[ Main Page of Report | Contents of Report ]
This chapter summarizes MDRC's conclusions based on the PFS pilot phase experience, highlights some of the implications of these results for the PFS service approach, and discusses a number of broader policy issues raised by PFS.
MDRC's decision to recommend a full-scale test of the impacts of PFS is based on three broad conclusions, reached after more than a year of close observation and the collection and analysis of considerable data on the operational performance of the pilot sites.
First, the pilot sites have been able to recruit diverse institutional partners, build programs around the four core PFS components, and ensure that participants receive a consistent message about the goals of PFS. Although smooth linkages took time to develop and organizational approaches varied substantially, the sites developed numerous strategies to reconcile divergent organizational cultures and ensure that participants do not receive mixed messages about the program's goals. As a result of these efforts, PFS is up and running in all sites, all four core components are in place, and the program has received consistent attention from staff in most of the key agencies and broad community and political support.
Second, most of the sites have developed effective procedures to identify eligible noncustodial parents, move them into program services, and encourage and enforce regular participation. Although the pace of referrals was slow at first most sites eventually developed routine systems that have generated more than 3,400 referrals to date. This required identifying a much larger number of nonpaying noncustodial parents because, as expected, many either failed to show up for scheduled hearings or appointments, or ackowledged employment and agreed to pay support without being referred to PFS.
About two-thirds of the noncustodial parents who were referred to PFS actually participated in an employment and training or peer support activity within four months of the referral. Considering that more than 90 percent of these parents were not engaging in any employment and training activities on their own at the time of the referral, this suggests a major increase in service receipt levels. Moreover, the PFS participation rate is higher than comparable rates measured by MDRC in mandatory employment programs for welfare recipients that also imposed participation requirements on large numbers of individuals who may or may not have been interested in pursuing these opportunities. Finally, most of the noncustodial parents who did not participate in PFS services either reported employment to program staff or were referred to CSE staff for further enforcement action. The number of parents who "slipped through the cracks" was relatively small.
Third, there are clear signs that PFS is changing both individuals and systems. The quantitative data show that a substantial share of those referred to PFS have been actively engaged in program activities. The most dramatic signs of change appear during the peer support sessions, which have emerged as the heart of many of the PFS programs. Many participants quickly develop a B attachment to this activity and both group members and staff insist that profound changes in both outlooks and behavior result. Participants frequently report rediscovering their B feelings for their children, which inspires them to work harder to obtain employment, pay child support regularly, and stay out of legal trouble.
Less dramatic but equally important changes have occurred in the key institutions involved in PFS. Child support enforcement agencies are "working" cases that were typically neglected before PFS provided a constructive enforcement option. In addition to generating large numbers of referrals to PFS, this new focus has uncovered a substantial amount of previously-unreported income. Most of the sites developed procedures to routinely reduce the child support obligations of noncustodial parents during the period they were participating in PFS, and to raise these orders quickly if a parent failed to cooperate or found employment. This represented a major change, since orders were rarely changed in the past. Finally, CSE agencies have developed new systems to expedite the implementation of wage withholding orders for participants who find employment and to follow-up quickly when noncustodial parents fail to participate in PFS as ordered. These individuals are returned to court and may be found in contempt. In most sites, these enhanced CSE efforts were facilitated by designating specific CSE staff to work with PFS cases.
Employment and training systems have been more difficult to change, perhaps because these systems perceived that they were already prepared to serve the PFS population. A variety of institutional barriers and program design choices have made it difficult for sites to build a broad menu of services options for PFS participants. As a result, the nature, quality, and intensity of the employment and training services participants received varied dramatically from site to site. In general, too many participants received only short-term job search assistance, and too few received classroom or on-the-job training or education. On the positive side, several sites have identified B nonprofit employment and training providers with more experience serving individuals facing serious barriers to employment and the menu of employment and training options has become broader over time. Moreover, the pace of OJT placement accelerated greatly during the latter months of the pilot.
Together, these conclusions confirm that PFS is indeed a feasible and promising policy approach that deserves a rigorous impact test.
[Go To Contents]
The pilot phase has generally confirmed the appropriateness of the PFS service approach. Thus, MDRC will recommend that all four core components -- with some modifications -- continue to be included during Phase II. In addition, MDRC will make some effort to reduce the level of variation in site approaches in order to create a more standardized "model" that builds on the lessons of the pilot phase.
Qualitative and quantitative information on the target population illustrate its diversity and confirm that an array of employment and training services -- including skill-building activities -- is indeed necessary. The experience of the pilot sites demonstrates that such a package -- including a focus on on-the-job training (OJT) -- is difficult, but not impossible, to assemble. In addition, sites have been encouraged to supplement OJT with other activities mixing training and work, and to increase the level of post-placement follow-up to bolster job retention.
The picture of the noncustodial parents that emerged during the pilot also confirms the rationale behind the peer support component. It seems clear that a program based solely on employment and training would have difficulty engaging many noncustodial parents and, in many cases, would not lead to regular child support payments. Several of the pilot sites would have had serious problems if peer support had not succeeded in building the credibility of PFS. Sites are now Bly encouraged to place participants into peer support groups before they begin employment and training activities and to create opportunities for participants to remain involved for long periods.
Mediation has clearly been underutilized during the pilot phase. Noncustodial parents have been reluctant to request services and, when they have, custodial parents have often failed to cooperate. This is consistent with the results of other studies of voluntary mediation programs. Nevertheless, MDRC remains committed to providing opportunities for mediation and sites are experimenting with ways to promote greater usage.
The need for enhanced child support enforcement has been confirmed by the very fact that sites have needed to adopt special measures such as designated staff and special court dockets in order to identify, refer, monitor, and follow-up on PFS cases. It is clear that the system could not have supported PFS without these steps.
[Go To Contents]
Parents' Fair Share has captured the imagination of many observers. Despite the lack of information on program impacts and cost effectiveness, the project has garnered an unusual level of media attention and has already begun to influence the national policy debate. PFS is appealing because it simultaneously addresses a number of vexing social problems and exemplifies the mutual obligation approach to social policy that has received broad political support in recent years. Specifically, the project offers the hope of achieving two highly desirable but seemingly contradictory objectives: toughening the child support enforcement system and offering assistance to poor noncustodial parents.
Although PFS clearly offers exciting possibilities, it seems important to step back and carefully examine the full implications of the PFS approach and the kinds of impacts it might produce before rushing headlong into Phase II or broader replication efforts. This examination reveals a set of unusually complex and in some cases surprising policy questions. Some of these questions can be answered, or at least informed, through careful impact and process research while others raise more fundamental issues of fairness and equity that must be resolved through a broad debate about policy objectives and societal values.
This section describes some of the key questions raised by PFS. It begins by discussing questions that the research can address at least to some extent, and then moves on to describe a set of critical issues on which research can offer only very limited guidance.
Now that the pilot phase experience has shown that PFS is feasible and promising, the next key question is clearly: Does it make a difference? The random assignment evaluation planned for Phase II of the demonstration is designed to answer this question by assessing the impact of PFS on noncustodial parents' earnings, employment rates, and child support payments; custodial parents' welfare receipt and household income; and measures of child well-being.
In order to make this assessment, individuals who meet the eligibility criteria for PFS will be assigned, at random, to either a Program Group, which remains eligible for the program, or to a Control Group, which is ineligible for the program. The two groups will then be tracked over time and compared on a number of relevant outcomes. Because the groups are created through a random process, there are no systematic differences between them except for the fact that only the Program Group is exposed to the program. Thus, any differences in subsequent behavior between the two groups can confidently be attributed to PFS.
The impact analysis described above will almost certainly yield reliable estimates of the program's effects. However, the story is likely to be complex. If PFS does generate impacts, it will be critical to understand the mechanism that produced them. In the two examples below, different interpretations might lead to quite different policy conclusions.
1. Smokeout versus Services. As noted earlier, PFS may increase child support payments in two ways. First, the employment and training and peer support services may lead to increases in noncustodial parents' earnings which are then translated into greater and more regular child support payments. Second, PFS may "smoke out" previously unreported earnings by offering CSE agencies a vehicle for working cases that were previously neglected.
The current PFS service approach implicitly assumes that both kinds of effects are possible. However, if the history of employment and training programs for men is any indication, there is a real possibility that the service-driven effects will be modest. That is why these services are supplemented with peer support in PFS. However, if all or most of the program's effect is driven by smokeout, one might conclude that an intensive, expensive service package is in fact unnecessary. Some type of service referral might be required to give courts and CSE agencies an option for disposing of cases, but the primary objective of the services would be to disrupt noncustodial parents' off-the-books work schedules. An option like unpaid community service might be an attractive way to achieve this goal. Of course, one might also conclude that services could produce impacts but that the package tested in PFS was inappropriate; this would lead to a different policy recommendation. MDRC's early planning for Phase II has focused on developing a research design that can separate these two kinds of effects in at least some of the sites.
2. Formal versus Informal Earnings and Child Support. It seems clear that both custodial and noncustodial parents of children on welfare have B incentives to handle child support informally and that a large, albeit unmeasurable amount of underground work and child support now occurs. It also seems likely that PFS will produce a shift from informal to formal activity by moving noncustodial parents from underground into mainstream jobs (or uncovering unreported earnings) and deducting child support directly from their paychecks.
Although this is clearly a positive result from the government budget perspective, the implications for participants and their children are less clear; a shift from one type of earnings or support to another is clearly not the same as a net increase. In fact, as discussed in the next section, given the current system for handling support payments for children receiving AFDC, a shift from informal to formal support could decrease the resources available to children, at least in the short-term. While it is fairly straightforward to measure increases in formal activity through administrative records, it will require some creative data collection techniques to assess whether these represent net increases or shifts from one sector to the other.
One might assume that a program that generates an increase in formal child support payments would have an unambiguously positive impact on children. After all, children are clearly better off when they have access to greater financial resources. Moreover, studies have shown that child support -- as opposed to other income sources -- is associated with other positive outcomes such as childrens' educational attainment.(1)
However, despite this evidence there are several reasons to remain cautious. First, as long as the custodial parent remains on welfare, she and the children will not receive more than $50 per month in child support. Thus, if payments increase but are still not high enough to close the custodial parent's AFDC case, the primary beneficiary will be the government. In fact, to the extent that PFS results in a substitution of formal for informal child support, the custodial parent and children may experience a net loss in income. Of course, when custodial parents do go to work and leave welfare -- most leave the rolls within two years of applying -- more regular child support payments may make the difference between poverty and self-sufficiency and held reduce recidivism. Those who do so will begin to receive the full amount of child support that is collected.
Second, although payment of child support seems to be positively associated with several other measures of child well-being, this does not necessarily mean that increasing child support payments will improve the lives of children. Under the current system, noncustodial parents unfortunately have some discretion about whether to pay child support because enforcement has not been universal. In this environment, those who pay are likely to be more Bly committed to their children and better able to get along with custodial parents than those who do not. Put another way, noncustodial parents who do not care or cannot get along with custodial parents often do not pay support. This suggests that it may be the characteristics of the noncustodial parents who pay (or of their children), rather than the child support payments per se, that are causing the positive child outcomes.
If this is true, policies that succeed in forcing some of the current nonpayors to pay more may not have positive consequences for children. Indeed, one study suggests that forcing more noncustodial parents to pay child support may reduce child well-being by increasing conflict between parents. For unmarried parents, the authors predict that the costs of conflict will outweigh the benefits of greater financial resources.(2) PFS implicitly acknowledges this possibility through its inclusion of peer support and mediation, which are designed to address the potential negative side-effects that may accompany increased payment levels. In Phase II, it should be possible to obtain some information about the childrens' well-being through a survey of custodial parents.
The enhanced child support enforcement system that is taking shape in the PFS sites is different in many ways than the current system. CSE agencies are working cases they previously neglected and, in many sites, temporarily reducing child support obligations to ensure that noncustodial parents' support orders reflect their ability to pay. Moreover, the system is actively encouraging these parents to build their earning power rather than attempting to maximize short-term collections. These measures may seem both far-sighted and fair. However, as models for national policy they raise several difficult issues that have been part of the national debate over child support enforcement for some time.
1. Working All Cases. Working all AFDC child support cases -- including those with no sign of income -- sends an important message about parental responsibility and is clearly consistent with current federal rules that require agencies to take some action on all cases within prescribed time limits. But is this level of effort cost effective? These cases are likely to demand repeated location efforts and enforcement activities; and, even with the combined effects of smokeout and PFS services, they may yield fairly modest collections. Some would argue that it makes more sense to focus scarce enforcement resources on the large group of noncustodial parents who live and work in the mainstream economy and simply refuse to pay support.
2. Temporary Reductions. Temporarily reducing the support orders of PFS participants is also consistent with recent trends to the extent that it makes child support obligations reflect the noncustodial parent's ability to pay. This notion seems attractive in principle but presents practical problems when considered as a broader policy. For example, it is not clear over what time period "ability to pay" should be assessed. A noncustodial parent who has recently lost his job may have a limited ability to pay support now. But it seems both impractical and unwise to immediately reduce a support order every time a noncustodial parent becomes unemployed; most of these parents will be working again soon and the state must be careful not create a perverse incentive for them stay out of work. Moreover, their children's financial needs will not disappear in the interim. Thus, most state's child support guidelines and practices do not support such an approach; they examine income over an extended time period or impute child support orders for unemployed parents based on the minimum wage.
A B argument can be made that a different approach is needed to encourage PFS parents to build their earning power and cooperate more readily with the system. However, in the environment described above, implementing the PFS policy of routine reductions has required judges to deviate from their state's guidelines and usual practices for handling these cases. Careful monitoring has been needed to ensure that noncustodial parents' orders are quickly increased in response to noncompliance or employment. Ironically, this represents a return to the case-by-case approach that characterized child support enforcement in the past, and suggests that replication may be difficult. Moreover, if routine reductions are not suitable in all instances where noncustodial parents become unemployed, how should the system determine when they are appropriate? The PFS process evaluation may provide some lessons on "best practices" in this area.
Finally, a serious focus on ability to pay has broader implications. For example, it suggests that states need to review their child support guidelines; most of these formulas make no allowance for the subsistence needs of noncustodial parents and many take a disproportionate share of income from noncustodial parents working in low-wage jobs. Other state and local policies, such as charging noncustodial parents for "confinement" and welfare costs incurred by custodial parents before paternity was established are not tied to ability to pay.
3. Mutual Obligation. The very notion of adding opportunities to match the existing CSE obligation has far-reaching and complex implications. This approach seems vital to alleviate the negative side-effects of a tighter enforcement system. As it becomes more difficult to work in the mainstream economy and avoid paying child support, noncustodial parents who are willing to live and work outside the system will be even more likely to do so. The social cost of this result seems high, but is it appropriate to "reward" noncompliance with special benefits? What sort of message does this send to noncustodial parents who currently work and pay regularly? Moreover, in the final analysis, it is not clear whether it will be possible to draw noncustodial parents above ground and keep them there as long as most of the child support payments collected for children on welfare are retained by the state.
PFS means different things to different people. Some see the project as a rare opportunity to provide employment and training help to low-income minority men, a group that has suffered disproportionately from economic shifts over the past two decades. Others see it as a means to make the mandate to pay child support real by giving CSE agencies a way to enforce cases they previous neglected. Still others see PFS as an opportunity to expose what they see as the CSE system's anti-father bias and its fundamental unfairness to low-income noncustodial parents.
One on hand, the program's chameleon-like quality has helped it gain support from individuals and groups with sharply different perspectives. This is possible as long as each group believes that the program's benefits outweigh its costs, as they see them. On the other hand, it is not always possible to maintain this delicate balance and PFS has been subject to several types of criticism.
For example, some advocates for children and custodial parents contend that PFS weakens the obligation to pay child support. These groups have fought for years to persuade society and government to take child support more seriously, and some see PFS as an inappropriate attempt to make this obligation "conditional" on government assistance for noncustodial parents. Others accept PFS in theory, but argue that in a world of scarce resources, funds devoted to noncustodial parents will inevitably be diverted from custodial parents. Ideally, PFS would be part of a two-parent JOBS program, but few of the pilot sites have consciously attempted to make this connection.(3)
In a sharply different kind of critique, some have suggested that PFS is an extension of a child support enforcement and criminal justice system that is unfair to members of minority groups and low-income people in general. While possibly conceding that PFS is a step in the right direction, they maintain that the program basically accepts the system as is.
In the final analysis, PFS is bound to cause controversy. Any initiative that delves into the relationships between mothers, fathers and children; society's ideas about parental responsibility; the welfare system; and the court system is likely to confront and in some cases conflict with deeply-held beliefs. Rigorous research may shed important light on many of the key questions raised by such an initiative. But it is vital to recognize that other issues can only be settled through informed debate in the political arena. This does not argue against pressing ahead, but it does suggest that this step should be taken cautiously, without unrealistic expectations, and with a solid research design.
1. Cite Ginger Knox's dissertation. [Back To Text]
2. McLanahan et. al. [Back To Text]
3. An exception is Montgomery County, Ohio. In this site the custodial parents of PFS participants are assigned to a single JOBS case manager who works closely with the PFS staff. The intent is to tie the two programs together to create Ber incentives for both parents to find jobs. [Back To Text]
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