in low-wage positions, and 3) those with little or no work history.
• A special focus on on-the-job training (OJT).  As part of the effort to ensure that skill-building activities would be available, MDRC required each site to commit to provide at least 100 on-the-job training slots for PFS participants.  OJT was seen as an ideal way to wed the need for training with the myriad pressures for immediate income (and child support payments).  Under OJT arrangements, individuals are placed in private sector jobs and public funds are used to provide their employers with a temporary (usually 3 to 6 month) wage subsidy for agreeing to hire and train a disadvantaged employee.  A recent national evaluation of JTPA found positive impacts over 18 months of follow-up for men who volunteered for the program and were deemed appropriate for on-the-job training (OJT).^[11]

Beyond these general guidelines, states were free to design the employment component to reflect their own objectives and views of the target population.  For example, while MDRC stressed the importance of having clearly-defined criteria for matching participants with specific types of services, sites were not required to adopt a particular approach to this vital task. Thus, the pilot phase presented an opportunity for sites to experiment with a range of employment and training strategies.

D. The Implementation Challenge

Developing the kind of employment and training component envisioned by MDRC presented sites with a set of difficult operational choices and institutional challenges.  Three examples follow.

1. Matching Participants with Services.  With a diverse target population and a broad menu of employment and training service options, strategies for matching participants with appropriate services assume great importance.  Previous research (focusing mainly on welfare-to-work programs) has highlighted the difficulty of this task.^[12] 

Employment and training programs use a variety of methods to match participants with services.  These include:

• Up-front Assessment. Some programs begin with a formal assessment of the participants' skills, education, work history, interests, and barriers to employment.  This process, which may involve a lengthy series of tests and interviews, results in an individualized service plan for each participant.  This approach is typically used in the JTPA system.
• Initial Appraisal.  Many large-scale, mandatory welfare-to-work initiatives make initial activity assignments on the basis of a few easily-measured factors, such as educational attainment or recent work history.  Information on these factors can be obtained during a relatively brief interview, sometimes called an appraisal.^[13]
• Fixed Sequence for all Participants.  Other mandatory programs assign all participants to a fixed sequence of services, typically beginning with job search.  This identifies participants who can find jobs on their own, and helps the others clarify their interests and goals.  Those who fail to find employment within a few weeks may be assigned to education or training activities, often after a formal assessment.  This was essentially the model used in the SWIM program.

Each of these methods has advantages and disadvantages, and each reflects the overall goals and philosophy of the program in which it is used.  For example, up-front assessment, which is typically used in programs stressing education and training, may provide a wealth of information about participants.  However, assessments tends to be quite expensive and research suggests that assessors' ability to accurately predict which participants will prosper in particular services is limited.   On the other hand, an up-front job search strategy, which may reflect a program goal of helping many participants enter the labor market quickly, is a relatively simple and inexpensive way to "sort" large numbers of participants.  However, critics note that it is difficult in practice to strike a balance between promoting immediate employment and identifying participants with training needs.  Thus, they suggest that these programs often encourage people to accept low-wage jobs that are unlikely to last.  The pilot sites needed to consider these issues in designing a service assignment strategy for PFS.

2. Institutional Barriers. Before PFS began, no existing employment and training service system saw poor, noncustodial parents as a priority target group or had fashioned a package of services designed to meet their needs.  Thus, PFS was almost certain to require new patterns of organizational cooperation.  Moreover, the focus on skill-building activities sought by MDRC was not necessarily consistent with "business as usual" in the relevant sectors of the employment and training system.

It was clear from the outset that both the JOBS and JTPA systems would be important partners in the PFS employment component.  The statutory roots of the demonstration assured that JOBS funding would support PFS employment and training activities.  JTPA agencies, which often play a role in local JOBS programs, are experienced providers of education and training (including OJT) for the disadvantaged. However, as discussed in Chapter 1, PFS asked both of these systems to make important changes in their standard operating procedures.

Most local JOBS programs have had limited experience serving men; AFDC-U recipients usually account for a small fraction of the AFDC caseload.  Moreover, on-the-job training, designed to be a core PFS employment and training service, is rarely used in JOBS programs.^[14]  Thus, PFS presented these agencies with both an unfamiliar target group and a new service approach.  Nevertheless, change was feasible because, as described in Chapter 2, JOBS is in fact as much a funding mechanism as a service organization.  In many jurisdictions, JOBS agencies contract for key services, especially those involving more intensive education and training. The challenge was to encourage JOBS agencies to rethink their institutional strategies to ensure that the organizations best suited to meeting the needs of mostly-male noncustodial parents were involved. 

Local JTPA agencies were natural partners in the service effort, especially given the PFS focus on skill-building activities and OJT.^[15]  However, the JTPA system has been drawn in conflicting directions in recent years.  On one hand, JTPA is the nation's major training program for the economically disadvantaged and, as such, claims to provide individualized, client-driven services that are more intensive than those typically offered in JOBS or other programs in which participation is mandatory.  On the other hand, the JTPA system is also subject to the criticism that it has -- at least until recently -- generally provided low-cost, short-term services to relatively job-ready clients.   This allegation arose because, in its initial years, JTPA had in place performance standards for local agencies that rewarded high job placement rates and placement wages and low costs per "success story."  Critics alleged that these standards discouraged many local JTPA programs from serving clients who were less likely to be placed in good jobs or providing more intensive, skill-building services.^[16]

As PFS was being developed, the JTPA system was under increasing pressure to target its services on those with greater barriers to employment and to provide more long-term services. Thus, the demonstration emerged at a time when the JTPA system, at least in theory, was seeking new ways to reach out to harder-to-serve clients.  Moreover, as in JOBS, local JTPA programs contract with an array of other local service providers so the system had the potential to be quite flexible in responding to previously unmet needs.  On the other hand, unlike JOBS, JTPA has a long history of outcome-driven performance standards that discourage risk.^[17]

3. Decreased Use of OJT. Despite the positive evaluation results described earlier, there were two serious obstacles to the heavy focus on on-the-job training in PFS.  First, while most JTPA Service Delivery Areas (SDAs) have extensive experience with OJT, this service is often perceived to be appropriate for participants who are judged to be "job ready" overall but lack a specific occupational skill. JTPA staff contend that employers are rarely willing to take risks on participants facing more formidable barriers.  It was not clear what proportion of the PFS population would fit this description, although the application guidelines implicitly recognized that many would not: sites were required to develop 100 OJT slots and to serve 300 participants. 

A second, more serious, problem concerned recent developments within the JTPA system that had pushed many SDAs to de-emphasize or even eliminate OJT as a service option.  Investigations by the Department of Labor's Inspector General in the 1980s had revealed that OJT was often used to subsidize employers in situations where little documented training occurred.  Thus, tighter performance audits and new federal regulations required JTPA agencies to carefully document the training that would be provided during the subsidy period, and to ensure that the length of the subsidized training reflected the characteristics of the job for which the employee was being trained.  Although these regulations were intended to prevent abuse rather than to reduce the number of OJT contracts written, many local JTPA agencies claimed that the new, tighter rules made the OJT "package" more difficult to sell to employers by potentially increasing the paperwork burden and decreasing the size of the subsidies they were able to pay.

4. Technical Assistance. In order to help sites develop and implement the employment component and address vital choices and obstacles like those described above, MDRC provided two technical assistance manuals during the planning period.  The first was a general guide discussing alternative strategies for determining what type of employment services are appropriate for particular participants.  The second paper described a "high support" version of OJT designed for participants who were somewhat more disadvantaged than the typical OJT candidate.  During a kick-off conference for representatives of all pilot sites, 2 sessions were devoted to OJT design and marketing strategies.  Additional technical assistance was provided during the course of the pilot phase, as discussed below.

[Go To Contents]

II. Employment and Training in the Sites

As might be expected, the sites responded to MDRC's general guidelines with a variety of employment and training strategies.  This section describes the nature of the employment and training services offered by sites, the staffing and structure of the component, and the patterns of participation in these activities, focusing mostly on the early part of the pilot phase. The next section discusses how the component has evolved and changed during the pilot.

A. Services, Staffing, and Organizational Approaches

Diversity in program approaches is a key theme of this report.  Starting with different institutional resources and linkages and distinct visions of the PFS, each site developed its own version of the program.  This variation was perhaps most evident in the employment component, which offers the greatest degree of site-level discretion.

1. Service Options. Following the application guidelines, each of the PFS sites developed a range of employment and training options for participants.  Although the content and schedules of these activities varied widely across sites, the overall "menu" typically included the following general categories of services:

• Group or Individual Job Search.  These activities varied considerably from site to site, but generally followed one of two patterns: 1) group job clubs that begin with 1-2 weeks of classroom instruction in job-seeking and job-holding skills followed by several weeks of structured job search, often using a phone bank; or 2) supervised individual job search in which participants work with a staff person to seek employment.^[18]  Most sites employ job developers who are responsible for identifying job openings for PFS participants.
• Job Readiness or Pre-Employment Training.^[19]  This typically involves 2-4 weeks of classroom instruction in "world of work" skills, self-esteem building, and motivational activities, but does not usually include job-seeking skills or direct job search.
• Classroom Skills Training.  This includes training courses for specific occupational areas, such as truck-driving, office skills, or electronics.
• Classroom Remedial Education. Also called Adult Basic Education (ABE), this usually includes basic reading and math instruction, often including preparation for the General Educational Development (GED) exam.  Some programs offer individualized, computer-based instruction, while others operate more like traditional high school classes.  Post-secondary education, though not heavily used in PFS, is also included in this category.
• On-the-job Training (OJT). As noted earlier, each site committed to develop at least 100 OJT slots for PFS participants.

Although almost all of the sites developed the capacity to deliver each of these types of services to at least some participants, a variety of factors led each site to emphasize only 1 or 2 of them in practice.  Thus, as will be discussed below, the menu of services actually available to most PFS participants has been more limited than the one described above.

2. Organizational Structure.  Although most of the PFS lead agencies were employment and training providers, each site needed to assemble a diverse network of agencies to provide the menu of services described above.   Table 7.1 illustrates that these have included JOBS and JTPA agencies, a variety of nonprofit organizations, school districts, vocational schools, and community colleges.^[20]  The organizational and financial linkages connecting these providers with the PFS lead agency are typically complex, and vary widely from site to site.

TABLE 7.1

In several sites, these networks grew directly from earlier linkages forged in the JOBS program or some other employment and training initiative.  For example, in structuring PFS Ohio relied primarily on providers that were already contracted to serve JOBS participants.  Similarly, the key employment and training agencies in Michigan had previously provided services for both the JOBS program and a previous initiative for noncustodial parents.  In other sites, such as New Jersey, the JOBS program served primarily as a funding conduit; the key PFS service providers had little or no previous association with the program although they did have experience relevant to PFS (e.g., they had served male populations).

In most sites 2 or 3 different agencies, including the lead agency, played major roles in providing employment and training services during the early months of the pilot phase.  The complexity of the provider network and the identity of the key members reflected the program's overall emphasis. In sites like Tennessee that stressed a wide range of services, including skill-building activities, a variety of different agencies tended to play key roles.  Sites with a more narrow focus -- including Minnesota, New Jersey, and Butler County, Ohio -- involved fewer providers.  As will be discussed below, the shape of the provider networks sometimes changed during the pilot as the overall nature of the employment and training component evolved.

Under the most common organizational model the PFS lead agency directly operated the employment activity that most participants attended first -- usually a job search or job readiness workshop -- and relied on outside providers for most education and training activities.  School districts or community colleges were typically responsible for basic education services; JTPA agencies were often slated to provide classroom and on-the-job training initially although, as described below, several sites have altered this arrangement during the pilot phase.^[21]

Although the financial and legal linkages connecting the employment and training providers were often complex, several of the sites developed innovative organizational approaches to integrate these services.  For example, in New Jersey, the State Department of Labor has outstationed a staff person in the PFS office to develop OJT positions for program participants. To the participants, this staff person is simply one member of the PFS team; her agency affiliation is probably not apparent.  As noted earlier, several sites have stationed staff from a number of different employment and training providers in a central PFS program office.

3. Staffing.   The staffing of the employment and training component depends largely on the structure of the service network and the funding arrangements that link the key providers.  In most sites, case managers employed by the lead agency are responsible for assigning participants to specific employment and training activities, monitoring their participation, arranging support services, and other related functions.

The staff who actually lead the employment and training activities may be the case managers themselves, other employees of the lead agency, or staff associated with other agencies. As noted earlier, job search and job readiness activities are often led by case managers or other lead agency staff.  In contrast, basic education and classroom training, typically provided by outside agencies, are almost always led by teachers associated with the school or agency that provides these services. 

In most sites, at least one staff person is assigned to develop job openings for PFS participants. In some sites, this individual develops both direct unsubsidized jobs (including, in some cases, part-time positions for participants who are in training or education) and OJT positions; in other sites different staff are responsible for each type of placement.  Once again, job and OJT developers may be associated with either the lead agency (they may be case managers) or a contracted provider (typically a JTPA agency).

B. Early Participation Patterns

Chapter 4 noted that about half of the noncustodial parents who were referred to PFS participated in an employment and training activity within 4 months of referral.  This section examines the patterns of participation among those noncustodial parents who attended at least one session of a PFS employment and training activity.  As in Chapter 3, this analysis uses program participation data reported for all noncustodial parents who were referred to PFS through February, 1993, and covers the first four months after the referral for each noncustodial parent.  Thus, these data primarily refer to the early part of the pilot phase.

1. Types of Services.  Table 7.2 examines the rates of participation in the four major categories of employment and training activities: job search/job readiness activities, basic education, classroom skills training, and on-the-job training.  These data show that it was difficult for most of the sites to deliver the broad menu of employment and training activities described in the application guidelines.  Overall, about three-fourths of PFS employment and training participants attended only one kind of employment and training activity during the four months following the referral.  Job search/job readiness activities were used most frequently; more than 80 percent of employment and training participants attended these activities, and more than half attended only these activities.  The next most common activity was basic education; almost a third of the participants attended these classes.  About one in six participants received classroom or on-the-job training.  Overall, just over 40 percent of participants attended some type of education or training activity.

TABLE 7.2

These combined figures mask considerable variation across sites.  The vast majority of participants in both Tennessee and Missouri attended basic education classes; few participants in these sites attended only job search/job readiness activities. Alabama, Florida, and Minnesota also had relatively high rates of basic education participation.  In the remaining sites, from 74 to 95 percent of participants attended only job search/job readiness activities.  The proportion of participants who were placed in OJT positions ranged from 0 in Butler County, Ohio to 17 percent in New Jersey, and the proportion attending skills training varied from a low of 1.4 percent in Butler County to a high of 25 percent in Minnesota.

2. Intensity of Participation.  Table 7.3 shows the total number of employment and training sessions attended per employment and training participant within 4 months of referral.  As noted earlier, the definition of a "session" varies from a full day of work in an OJT position to a few hours in a job club meeting.  Larger numbers of sessions do not necessarily signal a stronger program because many participants leave PFS to take jobs.  Nevertheless, this is a useful measure of the intensity of the employment and training experience. 

TABLE 7.3

The overall average number of sessions attended per participant is about 21.^[22]  Nearly 60 percent of employment and training participants attended at least 10 activity sessions; more than 25 percent attended 30 or more sessions.  Once again, however, there is considerable variation across sites.  The average number of sessions attended varies from more than 33 in Tennessee (where more than half of all participants attended 30 or more sessions) to about 8 in Florida (where nearly three-fourths of participants attended fewer than 7 sessions). The bottom row of the table suggests that the averages in sites like Tennessee, Massachusetts, Minnesota, and Missouri -- and, consequently, the overall average -- would be considerably higher if more months of follow-up data were available; a substantial fraction of participants in these sites were still attending employment and training activities in the final month of follow-up.^[23]

These site-level session attendance figures are clearly tied to the activity-specific participation rates in Table 7.2.  In general, basic education, skills training, and on-the-job training all involve substantially longer periods of participation than job search/job readiness activities.  This is illustrated in Table 7.4, which shows the average number of sessions attended among noncustodial parents who participated in each of the major types of employment and training activities and the fraction of these participants who were still active in the specified activity in the final month of follow-up.  The last column of the table shows that the average number of job club sessions attended among those who attended this activity was about 12, compared with averages of about 20 sessions for basic education attenders, 27 sessions for skills training participants, and 30 sessions (i.e. days of work) for those placed in OJT positions.  The proportion of participants still active in the final month of follow-up follows the same pattern, ranging from 27 percent of job club attenders to 65 percent of OJT attenders.  These results are not surprising because job search/job readiness activities are usually relatively brief and often lead to employment, while education and training activities are, by their nature, likely to involve longer time commitments.^[24]  Given these figures, it is not surprising that the average number of sessions attended is much higher in Tennessee, where most participants attended basic education, than in Ohio, where most attended only job search/job readiness activities.

TABLE 7.4

Although the association between activity-specific participation patterns and the overall intensity of the employment component (measured in terms of sessions attended) is clearly strong, the county-level data in Table 7.4 suggest that other factors are also at work.  First, these data suggest that the definitions and expectations for specific types of activities vary to some extent across sites.  For example, the site with the second highest average number of sessions, Massachusetts, relied heavily on job search/job readiness activities during the period covered by these data.  One-third of job search/job readiness participants in this site attended 30 or more sessions.  This would have been impossible in some other sites where these activities only lasted 3 to 4 weeks.

Second, there is strong evidence that some sites had more success than others in retaining participants in education and training activities.  For example, while nearly half of the basic education participants in Tennessee attended 30 or more sessions (and 54 percent were still active in the final month of follow-up), participants in sites like Alabama, Florida, and Missouri were much more likely to attend basic education for fewer than 20 sessions.

3. Short-Term Program-Reported Employment Rates.  Table 7.5 shows the percentage of noncustodial parents referred to PFS who reported employment to program staff within 4 months of the referral.  These figures do not provide a full picture of employment because they only measure job entries that are reported to PFS program staff.^[25] Staff are likely to know about jobs participants obtained directly through program activities but are less likely to hear about noncustodial parents who are referred to PFS and obtain jobs on their own without participating in the program (or participants who drop out of the program and subsequently find employment).  Although these "unassisted" placements may be partly attributable to PFS -- the noncustodial parent may have found employment to avoid the program -- they would often not be recorded in the MIS.^[26]

TABLE 7.5

The overall rate of program-reported employment is about 22 percent; the rate in individual sites ranges from about 8 percent in New Jersey to 41 percent in Dakota County, Minnesota. The overall average starting wage is just over $6 per hour; about one-fourth of the jobs paid $7 per hour or more and another one-fourth paid less than $5 per hour.  The average starting wage in each site ranges from less than $5 per hour in Alabama to $7.33 per hour in New Jersey.  It seems clear that the employment rates in certain sites would be higher if additional months of follow-up data were available; Table 7.3 showed that more than 40 percent of participants were still active in PFS employment and training activities in the final month of follow-up.  Indeed, the employment rate within 6 months of referral for an early cohort referred to PFS through December 1992 is nearly 28 percent. Although it is certainly possible that some sites were simply more proficient than others in helping participants find jobs, there are several other possible explanations for the wide variation from site to site in employment and starting wage rates.  These include:

• Data Issues.  The available data are not equally complete in all sites.  Specifically, it is very likely that the figures are understated in the sites where substantial numbers of nonparticipants did not enter the MIS, notably Montgomery County, Ohio and Tennessee.^[27]  Thus, in actuality, the cross-site differences may be smaller than is suggested by Table 7.5.
• Economic Conditions.  Table 1.2 noted that unemployment rates ranged from a high of 9.0 percent in the Springfield, Massachusetts to a low of 4.9 percent in the Kansas City, Missouri area during the study period.  It may have been more difficult for participants to find jobs in areas with depressed labor markets.  Similarly, prevailing wage rates, minimum wage laws, and the cost of living all vary in different regions of the country, which may help to explain the disparity in starting wage rates.
• Participant Characteristics.  The job readiness of the participants in different sites may have varied in measurable or unmeasurable ways.  A site working with a population facing more barriers to employment may have recorded a lower short-term employment rate.  For example, the New Jersey caseload was younger than average and had lower-than-average school completion rates.  On the other hand, a site like Minnesota, where a substantial proportion of participants were already employed at the point of the referral, probably had a "head start" on the other sites.

Finally, it seems clear that the employment rates are tied to some extent to the activity-specific participation rates in Table 7.2.  Although the pattern is far from perfect, the employment rate tends to be above average in sites like Alabama, Florida, and Butler County, Ohio where virtually all participants engaged in job search activities and lower than average in sites like Tennessee and Missouri, which assigned most participants to basic education.^[28]  This is logical since job search/job readiness activities are typically designed to lead directly to employment while basic education is not.  Additional follow-up would be needed to assess the payoff of longer-term services.

C. Explanations for the Participation Patterns

Overall, the data presented above confirm the feasibility of moving noncustodial parents into employment and training activities and monitoring their attendance.  However, the nature and intensity of the employment and training services participants received during the early months of the pilot varied considerably from site to site.  In several sites, virtually all employment and training participants received only relatively short-term job search/job readiness services.  Participants in most of these sites did not have a very intensive employment and training experience, although many did find jobs.  A few sites were more successful in exposing participants to skill-building activities (typically basic education) that tended to be more intensive. However, the ability of these sites to retain participants in these activities varied.  None of the sites was able to move as many participants as planned into OJT positions, and relatively few participants entered classroom skills training.  Several operational issues help to account for these patterns; three are discussed below.

1. Activity Assignment Policies.  The procedures used to determine participants' first employment and training activity are probably the single most important factor in explaining the early participation patterns.  No two sites used exactly the same process to make this determination; however, two general patterns emerged during the demonstration's early months:

• Up-front Job Search.  One group of sites assigned the vast majority of PFS participants to a fixed sequence beginning with a job readiness or job search activity.  For example, in Massachusetts, virtually all PFS participants began with a 3-week pre-employment training class.  After completing the class, participants either continued with supervised job search or were referred for an OJT placement or for other training or education.  Similarly, virtually all participants in New Jersey, Michigan, Florida, and Butler County, Ohio were referred initially to a group job club. Participants who completed this activity without finding employment could, in theory, be referred for education, training, or further job search, although the procedures for this transition were not always clearly articulated.
• Initial Assessment or Appraisal.  In the other sites, the first activity assignment depended on the results of an initial assessment or appraisal of the participant's characteristics, aptitudes and/or interests.  For example, in Missouri, participants were assessed during orientation and a three-week peer support program.  Those who lacked a high school diploma or could not read at the 10th grade level were generally assigned to Adult Basic Education classes; upon completion they were referred to the JTPA agency for OJT placement.  Those who did not begin with education were referred for OJT placement immediately after peer support.  Alabama, Minnesota, Tennessee, and Montgomery County, Ohio also used an initial assessment or appraisal to determine the first assignment, although the criteria in these sites were defined less sharply than in Missouri and could theoretically lead in any one of several directions.

The design and implementation of these procedures reflected the overall program philosophy and the previous experience of the lead agency in each site.  The sites that opted for an up-front job search strategy apparently felt that it was important to let the labor market identify those participants who could benefit from education or training; many of these sites had previously operated JOBS programs or other initiatives with a strong focus on job search activities.

The assignment procedures in the sites that used initial assessments or appraisals were, in principle, more flexible.  However, the day-to-day implementation of these procedures was similarly guided by the overall program emphasis and the previous experience of the staff involved.  Thus, in Tennessee and Missouri -- both sites that entered PFS with a strong focus on basic education -- the assessment almost always resulted in a referral to this activity.  In contrast, participants in Montgomery County, Ohio almost always began with job club; this was consistent with the usual pattern in the county's JOBS program.

The first activity assignment is critical because, as noted earlier, most employment and training participants attended only one type of activity.  This was particularly true of those who began with job search/job readiness activities; these individuals often found employment or left the program before moving to a second activity.^[29]  This result probably reflects the pressure on participants to find work quickly^[30], the inability of most sites to clearly define the next step for individuals who failed to find jobs during the initial activity, and the limited availability of other kinds of services (see next section).  It also underscores the difficulty of balancing a focus on immediate employment with an effort to identify participants with training needs.

2. Institutional Issues.As expected the organizational characterstics of both JOBS and JTPA agencies shaped the employment and training component in the PFS sites.  The early pilot experience illustrates the difficulty inherent in any attempt to change long-established institutional behavior patterns.

PFS reflects a broader debate between "labor force attachment" and "human capital development" approaches to the JOBS program.  In some sites, including Florida and Ohio, JOBS programs that had previously stressed job search activities continued this emphasis in PFS.   This shaped both the service network and the activity assignment procedures and ensured that most participants in these sites would receive job search services.  In contrast, the Tennessee and Missouri JOBS programs, which had previously stressed basic education, transferred this emphasis into PFS.  Sites like Massachusetts and New Jersey were less influenced by the previous nature of the local JOBS program because the PFS programs were run by nonprofit agencies that were only indirectly linked to JOBS.

As expected, agencies funded through the JTPA system were slated to be important providers of skill-building services in most sites.  This was particularly true of OJT, a service with which most JTPA agencies have extensive experience.  In nearly every site the local JTPA entity initially agreed to serve a specific number of PFS participants, either under contract using JOBS funds or directly using JTPA funds.

In fact, the level of JTPA involvement in most sites was lower than anticipated.  In some cases, JTPA agencies maintained that many PFS participants could not be certified eligible for JTPA because they were living with relatives who refused to provide household income information.^[31]  In other cases, JTPA staff frankly stated that the PFS population was "too risky" to serve in expensive activities given their performance standards and ongoing relationships with employers (other SDAs reported that the PFS population was fairly similar to their regular service population). 

The most immediate consequence of the lack of JTPA involvement was to limit participants' access to OJT and skills training.  As noted earlier, prior to the start-up of program operations, each site agreed to place at least 100 PFS participants into OJT positions during the pilot phase; JTPA agencies were usually made responsible for meeting this goal.  MDRC expected that the level of OJT activity would be low at first, but would increase within a few months after program start-up. However, by the end of January 1993, only 55 people across all nine sites had been placed in OJT positions.  One site, Missouri, had 12 OJT placements; no other site had more than 8.  Although the pace eventually picked up (see below) no site is likely to reach its initial goal.  The level of participation in classroom skills training was also fairly low, although this was less surprising given the participants' desires for immediate income.

A variety of explanations have been offered for the relative lack of success in this area.  Some JTPA agencies pointed to the regulations governing OJTs funded under JTPA (described earlier), contending that these requirements created onerous burdens and reduced the allowable length of many OJTs to the point that employers no longer found them appealing.^[32]  Anticipating this problem, several sites used JOBS funds to support OJTs developed by JTPA staff, since the rules governing this activity under JOBS are much less stringent.  However, OJT placement rates were relatively low even in these sites.

Other JTPA staff suggested that a relatively small fraction of PFS participants qualified for skills training or an OJT placement and questioned whether the emphasis on these services was appropriate for this population.  For example, they contended that many PFS participants could not meet the entry requirements for skills training classes (which may require a diploma or GED).  Similarly, some maintained that OJT is suitable for job-ready participants who lack a specific work skill, and said that most PFS participants faced more formidable barriers to employment such as very limited work experience, poor basic skills, and criminal records.  These agencies generally did not attempt to develop the "high support" OJT model described earlier, nor did they design alternatives for less job-ready participants.

Finally, in some sites, the activity assignment procedures described above effectively reduced the number of participants who were considered for OJT placement.  Some sites failed to develop clear criteria to identify participants who were suitable for OJT positions.  Others, concerned about "wasting" OJT slots on participants who could easily obtain good jobs on their own, reserved the OJT option for participants who failed to find unsubsidized employment during an up-front job search activity.  As noted earlier, these sites tended to discover that few participants remained active in the program after the job search component; most had either found employment or dropped out of the program.  This is not to suggest that employment is a negative outcome, but there was evidence that OJT placements tended to offer higher wages and better training opportunities than unsubsidized jobs obtained through job search programs.  As shown in Table 7.5, the average starting wage for participants placed in unsubsidized employment within 4 months of referral was almost identical to the average wage participants received in their most recent job before being referred to PFS (see Table 3.1).

3. Participants' Preferences.  Anecdotal evidence suggests that, as expected, many PFS participants have expressed a strong preference for activities that lead directly to employment, and have resisted classroom training or education.  In fact, staff in some sites have reported that some participants erroneously assumed that the program would provide them with a job and showed up to orientation asking when they would start work.  These participants were reluctant to wait several weeks or months for the next cycle of a training course to begin and, in some cases, were even unwilling to wait to meet with a staff member to discuss an OJT placement.  In a few sites, PFS case managers and other staff worked with participants during up-front peer support activities and succeeded in persuading many of them to consider other options.  In other sites, case managers, perhaps more familiar with programs that stressed job search, made only limited attempts to persuade participants to give skill-building activities a chance.

[Go To Contents]

III. The Evolution of the Employment and Training Component

The early patterns of participation in employment and training activities raised some important questions.  The primary concern was that most sites appeared to be having difficulty delivering the broad menu of options described in the application guidelines.  With relatively few noncustodial parents moving into classroom or on-the-job training, most participants were only receiving relatively short-term job search/job readiness activities; many were not remaining active in PFS for long periods. 

On the one hand, this may seem like a positive result, since many of those with short stays in PFS found employment.  On the other hand, there was concern that, without the benefit of skill-building activities, these individuals were not finding the kinds of jobs that would allow them to support themselves and meet their child support obligations.  This concern, which was supported by the sites own reports about the serious barriers facing many PFS participants, suggested that job retention would be a problem.  Moreover, field observation suggested that the job search/job readiness activities offered in several sites were not particularly intensive or interesting to participants; in some sites, participants seemed to drift into long periods of individual job search with little supervision.

The sites that assigned most participants to educational activities were focused on building human capital.  However, in these sites, there was a concern that participants were either becoming "stuck" in education programs for long periods without making progress or, at the opposite extreme, dropping out of these activities because they lacked income or commitment.  This problem was exacerbated by the fact that most sites did not have clear or realistic criteria to determine when participants had completed basic education.

This suggested that both sites stressing education and those emphasizing job search needed to expand and diversity their approaches.

A. MDRC's Efforts to Promote Change

Mid-way through the pilot phase, senior MDRC staff visited each site to conduct a Program Review designed to identify operational problems and devise solutions.  Following these meetings, sites were asked to developed specific plans to address the issues identified during the meetings.  Regarding the employment and training component, MDRC emphasized that the following goals should guide this effort:

• Broadening the Menu of Employment and Training Options.  All sites were encouraged to restructure their employment components to steer more participants toward skill-building activities, including OJT and activities mixing classroom training and/or education with income-producing work. Some sites clearly needed to find additional service providers to accomplish this task.  Sites stressing basic education were encouraged to review their criteria for completion of these activities and to consider whether participants were really committed to this course. 
• Strengthening Job Clubs.  MDRC hired, as a consultant to the project, a firm with extensive experience developing high quality job clubs.  Representatives of this firm visited several of the PFS sites to train job club providers. In additon, sites that were relying heavily on job search/job readiness activities were asked to make certain

[Go To Contents]

Endnotes

1. Although several studies have found that child support payments are associated with noncustodial parent income (see, e.g., Sonenstein and Calhoun, 1990), it is has not been shown that a program that increases the income of noncustodial parents will necessary lead to higher child support payments. [Back To Text]

2. Gueron and Pauly, 1991. [Back To Text]

3. Holzer, Sullivan. [Back To Text]

4. The AFDC-Unemployed Parent program provides assistance to two-parent families in which one parent is unemployed or incapacitated. [Back To Text]

5. Friedlander and Hamilton, 1993. [Back To Text]

6. Most of these individuals attended adult basic education, GED preparation, or English-as-a-Second- Language classes rather than skills training programs. [Back To Text]

7. Friedlander, Riccio, and Freedman, 1993. [Back To Text]

8. The SWIM and GAIN populations probably overlap with the PFS population in this respect:  Some PFS participants live with at least some of their children and may receive AFDC-U, and some of the SWIM and GAIN participants may have been both custodial and noncustodial parents. [Back To Text]

9. In the GAIN research counties that served the full population of AFDC-U clients, from 72 to 80 percent of these recipients reported being employed within the past 2 years. [Back To Text]

10. Some PFS participants may be recipients of General Assistance or AFDC-U, and others may be supported, in part, by an AFDC grant received a relative or partner. [Back To Text]

11. Cite JTPA report. [Back To Text]

12. Auspos and Sherwood, 1992. [Back To Text]

13. For example, in Florida's Project Independence (JOBS program) individuals who have a high school diploma or who have worked in at least 12 of the previous 24 months are determined to be "job ready" and are assigned to a sequence of job search activities.  Clients who are determined not to be job-ready (and job-ready individuals who fail to find employment in job search activities) are referred to a more complete assessment which may result in a referral to education or training.  See Kemple and Haimson, forthcoming. [Back To Text]

14. Lurie and Hagen, 1993. [Back To Text]

15. The role of JTPA in the PFS employment and training component was not specified in the application guidelines, although sites were encouraged to work closely with local JTPA agencies.  [Back To Text]

16. General Accounting Office, 1989; 1990. [Back To Text]

17. The current JOBS performance standards focus more on process than outcomes -- states are required to engage specific percentages of their AFDC caseloads in JOBS services.  PFS usually accounts for a small proportion of the total JOBS caseload in a jurisdiction and thus does not seriously affect the participation rate calculations. [Back To Text]

18 . These individualized efforts may be underreported in MDRC's participation data system. [Back To Text]

19. In practice, the distinction between the first two categories of activities has become blurred in many of the pilot sites.  For example, in reporting on participants' attendance in program activities, some sites have used the term "job club" to refer to an activity that closely resembles job readiness training in other sites.  In other cases, both job readiness and job search activities have been provided through the same agency during a single 2-4 week session, and it is not possible to separate them.  Thus, these two categories of activities are combined in most of the tables and figures that follow. [Back To Text]

20. This section and Table 7.1 focus on the agencies providing services, rather than their funding sources.  Most PFS services are supported to some extent with either JOBS or JTPA funds.  Thus, in a sense, the organizations whose staff lead these activities could be referred to as JOBS or JTPA-funded agencies (especially since some of them obtain funding almost exclusively through one of these sources).  However, this section distinguishes between situations where public agencies that administer the JOBS and JTPA programs in a jurisdiction provide services directly using their own staff, and those where JOBS or JTPA funds are used to pay for services provided by private nonprofit organizations, school districts, colleges, etc. [Back To Text]

21. In some cases, outside providers are under contract and receive PFS funds to serve a specified number of individuals.  In other situations, outside agencies have agreed to serve PFS participants using a separate funding stream such as JTPA or state education funding.  Expenditures for these services are part of the cost of PFS, although they may not be charged to the formal project budget. [Back To Text]

22. The average number of hours per session across all activity types is probably between 3 and 5; this translates into a total of 72 to 120 hours of participation over the first 4 months. [Back To Text]

23. As noted earlier, the figures for sessions attended do not necessarily capture all employment-related activities.  For example, in some sites, case managers met with participants individually for counseling sessions that often included job search help.  These meetings may not have been reported as formal activities. [Back To Text]

24 . To provide some context for these figures, a typical job club workshop may include 10 to 20 sessions over 2 to 4 weeks; basic education and skills training classes may meet anywhere from 3 to 5 times per week for several months; and OJT positions usually involve five days of work per week for 3 to 4 months.  However, in interpreting the figures in Table 7.4, it is important to note that individuals often do not start education and training activities or OJT positions immediately upon entering the PFS program.  Thus, the session attendance figures may in fact be capturing attendance over a period that is much shorter than 4 months. [Back To Text]

25. During Phase II, MDRC plans to track employment through state wage reporting data. [Back To Text]

26. Some of the jobs that are not reported to PFS program staff may be detected at some point later by the child support enforcement system; however, it is far from certain that this information would be transferred back to PFS staff and reported to MDRC. [Back To Text]

27. These sites completed the Enrollment Form only for noncustodial parents who attended a PFS orientation.  Since the MIS that provides information on attendance and job entries covered only individuals who completed the Enrollment Form, many of those who failed to show up to orientation are not included in the system in these sites.  As noted above, some noncustodial parents who are referred to PFS find jobs on their own and never attend a program orientation. [Back To Text]

28. It is also important to note that these employment figures include only unsubsidized jobs; OJT placements are not included. Thus, sites like New Jersey, which placed relatively large numbers of participants into OJT slots would be expected to have lower rates of unsubsidized employment. [Back To Text]

29. Although PFS participation is mandatory, some participants stop attending program activities.  As explained in earlier chapters, the consequences for noncooperation varied considerably. [Back To Text]

30. As described in Chapter 6, the treatment of participants' child support orders during the period of PFS participation varied considerably from site to site.  Some sites completely suspended the obligation, while others made no routine changes. [Back To Text]

31. Even older individuals who are living with their parents may in some cases be required to count the full household income in determining JTPA eligibility.  As noted in Chapter 2, few PFS participants live alone.  A similar result was observed in the Young Unwed Fathers Demonstration operated by Public/Private Ventures. [Back To Text]

32. Critics of the OJT approach have long maintained that the offer of a subsidy sends a signal to the the employer that the potential employee is sub-par and reduces his/her chances of being hired. [Back To Text]

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)

Last updated: 04/26/01