Working with Low-Income Cases: Lessons for the Child Support Enforcement System from Parents' Fair Share

Fred Doolittle

Suzanne Lynn

Manpower Demonstration Research Corporation

This report is based on research conducted for the Parents’ Fair Share Demonstration, a national demonstration project that combines job training and placement, peer support groups, and other services with the goal of increasing the earnings and child support payments of unemployed noncustodial parents (usually fathers) of children on welfare, improving their parenting and communication skills, and providing an opportunity for them to participate more fully and effectively in the lives of their children.

Funders of the Parents’ Fair Share Demonstration

U.S. Department of Health and Human Services  The Pew Charitable Trusts  W. K. Kellogg Foundation  Charles Stewart Mott Foundation  U.S. Department of Agriculture  The Annie E. Casey Foundation

U.S. Department of Labor  Smith Richardson Foundation  Ford Foundation  The McKnight Foundation  Northwest Area Foundation  State of California, Employment  Development Department

 

The findings and conclusions presented in this report do not necessarily represent the official positions or policies of the funders or the participating states. Interested readers may wish to contact the states for more information on the program. The sites and states in the Parents’ Fair Share Demonstration are Los Angeles Parents’ Fair Share Project, Los Angeles County (Los Angeles), California; Duval County Parents’ Fair Share Project, Duval County (Jacksonville), Florida; MassJOBS Parents’ Fair Share Project, Hampden County (Springfield), Massachusetts; Kent County Parents’ Fair Share Project, Kent County (Grand Rapids), Michigan; Operation Fatherhood, Mercer County (Trenton), New Jersey; Options for Parental Training and Support (OPTS), Montgomery County (Dayton), Ohio; and Tennessee Parents’ Fair Share Project, Shelby County (Memphis), Tennessee.

"

Preface

The child support enforcement system is facing new challenges, many of them driven by recognition of the increasing importance of such support as family structures have changed. Further, the existing level of child support payments is low: Less than one-fourth of those entitled to payment are receiving what is due them. The child support system is working hard to establish the paternity of more of the children who are receiving welfare (and were born outside of marriage), to put legally binding support orders in place, and to use many new enforcement techniques to raise the level of support payments. This reform effort comes at a time when changes in national welfare policy — most importantly, a time limit on the receipt of federally funded cash welfare — makes the issue even more critical to poor families.

Welfare reform and a desire to keep welfare expenditures down have been a motivating force for action. In reality, however, the agencies involved have been hard-pressed to find effective ways to deal with low-income, unemployed parents who are legally obligated to pay support. Traditionally, agencies have often decided that these parents would not produce enough to be worth a major enforcement effort, and the available enforcement tools were not effective with this group. Ironically, even though low payment rates in welfare-related cases were one of the main reasons the public supported stronger child support enforcement, these cases have been among the most difficult to address, and low-income, unemployed noncustodial parents have often been neglected in daily support administration. Courts and child support enforcement agencies have felt they must choose between one of two ineffective options — ordering noncustodial parents to seek work or jailing these parents for contempt of the court order to pay support. As a consequence, enforcement efforts have often largely focused on parents for whom there is evidence of current income.

The Parents’ Fair Share (PFS) Demonstration tests the feasibility and effectiveness of a new enforcement option for child support agencies in the seven participating counties. Under PFS, courts and agencies can refer parents who fall within program eligibility rules (i.e., noncustodial parents who are linked to a public assistance case, unemployed or underemployed, and not up-to-date with their support payments) to employment and training, peer support, and mediation services and can offer special flexibility in child support administration. Because of the availability of this new option (which was intended to increase noncustodial parents’ employment and earnings, child support payments, and involvement with their children), agencies and the courts moved aggressively to determine the status of cases they had previously not considered a high priority for action.

This report is about what happened when agency staff started to review cases, identify eligible parents, and refer those parents to the program. In the course of this effort, which produced more than 5,000 cases appropriate for PFS, local staff learned much about the status of child support cases: For example, some noncustodial parents were working and had not reported their employment; the lives of others had changed in ways that made their existing support order inappropriate; and still others were unlocatable. Furthermore, the effort produced lessons relevant to the broader child support caseload about how to conduct what might be thought of as "outreach" to the caseload of poor parents who are behind in their payments. In sum, this report provides an up-close look at what for many jurisdictions will be a new and important aspect of enforcement.

Two other reports from the PFS Demonstration will be published in the near future: a qualitative research report on the lives, attitudes, and experiences of a sample of PFS participants and a report on PFS’s implementation experiences and early impacts on employment and child support. Together these reports will add greatly to our knowledge of the lives of low-income noncustodial parents and the challenges the child support system faces as it seeks to increase payments for children receiving public assistance.

Acknowledgments

This report could not have been completed without the support and assistance of numerous people in the participating program sites, at the Parents’ Fair Share funders, and at MDRC.

Special thanks are due the child support officials in the seven Parents’ Fair Share sites who facilitated their sites’ participation in the project and reviewed a draft of this report: William Camden (Grand Rapids, Michigan), Annette Day (Jacksonville, Florida), Felicia Hogan (Memphis, Tennessee), Linda Jenkins (Los Angeles, California), Nancy Randolph (Dayton, Ohio), Mark Rogers (Trenton, New Jersey), and Norman Wagner (Springfield, Massachusetts).

Several of the project’s funders have also been actively involved in this part of the project from its inception. Especially active in this role have been Ronald Mincy of the Ford Foundation, Robert Harris of the Office of Child Support Enforcement at the U.S. Department of Health and Human Services (HHS), Linda Mellgren at the Office of the Assistant Secretary for Policy and Evaluation at HHS, and Mark Fucello at the Administration for Families and Children at HHS. Thanks are also due Vicki Turetsky of the Center for Law and Social Policy, who reviewed an early draft and suggested a number of ways to improve it.

At MDRC, Gordon Berlin, Judith Gueron, and Sharon Rowser reviewed drafts of this report and offered many helpful comments. Heidi Meyers did much of the background research on legal issues in child support enforcement, and Iris Reyes and Joan Johnson provided research assistance to the project. Sylvia Newman edited the report. Stephanie Cowell and Patt Pontevolpe did the word processing.

Finally, we owe a debt of gratitude to the child support enforcement staff in each of the Parents’ Fair Share sites who spent time discussing their experience and helped us track the results of the enhanced child support enforcement involved in the program.

1. Introduction

In the last 30 years, the American family has undergone dramatic changes owing to rising rates of divorce and nonmarital childbearing. Today, it is estimated that about half of all American children will spend at least some part of their lives in single-parent households. These families — most of which are headed by mothers — are more likely to be living in poverty than those headed by two parents and to be at greater risk of needing welfare. Regular child support payments would afford some of these families a measure of economic relief, yet many do not receive it, often because they do not have support orders in place or because the orders are not being enforced. The connection between more effective enforcement of child support obligations and reduction of welfare receipt has long been recognized and has led to a series of reforms over the last two decades. But despite improvements, the record of success remains mixed.

The recent overhaul of the nation’s welfare laws has made the task of improving child support enforcement (CSE) for low-income families more urgent than ever. The 1996 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) ended the federal guarantee of cash assistance under the Aid to Families with Dependent Children (AFDC) program and replaced it with Temporary Assistance for Needy Families (TANF), a federal block grant to states that carries strict work participation requirements and ends federal funding for cash assistance after a five-year lifetime limit for most welfare recipients. As a consequence, welfare-dependent families will face great pressure to replace lost benefits with new income streams, including child support payments. The new law also forces states to bear more of the fiscal consequences of their policy choices, since the level of funding for the block grant is essentially fixed for the first six years with no adjustment for inflation. Thus, states will increasingly view child support collections as one way to compensate for anticipated shortfalls in federal revenues in future years. Clearly, child support enforcement will continue to play a critical role in evolving welfare policy.

Collecting child support from low-income noncustodial parents (NCPs) poses special challenges for the CSE system. First, while advances in enforcement techniques have improved child support collections among the nonwelfare families served by the CSE system, they have been less successful in addressing the needs of welfare recipients. Frustrated in their efforts to reach this group, CSE agencies have tended to de-emphasize enforcement for NCPs without known income. Consequently, little is known about the absent parents in such cases. Second, it is uncertain that stepping up enforcement in this area will yield an increase in dollars sufficient to justify the added effort and expense. Research suggests that more effective enforcement would make a substantial difference overall, but says little about how the increment would be distributed across income groups. The NCPs of children on welfare are in large part an unknown quantity; how many could pay support if compelled to do so is a matter of some debate.

This report examines the experience of the Parents’ Fair Share (PFS) program, in which seven local CSE agencies attempted to work with their AFDC-related caseloads more intensively than before. The PFS experience suggests several things. First, while current CSE practices may not be ideally suited to dealing with the NCPs in these cases, they can be adapted to work better. Second, locating these NCPs may pay unexpected dividends by allowing CSE staff to sort out the "unwilling" from the "unable"; that is, the process may unearth enough employed fathers who can pay support to make the effort worthwhile. Third, a PFS-type program can be instrumental in improving enforcement against unemployed low-income NCPs by performing monitoring and follow-up functions that the courts or administrative agencies are not able to do well. Moreover, by providing a constructive alternative to the usual punishments available for noncompliant obligors (people with child support obligations), PFS-type programs help lend "political legitimacy" to aggressive enforcement efforts against an extremely disadvantaged group.

I. The Parents' Fair Share Demonstration

The Parents’ Fair Share (PFS) Demonstration was designed by the Manpower Demonstration Research Corporation (MDRC) to assist unemployed NCPs of children receiving welfare in securing employment, paying child support, and participating more fully as parents. The program was conceived in response to the 1988 Family Support Act’s authorization of a demonstration under the Job Opportunities and Basic Skills Training (JOBS) Program to provide services to unemployed NCPs of children receiving welfare who are unable to meet their child support obligations.

As shown in Exhibit 1, PFS offers employment and training services, together with peer support groups and mediation designed to strengthen parent-child bonds and to reinforce NCPs’ sense of obligation to pay support. Participants in the program usually have their monthly child support order reduced to zero or some small amount (often $50, the amount of child support that custodial parents (CPs) receiving AFDC were allowed to keep) while they are active in program services, but are also subject to much closer monitoring of the payment status of their case and their efforts to find work. For their part, PFS participants are expected to cooperate with the child support enforcement system, report employment and have their order raised appropriately, and make regular child support payments as soon as they are working.

The seven sites in the PFS Demonstration are Duval County (Jacksonville), Florida; Hampden County (Springfield), Massachusetts; Kent County (Grand Rapids), Michigan; Los Angeles County, California; Mercer County (Trenton), New Jersey; Montgomery County (Dayton), Ohio; and Shelby County (Memphis), Tennessee. The sites all contain a middle- to large-size city, with Los Angeles being one of the nation’s largest cities. Each site has developed a distinctly different CSE system, with different emphases and priorities, as well as varying degrees of effectiveness.

Exhibit 1

Parents’ Fair Share Program Model

---

• Employment and training. The centerpiece of Parents’ Fair Share programs is a group of activities designed to help participants secure long-term, stable employment at a wage level that will allow them to support themselves and their children. Since noncustodial parents vary in their employability levels, sites are strongly encouraged to offer a variety of services, including job search assistance and opportunities for education and skills training. In addition, since it is important to engage participants in income-producing activities quickly to establish the practice of paying child support, sites are encouraged to offer opportunities for on-the-job training, paid work experience, and other activities that mix skills training or education with part-time employment.

• Enhanced child support enforcement. A primary objective of Parents’ Fair Share is to increase support payments made on behalf of children living in single-parent welfare households. The demonstration will not succeed unless increases in participants’ earnings are translated into regular child support payments. Although a legal and administrative structure already exists to establish and enforce child support obligations, it is critical for demonstration programs to develop new procedures, services, and incentives in this area. These include steps to expedite the establishment of paternity and child support awards and/or flexible rules that allow child support orders to be reduced while noncustodial parents participate in Parents’ Fair Share.

• Peer support. MDRC’s background research and the pilot phase experience suggest that employment and training services by themselves will not lead to changed attitudes and regular child support payment patterns for all participants. Education, support, and recognition may be needed as well. Thus, demonstration programs are expected to provide regular support groups for participants. The purpose of this component is to inform participants about their rights and obligations as noncustodial parents, to encourage positive parental behavior and sexual responsibility, to strengthen participants’ commitment to work, and to enhance participants’ life skills. The component is built around a curriculum, known as Responsible Fatherhood, that was supplied by MDRC. The groups may also include recreation activities, "mentoring" arrangements using successful Parents’ Fair Share graduates, or planned parent-child activities.

• Mediation. Often disagreements between custodial and noncustodial parents about visitation, household expenditures, lifestyles, child care, and school arrangements — and the roles and actions of other adults in their children’s lives — influence child support payment patterns. Thus, demonstration programs must provide opportunities for parents to mediate their differences using services modeled on those now provided through many family courts in divorce cases.

---

II. An Overview of This Report and Its Central Findings

PFS research on child support enforcement has several goals. First, it seeks to provide insights into the interaction between local CSE systems and noncustodial parents whose children are on welfare. The approach taken in this report is to analyze what happened when the seven sites in the PFS Demonstration sought to identify low-income, unemployed NCPs appropriate for PFS and refer them to the program. The report carries this story up to the point of referral of appropriate NCPs to the program. Later reports in the project will continue the story, examining the implementation of PFS’s enhanced CSE for NCPs referred to the program and estimating program impacts on payment of child support and other key outcomes.

Efforts to improve CSE have largely focused on noncustodial parents with income and assets. Location and enforcement techniques, such as matches of administrative records and automated enhanced enforcement actions, work best with the NCPs whose residence, employment, and financial resources are stable. PFS research suggests that a significant portion of NCPs whose children receive welfare do not fit this profile. They continue to pose enforcement challenges precisely because it is often difficult to determine their residence and employment. The information available to CSE staff suggests that they have few financial resources and are unlikely to pay much in support. Hence, in many jurisdictions these cases remain a frustration to CSE agencies, causing them to turn their attention to other cases. For many sites, therefore, the effort to refer unemployed, low-income NCPs to PFS represented a shift in policy. In the process of making this change, the sites encountered a series of administrative and policy challenges that are not unique to PFS. The ways in which they sought to address them provide lessons for other CSE agencies.

In PFS, local CSE staff were asked to review their caseload of NCPs with established support orders (and, in some sites, cases in which paternity was newly established) to identify NCPs who fit the PFS profile: linked to a custodial parent who is receiving or has received welfare, behind in child support payments, and without evidence of employment. Local CSE staff then called in these NCPs who were potential referrals to PFS for an in-person review of their case status (discussed in detail later in this report) to determine whether they were in fact appropriate for the program. The results of this review reveal the diversity of NCPs and illustrate the complexity of the CSE problem. It suggests the potential for an intervention like PFS to help local CSE staff determine what is happening with absent parents and develop the most appropriate response.

Some NCPs could not be served legal notice of the review or did not appear at the review, suggesting that the first key step in these cases may need to be an intensive location effort. The experience of the sites shows why local staff may have difficulty compelling NCPs to attend hearings and how an enhanced location effort might be effected. Normally, CSE staff do not have an in-the-community presence. Typically, they use the mail, phones, administrative records, and so on, to locate NCPs. As explained above, there are limits to the usefulness of this approach for NCPs who are poor and unemployed. Even when local CSE staff reach the stage of an arrest warrant, they are often dependent on the actions of other agencies with many other pressing — in some cases, higher — priorities. In part, this office-based style of enforcement is due to resource constraints that do not allow CSE agencies to extend their investigative work into the community. But it also reflects what appears to be an organizational culture in some sites that relies more on the authority of the legal process and less on legwork in the community than is the case with some other public agencies charged with enforcing major obligations. The case study of one site presented later in this report illustrates what occurs when new approaches are undertaken.

Many NCPs, however, did appear at their status reviews and CSE staff were able to sort cases and respond appropriately. For a substantial proportion of the NCPs who appeared at the reviews (probably between one-fourth and one-third), the review produced an enforcement success; the NCPs reported employment previously unknown to the CSE staff and actions were taken to put in place a wage-withholding order. In other cases, local CSE staff collected information about NCPs that they had not uncovered in normal practices — for example, that they were disabled or incarcerated, living with their child, or even deceased. (See Chapters 3 and 4 for details.) For these cases, the added effort allowed local CSE staff to update their records and see that a current support obligation was inappropriate (though past arrears could still be owed). For the remaining cases, between one-fourth and one-third of those who were tracked during a period of PFS intake, the NCPs were appropriate for PFS and were considered for referral to the program.

These findings suggest that agencies could put in place a PFS-style program as a standard response to cases that appear to fit their eligibility rules (based on what is known through standard CSE practices) and not be overwhelmed with the cost of providing services. The costs of the initial stages of program outreach and referral — which should be seen as part of the program — might turn out to be more than covered by the upfront "smokeout" of jobs and resulting payments, and the absolute number of referrals is likely to be much smaller than the initial pool of potential referrals for the reasons cited above. In effect, part of what makes a PFS-like intervention feasible is its success in helping CSE staff determine the current status of cases that would otherwise be unclear.

A second implication of this drop-off is that those NCPs who turn out to be appropriate for PFS often face substantial barriers to employment. Many lack education credentials, have weak basic skills and a work history with substantial gaps and periods of unemployment, have a criminal record and have been involved in underground or even illegal activities, and suffer from great instability in housing and limited social support networks. Further, few are receiving any form of cash assistance, leaving a sizable proportion strapped for money. For these NCPs, the standard CSE measures such as seek-work orders, purge payments (defined later in this report), or the threat of incarceration may be inadequate if the long-term goal is to get them in a position where they could pay child support.

In sum, the initial stages of the demonstration strongly suggest that a commitment to offering PFS-like services when appropriate appears to have a beneficial effect on many aspects of enforcement. It provides a means of smoking out unreported employment and resources and of identifying those NCPs against whom enforcement is inappropriate. It offers a service option in cases in which the problem is not enforcement but lack of opportunity, skills, or job readiness. In addition, a PFS-style program can serve as an adjunct to the CSE system and the courts in cases in which information on their status is costly to obtain. PFS participation requirements and the program’s monitoring of compliance can put teeth into the mandate to seek employment and pay support.

Before moving to the details of the PFS Demonstration and its implications for CSE practices, this report presents some background on the child support system and the legal and organizational context in which CSE and PFS operate. While this might seem like old news to some readers, we believe that some recent research on CSE has ignored these facts as they apply to poor NCPs and — as a result — has drawn inappropriate conclusions about how CSE does and can work for poor, unemployed NCPs of children receiving welfare. We then turn to the findings from PFS and conclusions for policy and program operations.

2. An Overview of Child Support Enforcement

Child support was originally the domain of state courts, in which traditional precepts of family law were used to resolve what were considered private disputes. Historically, most cases were brought to the attention of the system only if the parent legally entitled to support filed an action to enforce this right. Judges followed no uniform standards in setting awards, so that there were wide variations in the amounts ordered, even among similarly situated parties. Recent research suggests that, as a result, awards for NCPs generally were set at lower levels than many absent parents (usually the fathers) were actually able to pay, and their children paid the price. As divorce rates and births outside marriage began to increase rapidly, the inequities and inefficiencies of the child support system became a matter of public concern.

It was with the aim of introducing greater standardization of enforcement, producing more child support payments, and reducing growing welfare costs that the federal government began to play a major role in child support, starting with the 1975 amendment to the Social Security Act creating the federal Child Support Enforcement, or IV-D, program. While states and localities retained primary responsibility for administering the IV-D system, the newly created federal Office of Child Support Enforcement (OCSE) was charged with approving state plans for delivering services and providing oversight, monitoring, and technical assistance. Federal funding for IV-D programs, in the form of matching funds and incentive payments, was also authorized. And AFDC recipients were required, as a condition of continued eligibility for benefits, to assign their rights to support to the AFDC program and to cooperate in finding their child’s father. Through these measures, it was hoped, some AFDC costs would be offset with increased child support collections, and some families might be prevented from going on assistance.

Congress has responded to the continuing need to improve CSE performance on the state and local level with a variety of measures that expand the federal oversight role. Many of the reforms are designed to strengthen the capacity of IV-D systems through automation. For instance, federal matching funds at a rate of 90 percent were made available to states to develop automated information systems. The Federal Parent Locator Service (FPLS), providing ways to match CSE cases against other federal administrative records, was established. The 1996 welfare law continues the trend toward easing access to information on individuals. Among other things, states are now required to obtain access to private and public records on individuals without the necessity of a court order, and state IV-D agencies must exchange information with an expanded list of state and federal databases. States must also give administrative subpoena power to their IV-D agencies.

Other measures addressed speed and standardization of results. Performance standards for paternity establishment were set; states were encouraged to develop expedited processes for establishing and enforcing support orders; they were required to develop and use uniform guidelines to determine award levels; and they had to review support levels and make appropriate modifications at least every three years in public-assistance-related cases. The new welfare law takes these efforts a step further by strengthening in-hospital paternity establishment requirements and mandating that states establish centralized collection and disbursement units.

Enforcement techniques have been progressively strengthened over the years. Immediate and automatic income-withholding has been made the rule; and tools such as tax intercepts, property liens, and license revocations are becoming more widely used. The 1996 welfare law requires states to use a variety of administrative enforcement mechanisms that depend on automated case processing instead of case-by-case handling. Liens must now be able to be issued administratively, and states must have the authority to withhold, suspend, and restrict occupational and driver’s licenses.

I. CSE Reform: A Middle-Class Paradigm

Although advances in CSE techniques have been driven to a large extent by the desire to reduce welfare spending, they have proven to be best suited for identification and location of NCPs who are stably employed and housed, with income and assets. PFS research reveals that while some of the NCPs in welfare-related cases do in fact fit this profile, others live at the margins of society. Many PFS participants have sporadic work histories, characterized by frequent job changes. Some have no fixed place of residence, living in a succession of relatives’ and friends’ homes. Understanding how these NCPs’ lives differ from the norm is key to successfully adapting CSE practices to reach them.

The importance of NCPs’ socioeconomic status can be illustrated by looking at the problem of location. Parent location services that establish links to information sources such as credit bureaus, tax authorities, employment security agencies, and motor vehicle bureaus work only for NCPs who earn a regular income or own assets. Moreover, these automated systems cannot operate in "real time" because of lags in the reporting of information. Thus, they are effective only when NCPs stay put long enough for the records to reflect their current circumstances.

To a large extent, unemployed low-income men live outside many of the systems on which CSE-related information-gathering depends. When they find work, they tend to hold onto those jobs for shorter periods of time than their higher-income counterparts, frustrating the IV-D system’s attempts to keep track of them. Further, they are much less likely to be tied to other government institutions than higher-income NCPs or low-income women. Experience from PFS suggests, for example, that unemployed low-income men are much less likely to receive public assistance or other social services or to participate in government-funded employment programs than are their female partners. Thus, reforms that focus on increasing and speeding up the IV-D system’s access to public and private databases may be irrelevant to locating many low-income men. Indeed, the most reliable source of information about unemployed low-income NCPs’ current status may be the family and friends of the NCPs or the custodial parents (CPs) themselves.

Besides making them hard to locate, the lack of stability in many unemployed low-income NCPs’ lives renders problematic application of the usual standards for determining appropriate award levels. For instance, state guidelines that fix awards based on a proportion of income assume a steady income over time, punctuated at most by only brief periods of unemployment. But periodic unemployment may be an unavoidable fact of life for many low-income NCPs, especially if they are men of color facing job discrimination in inner cities with weak job markets. For them, imputing steady earnings — even at a low wage — can dramatically overstate their "potential income." The difficulty of determining potential income may be greatest among unemployed low-income NCPs; some are in the midst of temporary spells of unemployment and low earnings, while others will suffer through long periods of poverty.

The use of state income guidelines to determine award levels may not make adequate provision for those cases in which NCPs are truly destitute. Many PFS participants complain that the award levels mandated by the guidelines are unrealistically high, and that after child support is deducted they do not have enough to live on. While some PFS participants acknowledged that they did not make regular support payments, they claimed to make occasional cash contributions or to help in other ways, by buying food, diapers, and other necessities, as their cash flow permitted. The fact that direct payments to CPs on welfare are treated as fraud, or that in-kind contributions are not counted by the IV-D system toward support, is widely viewed by these NCPs as inequitable.

The IV-D agencies participating in PFS tried to respond to these concerns in a variety of ways. Existing awards for PFS participants were either reduced or suspended as long as they complied with program requirements; and procedures for upward or downward modification of awards in response to changes in the NCPs’ employment circumstances were streamlined. Other areas of concern, however, such as treatment of accumulated arrearages and in-kind contributions, were beyond the scope of this project.

In sum, welfare-related child support cases present special challenges for IV-D agencies. Despite efforts to promote the enforcement of all cases, those in which the obligors have no discernible employment or assets are not considered a high priority because they offer little prospect of return and are labor- and time-intensive as well. This is especially true when the caseworkers carry a huge caseload, as in all of the PFS sites. But in giving short shrift to AFDC-related cases, the IV-D agencies sacrifice their ability to distinguish between the "unwilling" and the "unable" or to identify those obligors against whom enforcement is simply not appropriate.

Chronically unemployed or underemployed NCPs also present a challenge to CSE policies designed to standardize support payment levels. Reforms such as state guidelines on award levels and practices such as the imputing of income were in large part a response to historic abuses, and in the case of most obligors they work reasonably well. But the experience of the PFS sites suggests that when dealing with NCPs who are living on the fringes of society, some flexibility on the part of the IV-D system in terms of setting payment levels, imputing income, and implementing modification procedures may be appropriate.

At the policy level, investing resources to locate and enforce appropriately against the most disadvantaged NCPs goes against the national trend toward standardization and mechanization of enforcement. But at the ground level, a more individualized and flexible approach is not entirely inconsistent with current practices. We found that line staff in PFS sites still exercise a certain degree of discretion, even in jurisdictions where computerized systems are designed to prompt caseworkers to take specific measures depending on the status of the case. And in those sites where staff do not have automated tickler lists, the potential for individual variation is even greater. We also observed CSE staff working with NCPs who appeared to be trying to meet their obligation, giving them some leeway because of their good faith efforts. These findings suggest that those who work in the IV-D system want to be able to deal more flexibly with unemployed low-income obligors.

II. The PFS Participants’ Experience of the IV-D System

In PFS, where over three-fourths of the research sample is African-American or Hispanic, CSE issues are inextricably entwined with race. Communities of color, especially the African-American community, have long had a troubled history with law enforcement. The days of state-sanctioned police violence against members of these communities are within living memory; indeed, newspaper headlines remind us daily that police brutality remains a serious problem. African-American men are arrested, convicted, and incarcerated disproportionately in relation to their numbers, and in many black communities it is an article of faith that the criminal justice system is out to get black men. For those who share that belief, it is a small leap to the conviction, expressed by many PFS participants, that the CSE system is simply part of a larger pattern of persecution.

The institutional links between the CSE and regular criminal justice systems are, of course, real. Most PFS sites report that the majority of NCPs arrested for outstanding child support warrants are originally detained as a result of regular police arrests for offenses unrelated to child support. Moreover, the racial and gender makeup of the staff in various agencies involved in the enforcement process echoes that of the criminal justice system, increasing the NCPs’ feelings of alienation. Court or administrative staff who process NCPs through the bureaucratic intricacies are often female, with an increasing proportion of males and a larger percentage of white staff as one goes up the hierarchy within the CSE agency and the courts. Most formal hearings are held in courtrooms presided over by white judges. African-American NCPs, in particular, are dubious about getting a fair shake from a system that looks very much to them like a part of law enforcement in general.

The NCPs in PFS overwhelmingly report feeling that they are not given a chance to be heard at their hearings. They perceive that the dice are loaded in favor of the CPs, not understanding that the lawyers facing them across the table are representing the state (more specifically the CSE agency and welfare department) and not the CPs. Eager to tell their side of the story, these NCPs often find that judges and court personnel are indifferent to the facts of their particular case, often because they are legally irrelevant to the issue of whether NCPs have met child support obligations. For instance, allegations that a CP refuses to grant visitation with the child or is squandering her money rather than using it to provide for the child are not germane to the question of whether there is a legal support obligation. Not understanding these concepts can lead to feelings of frustration and humiliation among NCPs.

Not surprisingly, many PFS participants display cynicism and hostility toward a system they view as unconcerned with the harsh realities of their lives and interested only in squeezing money out of them. Some, for instance, cited payment allocation rules under AFDC, requiring all but the first $50 of child support collected to go to the state to pay off the welfare debt, as justification for their noncooperation with the system. They were vociferous in complaining about the unfairness of having the bulk of their payments go to the state and federal government rather than to their children.

Such negative attitudes undoubtedly have an impact on the CSE system’s ability to gain even minimal cooperation from NCPs. In conducting the initial outreach to potential PFS-eligible NCPs, the program sites sent out thousands of hearing notices. But, as discussed elsewhere in this report, the appearance rate was generally low. Without the means to go after the nonresponders, the PFS sites were forced to rely on "voluntary" compliance in what appears from the outside to be among the most mandatory of institutions. No doubt many of those who did not show up were motivated primarily by the desire to avoid their obligations; for them, the most appropriate response may be to strengthen available sanctions with greater certainty of punishment. But these findings also suggest the possibility that some NCPs’ noncooperation is rooted in negative attitudes about the IV-D system that are carried over from their experience with law enforcement in general. For these NCPs, better outreach and education efforts might engender more cooperation with the system.

3. An Overview of the PFS Intake Process

At this point, it is useful to summarize (1) the steps in the process of identification and referral of noncustodial parents to PFS, (2) the way in which this process facilitated the sorting of CSE cases, and (3) results of this sorting process. During the PFS Demonstration, NCPs might be identified for the program in two ways: intensively working a random sample of CSE cases (which occurred in three sites) and reviewing the status of NCPs who attended review hearings (which occurred in all seven sites).

illustrates these two methods of PFS intake. In three sites, local staff provided MDRC with a list of CSE cases that appeared to meet the PFS eligibility criteria: the CP is receiving or has received AFDC, there is at least one child in the household for whom current child support is owed, the NCP has an address on file within the PFS program service area (usually the county), and child support payments are not current. MDRC staff then randomly assigned one-third of these cases to a research group designated to receive standard CSE (the "standard" group) whose members were not eligible for referral to PFS. The other two-thirds of these cases were randomly assigned to a research group slated for enhanced child support enforcement efforts (the "enhanced" group); members of this group could potentially be referred to PFS. Local staff made special efforts to determine whether these cases were appropriate for PFS, as discussed below. In order to test the effectiveness of PFS program services, those NCPs in the enhanced group who appeared for a review hearing and appeared to be appropriate for PFS were randomly assigned to the program group (referred to PFS) or the control group (subject to standard CSE). The details of this process are discussed below in section I.

In all seven sites, local CSE staff worked to identify through a variety of means (besides drawing a random sample from the caseload) other NCPs who appeared to be potential referrals to PFS and to determine at a review whether this was so. Enforcement staff reviewed cases they were handling and brought potential referrals in for a hearing or reviewed the regular court dockets for enforcement cases in which NCPs appeared to be potential referrals. When NCPs appeared at a review hearing, local staff determined the status of their case and whether referral to PFS was appropriate. As discussed above, those NCPs for whom PFS appeared to be an appropriate response were randomly assigned to the program group (referred to PFS) or the control group (subject to standard CSE). The details of this process are discussed below in section II.

I. Intensively Working a Random Sample of the CSE Caseload

In three sites (Ohio, Michigan, and Tennessee), local CSE staff started with a randomly

Exhibit drawn list of NCPs from their CSE caseload who appeared to be eligible for PFS. Cases in this group were slated for enhanced CSE: local staff were asked to make special efforts to review the status of each case, locate the NCP and serve him with legal notice of a hearing, hold the review of the case, determine current employment status and employability, determine current support order and payment status, and identify appropriate cases for possible referral to PFS. Thus, the information from these sites illustrates what happens when local CSE programs aggressively work a random sample of welfare-related CSE cases in which the NCPs have an address in the county on file and payments are not current.

In these three sites, local staff determined that between 8 and 24 percent of the cases on the enhanced enforcement lists were appropriate for referral to PFS during approximately two years of follow-up efforts. Details on this effort are presented later in this report, but data from Montgomery County, Ohio (the middle site in the range), and Kent County, Michigan (the high end of the range), illustrate the difficulties and successes encountered in this enforcement effort, as shown in Exhibit 3.

• Location problems. Local CSE staff were unable to make contact with slightly over 20 percent of the NCPs and/or get them to attend a hearing. In Kent County, this was primarily because of difficulties in locating and serving them with legally sufficient notice of the hearing. In Montgomery County, it was a mix of inability to serve and nonappearance at the hearing.
• Discovery of previously unreported employment or resources. In approximately 25 percent of the cases, NCPs either reported a job previously unknown to the CSE agency (for which an income deduction order was possible) or otherwise made child support payments without any referral to PFS. This smokeout of previously unreported resources does not automatically produce increased support payments, for reasons discussed later in this report, but calculations by site staff suggest that the resulting payments have been substantial.
• Identification of cases appropriate for PFS referral. In the two counties, 20 percent (Montgomery) and 25 percent (Kent) of the NCPs attended the review hearing and were identified as appropriate for PFS referral, thus meeting all the eligibility grounds for the program. As mentioned earlier in this report, these NCPs tend to have serious barriers to employment (lack of education credentials, unstable work history, high rates of criminal arrests, and so on), reinforcing both the need for the PFS option for a portion of the caseload and the challenge for agencies providing program services.
• Identification of cases inappropriate for PFS referrals. In approximately 25 percent of the cases, the NCPs turned out to be inappropriate for PFS for reasons previously unknown to the CSE agency. Some NCPs (1 in 3, or 1 in 12 cases overall) were disabled and receiving SSI, Veterans, or Social Security disability payments; thus they should not have been facing a current child support obligation. Others no longer faced a current child support obligation because they were living with the CP and/or child or the child was legally emancipated. Some NCPs were ineligible for the program because they were incarcerated and thus unable to work. The remaining NCPs in this category either lived outside the county or state (and thus could not be served in the PFS program), were deceased, or had a support order in which there were legal procedural problems in enforcement.

Exhibit 3

Resolution of Enhanced Enforcement Group Cases in Two Parents' Fair Share Sites

Source: MDRC calculations from enhanced-group logs completed by local Child Support Enforcement (CSE) staff.

II. Screening Through Other Means for NCPs Appropriate for PFS

All seven PFS sites identified NCPs who were appropriate for PFS by screening NCPs who appeared for various types of CSE reviews. The mechanisms by which hearing dockets were set varied from site to site. In some sites, special support enforcement hearing dockets were established just for NCPs that local CSE staff identified as potential referrals to PFS: that is, the CP was receiving or had received AFDC, and the NCP was not making required payments on a child support order. In others, local staff routinely screened the regular paternity establishment, order setting, and support enforcement hearing dockets for NCPs appropriate for the program. Once again, the details of how this was done vary and are explained later in this report.

To assess the overall picture of NCP identification, review, and referral, staff in six sites tracked what happened to NCPs whom they initially identified as potential PFS referrals as the hearing process unfolded. Over the period in which this tracking was in place (which varied from one to six months among the sites) the basic experience was as follows:

• Appearance rate. The rate of appearance at hearings varied widely, reflecting differences in the nature of the enforcement hearings, the accuracy of the addresses in the CSE database, the notice sent to NCPs, and local enforcement practices. Across the sites, 5 to 70 percent of NCPs initially identified by local staff as potential PFS referrals actually appeared as scheduled for the hearing. The factors influencing the appearance rate will be discussed in more detail later in this report.
• Smokeout of previously unreported employment. Across all sites studied, about one-third of all those NCPs tracked who appeared for a hearing reported employment previously unknown to CSE staff. The rates ranged from 8 to 55 percent. CSE staff were then in a position to institute a wage-withholding order. As before, this smokeout represents an effect of the added enforcement efforts connected with PFS intake.
• Appropriateness rate. The percentage of those NCPs appearing at the hearing who were judged appropriate by local staff for PFS ranged from 10–15 percent in two sites to over 70 percent in another. As a general rule, as sites were able to draw PFS referrals from dockets specially assembled to produce referrals to PFS, this percentage rose sharply.

This summary of the PFS intake process illustrates that the sites were able to improve their knowledge of the status of cases once they stepped out of their usual enforcement paradigm and developed methods to more effectively sort CSE cases based on NCPs’ circumstances. The next chapter presents more detail on how the sites accomplished this.

4. The Steps in the Enforcement Process

This chapter describes the specific steps that local agency staff take to enforce a child support obligation, assessing the challenges they face, pointing out the administrative resources available to address them, and suggesting promising solutions. In the course of this discussion, we seek to offer insights into how a program like PFS increases the ability of local CSE staff to identify cases in which NCPs are unwilling to provide support (in which a pure enforcement strategy is appropriate) or unable to pay support (in which a service approach like PFS can be appropriate).

This emphasis on steps in the process is consistent with the way many of the PFS sites organize their staff to conduct the CSE task. Specific CSE functions of a case are often assigned to different staff members rather than giving one individual responsibility for an entire case. This is especially true of paternity establishment, location, and enforcement. Five PFS sites follow the specialized model (Duval, Hampden, Los Angeles, Montgomery, and Shelby counties), while two follow the individual model, although with a limited amount of specialization (Kent and Mercer counties).

There are trade-offs with both the specialized and individual case assignment models. While specialization allows staff to build up expertise about one part of the CSE process, it also creates incentives for staff members to finish their part of the case and pass it on to someone else for the next step. In one site, for example, it had been the practice early in the demonstration for enforcement workers to refer cases with addresses but no information on employment to the Parent Locator Service (PLS), and then wait for information to be uncovered through the monthly matches of administrative records. In essence, because the PLS was often unable to turn up current information on employment for low-income NCPs, it included a growing proportion of these cases.

I. Identifying Nonpaying NCPs

The outreach process began with the overall child support enforcement caseload in each of the PFS sites. For PFS purposes, this overall caseload was made up of two types of cases: (1) those in which CPs were currently receiving AFDC or for whom past AFDC receipt had generated an arrearage of child support owed the government (CSE agencies track these cases separately because some or all of the child support payments go to the public agency rather than the CP) and (2) those in which there was no obligation owed the public agency; these cases were in the IV-D system at the request of the CP.

The size of the overall CSE caseload in a jurisdiction is a function of its population, the income of families and the number of families receiving public assistance (for which CSE agencies are legally involved), and the proportion of the non-public-assistance-related cases in which CPs request that CSE agencies be involved. Within the public-assistance-related cases, PFS is an option for cases in which paternity is not at issue and a child support order exists or is about to be set.

A. The Usual Methods Used to Identify Cases for Enforcement

When local CSE agencies begin to work this caseload to enforce child support obligations and — in the PFS Demonstration — identify potential referrals to the program, their approach is heavily influenced by three factors: (1) the staffing available for routine enforcement activity, (2) the reports produced by their automated data system that allow them to focus their enforcement efforts most efficiently, and (3) state or local enforcement priorities. Exhibit 4 provides some background information on these issues.

Across the seven PFS sites, two CSE programs are state-administered, with regional offices (Duval County, Florida, and Hampden County, Massachusetts), and the remaining five are county-administered. Choices made at the state and local level about enforcement emphasis and staffing deployment affect the caseload per "frontline" enforcement worker. In six of the seven sites, local CSE staff identified a special worker whose job was to monitor payments on cases, and take enforcement actions as appropriate. As Exhibit 4 shows, the caseloads of these workers vary considerably, but in all cases each worker handles a minimum of several hundred cases, and staff in one site (Shelby County) handle 8,000 to 10,000 cases. Many of these cases involve NCPs who pay support regularly and need little daily enforcement attention, but these large caseloads are likely to force line workers to make choices about how to spend their time and which cases or types of cases to emphasize.

Automated CSE data systems ideally could help frontline enforcement staff make choices — for example, by providing them with listings of cases falling into various payment statuses, such as current in payments, paid recently but not the entire obligation, and no payment received recently. In accord with federal law, all PFS sites have made progress to varying degrees on developing statewide automated data processing and information retrieval systems.

Unfortunately, the complexities of the data-processing tasks involved in these large-scale agencies have prevented some of the PFS sites from having such a resource routinely available for frontline enforcement staff, as the third row of Exhibit 4 shows. In three sites (Kent, Mercer, and Montgomery counties), frontline staff do get such reports on a monthly basis and use them to focus enforcement activities. In two sites (Duval and Shelby counties), reports are produced but they are not routinely used because either the format in which they are provided (cases not grouped by payment status) or their sheer size per worker makes them difficult for frontline staff to use. In two sites, local staff do not get such reports, in one case because much enforcement is done at the state level.

Exhibit 4

Staffing and Information Sources on the CSE Caseload

Source	Duval County, FL (Jacksonville)	Hampden County, MA (Springfield)	Kent County, MI (Grand Rapids)	Los Angeles County, CA (Los Angeles)	Mercer County, NJ (Trenton)	Montgomery County, OH (Dayton)	Shelby County, TN (Memphis)
State or county enforcement agency	State agency: Department of Revenue, with regional offices	State agency: Department of Revenue, with regional offices	County agency: Kent County Friend of the Court	County agency: Bureau of Support Obligations within Los Angeles County district attorney’s office	County agencies: Mercer County Board of Social Services for paternity establishment and order setting and Probation Department for enforcement	County agency: Support Enforcement agency within Montgomery County Department of Human Services	County agency: Child Support Bureau within Juvenile Court of Memphis and Shelby County
Average enforcement caseload per "frontline" support enforcement worker	Case analysts have caseloads of 1,500, which may include some closed cases.	Not applicable, because of local office emphasis on establishing new paternities and support orders; four staff members in each region handle enforcement on high-profile cases.	Friend of the Court caseworkers have caseloads of 4,600.	Family Support representatives have caseloads of approximately 1,300.	Probation Department investigators have caseloads of 700-800 and probation officers have caseloads of 400-500.	Child Support specialists have caseloads of 1,500.	Mediators have caseloads of 8,000-10,000.
Routine reports on nonpaying NCPs produced for frontline workers	System produces reports on all cases, not separated by paying and nonpaying so time-consuming to use given size of caseload.	Not routinely done for local office staff; possible on a special project basis out of state office; state staff have automated reports.	Yes; Friend of Court staff can also produce custom reports by extracting cases with specified criteria.	No, but system does produce information on next administrative action needed in individual cases depending on information about case entered into system.	Yes, system produces "tickler" list of priority enforcement cases.	Yes.	Possible, but hard to use for daily enforcement activity so rarely produced or used.
(continued)

Exhibit 4 (continued)

	Duval County, FL (Jacksonville)	Hampden County, MA (Springfield)	Kent County, MI (Grand Rapids)	Los Angeles County, CA (Los Angeles)	Mercer County, NJ (Trenton)	Montgomery County, OH (Dayton)	Shelby County, TN (Memphis)
Usual frontline staff approach for identifying cases in need of enforcement action	Case analysts typically respond to contacts from CP, to mail received providing location or employment information, to alerts from computer system about needed administrative actions or case status changes, and — when they have time — to information they cull from reports on payment status.	Generally, enforcement on existing orders handled by state central-level staff. Computer-automated enforcement system identifies cases with more than a certain threshold in arrearages and performs a variety of matches against other data sources to identify assets or employment. Regional office staff may get involved in pursuing particularly egregious cases.	Caseworkers respond to contacts from either party; phone calls from either party; mail with address, employment, or asset information; PLS information; computer-generated enforcement lists; and reports of case status changes. Some initial steps are automated and formal hearings are not first resort.	Family Support representatives review their caseloads and recommend cases to deputy district attorneys for court hearings.	Probation Department worker reviews priority list of cases and is prompted to take next appropriate action, including further investigation, filing of court motions, etc.	Child Support specialists review default lists for their caseloads and seek to contact NCP and collect information on employment. Generally reach all these cases within a month or they are referred to PLS.	Generally, mediators respond to contacts from CP seeking enforcement action. Certain categories of public-assistance-related cases with large arrearages identified by computer runs are referred to collection agency.

The task of identifying potential referrals for PFS was piggybacked with the existing local strategy for identifying cases in need of enforcement action and these reflected local efforts to set priorities among the NCPs on the caseload. The fourth row of Exhibit 4 describes how staff in each site identify cases for enforcement action. Not surprisingly, in the site with the largest caseload per frontline worker they operate largely on a "squeaky wheel" basis, with CP complaints getting highest priority.

An experienced case analyst in a site with a midrange caseload per frontline worker and modest use of automated reports of nonpayers described how staff in her unit set priorities:

• Respond to phone calls from parents. If you do not, they will call your supervisor.

• Check your mail and take appropriate actions on postal address verifications, reports of employment, and so on.

• Respond to any alerts produced by the automated data system: for example, ticklers for future follow-up put on the system by the case analyst, changes in the status or configuration of public assistance cases, and data exchange alerts when the staff member has done location checks by matching against other databases.

• If you have time, check payment lists from the state data system for your cases, looking for nonpayers for whom you have adequate location information to take further action.

Informal prioritization takes place even in the PFS site with the smallest caseloads (Mercer County), 700-800 each for the welfare agency investigators handling paternity establishment and first-time orders and 400-500 each for enforcement investigators. As the local CSE staff pointed out, they still face choices about which cases to work most aggressively.

B. Special Methods Developed to Identify Cases for Possible PFS Referral

Exhibit 5 summarizes the methods used in each site to identify potential PFS referrals. In two of the PFS sites (Duval and Hampden counties), local staff reviewed the court dockets set through normal procedures to identify potential referrals. Because the NCPs who are potential PFS referrals are often seen as having little capacity to pay support in the short run, they are often a low priority for CSE workers and relatively few appear on court or other review hearing dockets. Hence, relying solely on existing practice for potential program referrals limits the scale and impact of the program.

Exhibit 5

Process for Identifying Potential Referrals to PFS

Site	Process
Duval County, FL (Jacksonville)	PFS staff reviewed court contempt hearing dockets developed by CSE staff to identify welfare-related cases in which NCPs had orders, had not paid their obligation, and there was no evidence of employment. When NCPs appeared, they completed forms asking if they were working. If not, PFS staff person recommended them for possible referral to PFS. No special outreach or court dockets for likely referrals was instituted.
Hampden County, MA (Springfield)	State emphasis on establishing new paternities and orders meant that approximately 85 percent of cases heard in court were new paternities, so these were major source of potential PFS referrals. CSE staff reviewed court dockets prior to hearing and identified welfare-related cases where NCPs were likely referrals. When NCPs appeared they were asked if they had means to pay child support. If not, they were identified as potential referrals to PFS. Efforts to generate a list of welfare-related cases with existing orders where NCPs were not paying and allocate a portion of court docket to these cases did not produce many referrals.
Kent County, MI (Grand Rapids)	PFS and CSE staff reviewed special lists of potential referrals to program randomly drawn from caseload, identified those employed by checking wage reports and took action to institute wage-withholding, identified others who appeared inappropriate for other reasons (for example, incarcerated or living outside jurisdiction), and notified remaining NCPs that they should appear at Friend of the Court for an appointment about their case status. In addition, staff reviewed status of NCPs attending conciliation meetings and administrative, contempt, and arraignment hearings to identify potential referrals. Also, they attempted to locate potential referrals by tracking births in hospitals where medical expenses were covered by Medicaid.
Los Angeles County, CA (Los Angeles)	CSE staff prepared specially run lists of welfare-related cases in which NCPs were not meeting their support obligations and mailed a notice to these NCPs urging attendance at a group hearing to determine eligibility for PFS. Hearings were set to come shortly before start of a PFS service cycle in one of six different offices around county offering PFS services. NCPs who were called to attend a hearing lived near service center about to begin a cycle. Follow-up letters were sent if NCPs did not appear. CSE staff also screened cases on paternity and order modification hearing dockets for potential PFS eligibility.
Mercer County, NJ (Trenton)	Local County Board of Social Services staff reviewed paternity and order establishment hearing dockets for welfare-related cases in which NCPs were potential referrals to PFS. In addition, Probation Department enforcement staff reviewed their caseloads for potential referrals and a specially dedicated PFS enforcement worker supplemented review efforts of normal enforcement staff. When potential referrals to PFS were identified, they were scheduled for a court hearing, called a motion to enforce litigant rights. When they appeared, NCPs were screened for PFS eligibility.
Montgomery County, OH (Dayton)	CSE staff reviewed special lists of potential referrals to program randomly drawn from caseload, identified those employed by checking wage reports and took action to institute wage-withholding, identified others who appeared inappropriate for other reasons, and notified remaining NCPs that they should appear for a review hearing to determine their eligibility for PFS. CSE staff also reviewed monthly lists of NCPs not meeting their obligations to identify potential referrals to program, and these NCPs were also notified that they should attend a review hearing. Prosecuting attorneys also reviewed NCPs appearing in contested paternity, support, or contempt hearings for possible PFS referrals.
Shelby County, TN (Memphis)	CSE staff reviewed special lists of potential referrals to program randomly drawn from caseload, identified those employed by checking wage reports and took action to institute wage-withholding, identified others who appeared inappropriate for other reasons, and notified remaining NCPs that they should appear for a court hearing. Efforts to work this list produced fewer PFS referrals than in Michigan or Ohio, in part because many addresses of NCPs on list were not current and location was difficult. In addition, CSE staff reviewed special lists of welfare-related cases with no recent payments, lists of NCPs about to exhaust unemployment insurance benefits, and lists of new welfare-related CSE cases to identify potential referrals to PFS and schedule them for court hearings. Finally, CSE staff reviewed the contempt hearing dockets (largely made up of cases in which CP filed a complaint) and paternity establishment dockets to identify potential PFS referrals. CSE and PFS staff interviewed those NCPs who appeared at hearing to determine eligibility.

In the remaining five PFS sites, special outreach efforts were made to increase the flow of potential PFS referrals. These efforts included:

1. Conducting special reviews of cases on the existing caseload to identify NCPs with orders who appear to meet the PFS eligibility rules and scheduling them for individual hearings at which eligibility could be determined. This was done in Mercer and Montgomery counties, for example, where designated staff coordinated the efforts of other frontline staff with caseloads of NCPs and conducted reviews themselves of the entire caseload.
2. Reviewing other lists in search of potential referrals. Examples included new referrals from the welfare agency to the CSE agency of families receiving public assistance, listings of individuals about to exhaust unemployment insurance benefits, and Medicaid-supported births in local hospitals. One site also retrieved several thousand cases from a special PLS caseload, cases that had an address but no information on employment and had been sent to the PLS to determine if any current job could be found. These tactics were used most extensively in Kent, Montgomery, and Shelby counties.
3. Developing hearing procedures to review the status of large numbers of NCPs on the caseload. Mass hearings of cases for NCPs who appeared to be potential referrals proved to be the most effective and efficient way to identify large numbers of parents in need of program services. This was done in two ways. In Kent, Montgomery, and Shelby counties, MDRC worked with the sites to produce randomly drawn lists of NCPs with orders who appeared to fit PFS eligibility rules; and local CSE and PFS staff worked to determine if they were really potential referrals to PFS and to get appropriate NCPs to a hearing where their eligibility for the program could be determined. In Kent and Montgomery counties this approach produced a substantial portion of the sample. In Los Angeles, local CSE staff on their own developed similar lists of NCPs who were sent letters telling them to appear at a specially scheduled mass hearing for possible referral to the program. The hearings produced the vast majority of NCP referrals. In the course of conducting these large-scale reviews, substantial smokeout of previously unreported employment also occurred, as discussed more fully in a later section of this report.

II. Arranging Reviews of NCPs’ Status and Eligibility for PFS

One of the primary aims of federal CSE legislation has been to introduce speed, regularity, and predictability of results to individual child support disputes. The 1984 CSE amendments required states to adopt procedures to process child support matters expeditiously, noting that courts had been unable to handle the growing volume of child support cases without long delays. Expedited processes have been interpreted to include both administrative and judicial processes, and many states have moved in varying degrees to reduce their reliance on the courts. While some have continued to locate primary responsibility for CSE in the courts, others have moved to a quasi-judicial system in which hearing officers or referees typically hear cases in a courtroom setting; and still others have moved to an administrative system in which most aspects of child support cases are handled by the CSE agency, with minimal involvement by the courts.

The seven PFS sites represent in microcosm the diversity of child support systems nationally. Alone among the PFS sites, Hampden County, Massachusetts, has a court-based system, in this case Probate Court, in which only judges can preside over child support cases. The courts, however, generally handle only cases involving paternity establishment and the setting of initial orders, since enforcement — with a few high-profile exceptions — is handled through the state Department of Revenue.

Four of the sites operate a quasi-judicial system in which referees or hearing officers preside, but under close supervision by the courts. Memphis’s CSE agency is housed in the Juvenile Court, and its staff are answerable to the chief judge. All child support matters are heard by referees in courtroom settings, although the CSE agency relies on nonlawyer staff to represent the county’s interests. In Jacksonville, where all cases are heard in the Family Court, the system was changed during the course of the demonstration to allow a hearing officer to preside over child support cases in addition to the regular judges. Contract attorneys are responsible for service of process and scheduling of court dockets. In Trenton, hearing officers hear cases in the Family Court. Court personnel schedule paternity establishment and first-time orders, while the Probation Department, a unit of the court system, schedules and handles all enforcement cases. In Los Angeles, CSE functions are lodged with the district attorney’s office, whose lawyers oversee the scheduling of court dockets and represent the county. Cases are heard in the Superior Court by referees.

The systems in two of the PFS sites are primarily administrative. In Montgomery County, Ohio, the local CSE agency administratively processes paternity and new support orders, as long as they are uncontested. Contested actions are heard by referees in the Juvenile or Domestic Relations Court, but these tend to be time-consuming since the prosecutor’s office, which represents the county, requires that delinquent obligors be prosecuted for contempt, a quasi-criminal proceeding. Kent County, Michigan, has the most thoroughly administrative CSE process. The local CSE agency, which is actually a part of the court system, administratively handles all aspects of child support cases, with referees (employed by the agency) presiding and hearings held in the agency. The only exception is for contempt actions, which are referred to the Circuit Court.

Because of these reforms, local child support agencies, especially those that have moved away from a strict court-based model, have a variety of options for in-person reviews of the child status of NCPs who are potential referrals to PFS. For example, in Michigan options run from requests to contact staff from the Friend of the Court office, by phone or in person, to formal meetings with these staff in the Friend of the Court office, administrative hearings before referees, judicial hearings before a judge of the Circuit Court, issuance of bench warrants for arrest of NCPs for contempt of court, and formal arraignment before a judge once arrests are made. The choice of the type of forum has a variety of implications.

A. Perceived Seriousness of the Review

Less formal reviews (for example, case conferences or meetings to see whether NCPs are eligible for PFS) may be less threatening to NCPs and more likely to convey the message that PFS represents an opportunity rather than enforcement as usual. At the same time, they may also be less likely to focus the NCPs’ attentions on the seriousness of the issue (of nonpayment) and may produce a much lower appearance rate. Finally, very nonthreatening, service-oriented announcements may not be believed; NCPs in several sites initially saw such notices as potentially a sting operation in which they were being lured into attending a meeting and would then be subject to jail or other harsh enforcement tactics.

B. Case Reviewers and Associated Required Facilities, Paperwork, and Docket Space

Using judges (or, to a lesser extent, referees or hearing officers) to review cases sends a strong message about the seriousness of the issues involved. However, in all the PFS sites, these officials handled a variety of types of family law and child support cases, not just referrals to PFS, and it proved complicated (though many of the sites found a way) to institute special PFS dockets just to consider referrals to PFS because of the busy schedule of these officials and the competing demands for courtrooms. Further, more formal hearings often involved more paperwork to schedule. These factors all contribute to the lengthier "lead time" involved for more formal hearings. For example, in one site it took three to four months to bring a contempt action, while a less formal review hearing could be conducted within one month.

C. Requirements for Legally Sufficient Review Notice to NCPs

The more authority the reviewer of the case has to impose sanctions for nonappearance or other noncompliance, the more the CSE agency must comply with specific notice requirements to assure that the NCPs are aware of the review. Typically, court hearings have specific requirements as to the length of advance notice (often 20 days or more), the language needed to inform NCPs of the nature of the proceeding, and the method of delivery (sometimes personal service, but usually at least certified mail). The notice for less formal group meetings to screen for eligibility can be sent with less advance planning, through the regular mail, and with more freedom to emphasize the opportunities that PFS offers.

D. Staff Needed to Represent the Agency

At formal hearings before a judge or — in several jurisdictions — a referee, the CSE agency’s case typically is presented by an attorney, which increases the cost and scheduling difficulties of the review. At other types of reviews, CSE staff who are not attorneys have assumed this role.

E. Enforcement Expectations Triggered by the Choice of Forum

In one of the PFS sites, the lead county attorney had established an expectation that child support cases in which NCPs were found in contempt of court should produce either an immediate payment of some or all of the past-due child support (typically known as a purge payment because it purges NCPs of the contempt of court) or jailing of the NCP. Since reviewing cases to determine eligibility for PFS could produce a variety of outcomes (referral to the program, a realization that enforcement of current support was not appropriate, smokeout and payment, or a seek-work order), using the contempt process for PFS reviews was inconsistent with the goals established for this type of hearing. While it was never certain how much this contributed to the site’s decision to use a less formal review forum, it was part of the context as local staff developed their program.

F. Practices in the PFS Sites

Exhibit 6 describes how the PFS sites resolved these trade-offs during the demonstration period. In four sites, PFS referrals occurred during the normal hearings held for paternity and initial order establishment or contempt of court for nonpayment (or the local equivalent). In three of these four jurisdictions (Duval, Mercer, and Shelby counties), the reviewer of the case is a referee or hearing officer, while in one (Hampden County) a judge hears the cases. In Kent County, referrals are made out of the variety of review forums mentioned earlier in this section.

In Los Angeles and Montgomery counties, special group or "mass" review hearings (with less formality and authority than judicial proceedings) were put in place to determine eligibility for PFS and make referrals. This proved to be an effective way to sort CSE cases and make referrals to PFS. Large numbers of cases could be scheduled with a block of time reserved for a referee or hearing officer. After the first such hearing, agency staff could gauge how many NCPs to schedule (overbooking was the rule) to utilize the available time for actual reviews and the final summary paperwork.

Before attempting to draw lessons from the site experience, it is useful to present information on the appearance rate of NCPs across the sites, the subject addressed in the next section of the report.

III. Getting NCPs to Appear at the Reviews

In planning the PFS demonstration, local CSE staff were aware that the appearance rate of NCPs called to a paternity or enforcement hearing is far from 100 percent and many factors affect the proportion who do attend. As the demonstration developed, monitoring the appearance rate of NCPs who were potential PFS referrals and seeking ways to increase it were high priorities.

Exhibit 6

Forum for the Review of PFS Cases

Site	Type of Review Used
Duval County, FL (Jacksonville)	Most referrals came from hearings before a commissioner of Family Court for Duval County, who was not a judge. Cases were brought on a motion for contempt of court for failure to comply with order to pay support. Cases were presented by a private attorney under contract with state, with assistance from CSE staff. Child support mass hearings (for cases not expected to have complications) were scheduled for half-day sessions, approximately 15 days a month. Notice of hearings was typically served in person, by substitute service (to another adult at address), or by mail, and the notice process was usually begun about 30 days before hearing. Some cases were referred out of paternity hearings, also in Family Court, and a few from order modification hearings.
Hampden County, MA (Springfield)	Most referrals came from hearings before a judge in Probate Court to establish paternity and an order. The case was presented by a CSE agency attorney or other staff. Block of time was set aside on one day of week for hearings of new paternities and potential PFS referrals were mixed in with other cases. Notice of hearings could be served in person or — more usual — by mail. Hearing had to be set at least 20 days after CSE agency received return receipt that service had been made, except for contempt hearings, which could be scheduled sooner. Other cases came from order modification hearings, or — to a small extent — from contempt hearings.
Kent County, MI (Grand Rapids)	Referrals most commonly came from meetings with CSE staff in their offices, which had no docket schedule, judicial involvement, or notice requirement, and additionally from NCPs brought into court on bench warrants for their arrest for nonpayment of child support. Other cases also came from hearings on an order to show cause why NCPs should not be held in contempt that occurred before referees, who typically had 10-15 cases an hour scheduled (this was on assumption that not all NCPs would appear) and CSE staff presented the case. Notice was sent by regular mail and hearing could be no more than 28 days after date of mailing. Cases heard in court before a judge on an order to show cause were presented by a CSE staff attorney. Notice was similar to referee hearing. Bench warrant cases were presented by a CSE staff arraignment officer.
Los Angeles County, CA (Los Angeles)	Referrals came out of a group screening and stipulation process. NCPs on CSE caseload who appeared to be eligible for PFS were sent a letter telling them to report on a date approximately one week later to civil court house in downtown Los Angeles for screening to determine eligibility for PFS. Screening was done by PFS and CSE staff, and those found eligible for program were asked to sign stipulations agreeing to participate in PFS. Those who signed then appeared as a group before a referee, who confirmed their understanding of meaning of stipulation, signed stipulation, and ordered them into program. Since group screening was not a formal hearing and was done without usual legal notice, staff had to institute a separate process to enforce obligation of those who did not appear and/or did not agree to stipulation.
Mercer County, NJ (Trenton)	Referrals came from hearings before hearing officers of Family Court. Special PFS hearing dockets were set for both paternity and order establishment and enforcement of existing orders (called motions to enforce litigant rights). Notice of hearings was normally served by mail (both regular and certified), though sometimes in person. Usual lag between service and hearing was 30 to 60 days. PFS staff also sent a letter to NCPs telling them of program. PFS staff also attended other Family Court dockets to identify potential referrals to program. If service was adequate and NCPs did not appear, hearing officers could issue a default order in paternity and first order cases or a bench warrant in enforcement cases.
Montgomery County, OH (Dayton)	Most referrals came from special group review hearings before a referee conducted solely to determine if NCPs should be referred to program. County prosecuting attorney prepared dockets and handled cases for two courts involved in CSE: Domestic Relations Court for cases in which there had been a divorce and Juvenile Court for cases in which paternity had to be established. Seven-day notices were sent by regular mail, which was supplemented starting in mid-1995 by home visits made a few days before hearing. Prior to beginning of home visits, cases of nonappearing NCPs were dismissed. After home visits began, referees accepted recommendation to institute a contempt action against NCPs for whom there was evidence of actual notice of hearing.
Shelby County, TN (Memphis)	Most NCPs were referred to PFS at hearings before a referee of Juvenile Court for contempt of court for nonpayment of support. CSE staff prepared material for hearing and "presenters" (who are specialists in hearing process but not attorneys) presented agency’s case. Usually notice of contempt hearing was served by mail, though sometimes in person. Usual lag between identification of a case and the scheduled hearing was 2 to 4 weeks. If NCPs failed to appear at a contempt hearing, court usually issued warrant for their arrest. Some referrals were made out of paternity establishment process at the point at which court was setting an initial child support order. If NCPs were unable to pay support and meet PFS eligibility requirements, they could be referred to program.

A. Appearance Rate at Reviews

Exhibit 7 shows the appearance rate among potential PFS referrals and two pieces of information important in interpreting this rate: the point in the hearing process at which potential referrals are identified and tracking begins (the later in the process that NCPs are identified and tracked, the higher the appearance rate) and special features about the local context likely to affect the appearance rate. The rate ranges from an estimated 5 to 10 percent in Los Angeles County to nearly 70 percent in Montgomery County, a range that does not appear to be explainable by the different points at which tracking begins in the various sites. As the table shows, some key features of the local enforcement process help explain these differences: the geographic distance to the hearing site, the use of additional (nonjudicial) contacts prior to the hearing, and the likelihood of sanctions for nonappearance seem especially important.

The low appearance rate in Los Angeles may be related to the scale of the outreach effort undertaken (thousands of letters were mailed to NCPs), which worked against intense individual follow-up with each NCP, the less formal notice, and the difficulty that NCPs encountered in getting to the site of the review (NCPs from across the county were called to a hearing in downtown). Findings from the ethnographic research on the project suggests that many NCPs are short on funds, lack access to a reliable car, and find the public transit system difficult to use for long trips. This may also be a factor in the Duval County appearance rate, since that county is geographically quite large.

Two sites (Mercer and Montgomery counties) at various times in the intake period put in place an additional contact with the NCPs between the time of the official notice of the hearing and the date of the hearing. In Mercer County, PFS program staff sent a letter to potential PFS referrals alerting them to the opportunities the program offers and the possibility they might be able to participate, and encouraged them to attend the hearing. In Montgomery County, the PFS staff attempted to make a home visit to the last known address of each NCP a few days before the hearing to encourage attendance. (This is discussed in more detail later in this report.) Both sites had higher-than-average appearance rates during the period in which these policies were in place.

B. Difficulty of Imposing Sanctions for Nonappearance

Instituting an action to hold noncooperating NCPs in contempt of court (or similar procedures under local rules) is a remedy theoretically available in all PFS sites, but actually applying it is difficult and time-consuming. First, the delinquent obligors who are the most appropriate candidates for this sanction are usually the ones who cannot be found. Second, invoking contempt invariably requires the involvement of judges, so that even those sites that moved more toward an administrative model (for example, Kent County) have to go to court. The IV-D agencies are usually represented by lawyers, as opposed to nonlawyer staff who normally represent the agency at hearings, and in two sites, where contempt is considered a quasi-criminal proceeding (Los Angeles and Montgomery counties), NCPs are entitled to counsel as well.

Exhibit 7

Appearance Rate of Potential PFS Referrals and Factors Affecting It

Site	Appearance Rate at Reviews	Point at Which Tracking Begins	Factors Affecting Appearance Rate
Duval County, FL (Jacksonville)	27%	When court dockets were reviewed and public-assistance-related cases were identified	Some problems with serving notice reported and large geographic area for county
Kent County, MI (Grand Rapids)	30%a	When NCPs from public-assistance-related cases were scheduled for a review meeting	Many NCPs called to less formal reviews
Los Angeles County, CA (Los Angeles)	estimated 5 to 10%	When letters were sent to NCP in public-assistance-related cases within service area of a PFS service center	Less formal hearing and notice; NCPs drawn from across large urban county with all reviews held in downtown Los Angeles location; mobility of NCPs meant many addresses were outdated; short notice; notice emphasized opportunity, not sanctions; sanctions for nonappearance difficult to impose; fear among some NCPs that this was a sting operation
Mercer County, NJ (Trenton)	56%	When the staff scheduled an NCP from a public assistance-related case for a review meeting.	Significant proportion of NCPs involved at stage of paternity establishment when fathers were more involved with children; follow-up letter by PFS program; most NCPs lived near the location of hearing; formal hearing forum and notice
Montgomery County, OH (Dayton) Prior to prehearing home visit	42%	When notices were sent to NCPs ordering them to appear for a review hearing	Less formal hearing and notice; short notice; lack of sanction; cases dismissed if no appearance
After prehearing home visits begun	69%	When notices were sent to NCPs ordering them to appear for a review hearing	Similar hearing and notice, but supplementary notice was left at prehearing home visits; when home visits yielded proof NCP was given notice, then CSE staff recommended contempt proceedings for those not appearing
Shelby County, TN (Memphis)	67%	When court dockets were reviewed and public-assistance-related cases were identified	Formal court hearing and notice; many NCPs lived in downtown area; strong threat of jail for nonappearance

NOTE: Data for Hampden County are not available because about 70 percent of referrals were from paternity hearings, which made it difficult to track potential referrals prior to the hearing.

^aThis includes people called in for informal meetings.

When contempt proceedings are brought, judges will typically issue bench warrants for the NCPs’ arrest. But effecting the arrests is not always easy, since arresting officers are usually employed by agencies not under the direct control of the courts or the IV-D agency. Often the sheriff’s office is responsible for CSE arrests, but because it is also involved in a wide range of cases, child support cases are usually assigned a low priority. (Kent County is the one exception where a part of the Friend of the Court is dedicated to making arrests in domestic relations matters.) In three jurisdictions, contempt was rarely invoked in connection with PFS for slightly different reasons, all of which were related to the time-consuming and labor-intensive nature of the process. Because few PFS sites had the resources or inclination to specially target child support violators for arrest, most of those who were arrested were picked up on charges unrelated to child support, such as traffic violations.

Even when arrested, most NCPs spend little or no time in jail. In all sites, NCPs who are arrested face jail, but in five of the seven sites they can be released if they pay a specified amount, set to be within their power to pay. Sometimes, when an NCP has spent enough time in jail to convince court officials that he cannot pay the originally specified purge payment, a lower amount is negotiated. In the remaining two sites, NCPs can be sent to jail for a specified time; in Los Angeles, this is because the CSE agency institutes a criminal contempt proceeding and in Montgomery County the court often sets definite terms, although release is often negotiated once payment is made.

In most of the PFS sites, there is a low probability that NCPs who do not appear in response to a summons will get caught or that if they are caught they will spend time in jail.

C. Other Features of the Intake Process

Some features of the intake process that initially were expected to be important had little effect in the final analysis, at least compared with the more systemic or macro features already discussed. Most important, efforts to fine-tune the message in the notice of the review (for example, "toughening" it by emphasizing obligations and potential sanctions and downplaying services and the opportunity afforded by PFS) appear to have made little difference in the appearance rate. For example, in Los Angeles County CSE staff began program outreach using a letter that told NCPs they had "been selected as a potential candidate" for PFS, an "exciting new program developed to assist you in becoming gainfully employed so you will be able to contribute to the support of your children." Although the final section of the letter stated clearly that this was a "mandated program" and failure to attend "may result in court action," the tone of the letter was one of opportunity and encouragement. During this early period, the appearance rate hovered just below 5 percent. With this experience, local staff decided to toughen up the letter, emphasizing the seriousness of the child support obligation and making more explicit the threat of court action (mentioning the possibility of instituting criminal contempt proceedings). Following this shift, the appearance rate moved up to between 8 and 10 percent.

IV. Using the Review to Assess NCPs’ Circumstances

In planning PFS, one of the most frequently mentioned frustrations of CSE — for agency staff and judges — is the inability to be certain whether NCPs brought into a review for nonpayment present a support enforcement problem (they have the means but will not pay) or are there because of limited employment opportunities (unemployed or underemployed because of their inability to get and keep a job). With traditional enforcement tools (sanctions or seek-work orders), the judge or agency is forced to choose which type of error to risk: letting NCPs presenting an enforcement problem off the hook or subjecting those with few employment opportunities to sanctions. Judges and agencies traditionally had to resolve this dilemma based on local and state laws and procedures, political pressures, and their own implicit theories about the characteristics and behavior of NCPs.

PFS offers a way for judges and agency staff to avoid this dilemma: assume an NCP’s problem is one of limited opportunity and refer him to the PFS program; if, in fact, he has a job or the ability to get a job easily, the program participation mandate can smoke out this fact. NCPs who are employed will face a decision: report their job to the system and begin paying support or fail to comply with the PFS service participation mandate (because they are busy with work) and be referred back to regular CSE where staff will now have evidence that this case presents an enforcement problem. In essence, the PFS program staff become a monitoring arm of the court, a capacity that court staff have previously not had. Judges and hearing officers value the assurance they receive that someone (the PFS staff) is following up to track the NCPs and prod them into working to meet their support obligation.

While PFS does lessen the complexity and uncertainty of the review, the PFS sites still faced many issues in conducting reviews. This section presents an overview of how the sites assessed the circumstances of NCPs appearing for hearings and describes how staff resolved various problems that arose in conducting the hearings. It then summarizes what local staff discovered about the circumstances of these NCPs.

A. Review Setting and Procedures

Typically, the reviews were held in a clean, reasonably well-maintained building, in a courtroom setting. In some buildings, there were metal detectors, armed bailiffs, and other trappings of the justice system reminiscent of criminal hearings. Usually, an area was set aside (either within the courtroom or in a nearby area) for prehearing interviews. In most sites, the reviews were held in an area also used for other types of hearings (most commonly other family law cases) so there were often crowds of people in the building and parents with young children.

Generally, NCPs called to a review were told to appear at the beginning of the morning or afternoon hearing docket. Often many NCPs (and in some cases CPs) would arrive at one time, so there was a sense of frantic activity to prepare the cases for their appearance before the hearing officer (or sometimes a judge). When NCPs arrived, they were logged in and told to fill out information forms about their current residence, employment, and resources. Usually, individuals were then interviewed by a CSE staff member and — if they appeared to be potential PFS referrals — a PFS staff member would seek to learn whether they were eligible for the program.

In most reviews, the judge or hearing officer was willing to allow cases to be heard as staff were ready to present them and did not insist on their appearing in the order listed on the court docket. This allowed CSE and PFS staff time to determine the appropriateness of NCPs for PFS and, in the context of the demonstration, conduct random assignment. (If the PFS research procedures were not part of intake, this potential for a bottleneck would not exist.) One approach was to hear non-public-assistance-related cases first or to give staff the discretion to hold the cases of NCPs who looked appropriate until a final determination was made. Given the usual scarcity of courtroom time and the busy schedule (and status) of hearing officers and judges, staff were aware that PFS procedures would be overridden if they did become a bottleneck. To avoid this, reserve staff were usually available and in some sites the number of staff committed to the intake process was set to be able to cover peak flows of cases. This led to an appearance of overstaffing during other times of the day.

As individual cases were ready for a hearing, they were brought before the hearing officer. Each hearing usually lasted 5 to 15 minutes and the NCP was not represented by counsel. Normally, the hearing began with the hearing officer asking the representative of CSE (either an attorney or other staff person) the facts of the case. This short recitation usually focused on the amount of the order, payments, and arrearages. The hearing officer would then ask the NCP whether this was correct, and in most cases the NCP would not dispute the CSE records. The hearing officer would then ask the NCP about his financial and employment circumstances. At this point, the NCP might inform the court of employment, plead unemployment or insufficient funds to pay support (which do not override his obligation to pay support), or present information about inability to work (disability, medical conditions, recent incarceration).

Typically, the court and CSE staff did not have the means to independently verify information received from NCPs during a hearing. Administrative records on employment (the unemployment insurance wage reporting) were often several months old (though recent reforms requiring reporting of new hires will make them more current), and normal court routines did not allow for confirmation calls to employers.

In certain circumstances, there were reasons why an NCP might falsely report a job when he in fact was not working or not working at the job reported. If the court’s response to a report of employment is to order the CSE agency to issue a wage-withholding order and the court does not require a purge payment and/or jail, then an NCP who wishes to evade child support could fabricate a job and be released with an admonition to keep the CSE agency informed of future job changes, and the CSE staff would be left holding the bag when the job turned out to be invented. Further, if an NCP wished to avoid being referred to PFS because of its participation requirements, he might similarly invent a job to appear ineligible for the program.

In all sites, if a job was reported, court and CSE staff tried to get enough specific information from the NCP on the employer (name, address, phone, supervisor) so they would later be able to issue a wage-withholding order. After recognizing the problem of false reporting, in at least one site staff changed their usual practice and would put a case on hold briefly while they telephoned the reported employer to verify the job.

B. Circumstances of NCPs Who Appear for Reviews

Exhibit 8 presents information on a sample of NCPs who were tracked by sites from the point at which they were identified as potential PFS referrals. (The same sample was used in Exhibit 7.) The "percent eligible and appropriate for PFS" represents those tracked in each site who were identified as appropriate for PFS and went through random assignment for selection for referral to the program or to a control group. The NCPs identified as appropriate were linked to a public-assistance case; not currently living with the child for whom support is owed; behind or unable to pay support payments; able to work but without a job; and within any of the age, language, or geographic restrictions that the local program imposed.

As Exhibit 8 shows, the percentage of those appearing who were judged appropriate for PFS varied widely, from over 70 percent in Kent County to about 12 percent in Duval County. Overall, slightly more than one-third of the NCPs who appeared at a review were deemed appropriate for the program.

The remainder of Exhibit 8 shows the reasons why other NCPs were deemed inappropriate by local staff for the PFS program.

Exhibit 8
Disposition of NCPs Who Appeared at Review Hearing, Percent by Site
			Percent
			Duval Countya	Kent County	Los Angeles County	Mercer Countyb	Montgomery County Early Cohortc	Montgomery County Later Cohortd	Shelby County	Total
Eligible and appropriate for referral to PFS			12.4	72.2	41.5	13.6	40.6	43.9	17.3	33.7
Not appropriate for referral to PFS			81.0	27.8	58.5	86.4	59.4	56.1	82.7	65.4
Reasons not appropriateb
	Employed		36.9	5.6	31.8	31.9	33.3	35.4	48.8	33.5
	Purge payment amount set		6.5	5.6	0.1	0.0	0.0	5.1	14.2	1.9
	Currently unable to work
	Disability		6.5	0.0	5.6	5.2	5.5	1.7	1.2	5.2
	In jail/prison		4.6	0.0	0.0	2.7	0.0	4.2	0.0	1.2
	Living with CP		4.4	0.0	3.8	2.2	0.0	0.0	0.0	2.9
	Ineligible for PFS Demonstration under local rules
	Living outside jurisdiction		7.6	0.0	0.7	19.0	0.0	0.8	0.0	3.5
	Too young or too old		1.0	0.0	0.1	7.4	0.0	0.0	0.0	1.0
	No required link to AFDC		0.0	0.0	8.4	18.8	0.0	0.0	0.0	6.6
	Does not speak English		0.0	0.0	2.0	3.5	0.0	0.0	0.0	1.5
	No legal right to work (LA)		0.0	0.0	1.8	0.0	0.0	0.0	0.0	1.0
	In pilot phase		15.0	0.0	0.0	0.0	8.7	8.1	0.0	2.9
	Othere		4.2	16.7	4.2	41.7	11.0	25.1	2.5	9.0
	Missing Information		6.7	0.0	0.0	0.0	0.9	8.1	16.0	0.8
Sample size			526	18	2142	405	345	237	162	3835

SOURCE: MDRC calculations from child support tracking logs.

NOTES : Data for Hampden County are not available due to the way in which PFS intake works there. Most referrals were from paternity hearings, which made it difficult to track potential referrals prior to the hearing.

a In Duval County potential referral logs were collected from January to March 1996. More than one reason may have been recorded in Duval County for not randomly assigning an NCP.

b More than one reason may have been recorded in New Jersey for not randomly assigning an NCP.

c Early cohort data were collected from January to June 1995.

d Late cohort data were collected from January to May 1996.

e As of September 10, 1994, Mercer County is only randomly assigning cases in which the custodial parent is no longer on AFDC, but there is both a state debt and a current child support obligation.

Employment. About one-third of those appearing provided previously unreported information on employment. This ranged from a low of about 6 percent in Kent County (the jurisdiction in PFS that other evidence suggests has the most stringent existing level of CSE) to 49 percent in Shelby County. In Shelby County, a large percentage of the cases called for hearings were those in which the CPs filed a complaint with the CSE agency requesting enforcement action. In these CP-generated cases, the chance that previously unreported employment existed could well be quite high; in fact, it might be what stimulated the CPs to lodge the complaint. The remaining four sites fell within the range of 32 to 37 percent smokeout.

While these reports of employment do not automatically translate into child support payments for the reasons cited above, analysis of payments generated by these NCPs has convinced staff that there is a payoff in increased support from these cases. But these early indications suggest strongly a potential for increased child support payments through the case review process that a PFS-like program stimulates.

Purge payment made at review. About 2 percent of those NCPs appearing at a review made purge payments at the time of the hearing. But these payments occurred primarily in two of the sites (Duval and Shelby counties), where about 7 and 14 percent (respectively) of those appearing made purge payments. Such payments would typically occur only in a formal civil contempt hearing, when the judge or hearing officer had found an NCP in contempt of the court order and sentenced him to jail. Most contempt actions for failure to pay child support are civil contempt. In these procedures, the goal is to force compliance with the court’s order rather than to punish for failure to comply, as would be the case in a criminal contempt action. Thus, in civil contempt actions an NCP has the option of paying a specified amount and being released, and the judge or hearing officer is required to set the purge payment at a level the NCP is capable of paying. In the folklore of the law, the defendant must have "the keys to his cell."

In the remaining sites, purge payments were basically not imposed as part of the hearing process leading to referral to PFS. In several sites (Los Angeles, Montgomery, and Kent counties), a finding of contempt and imposition of purge payments was not within the powers of the staff conducting the initial review of most NCPs considered for PFS because of the form of the review. Consequently, the low overall level of purge payments is not a real indication of how prevalent they might be if a jurisdiction had chosen to hear the cases of potential referrals to PFS in a forum with the power to find NCPs in contempt.

Current inability to work. About 5 percent of those NCPs tracked were disabled to the extent they could not be expected to work and another 1 percent were incarcerated. For both categories, there was thus evidence that the support order should be adjusted downward. The PFS review process typically provided the information needed to make this possible, but such a modification was often not possible with the review.

Living with the custodial parent. At the time of the review, staff learned that about 3 percent of the NCPs appearing were currently living with the CP and the child for whom the order had been issued. This illustrates the dynamism in the family situations of NCPs and CPs, as would be the case in a sample not restricted to low-income households. In these cases, the NCPs should have their order adjusted to reflect these newly discovered living circumstances.

Ineligible for PFS under local program rules. About 15–20 percent of NCPs who appeared at reviews fell outside specific local PFS rules. Especially important were local rules on the required link to a public-assistance case; during part of the demonstration, some sites did not serve NCPs in PFS if there was not a current AFDC case for the CP. Other sites required NCPs to be older than a specified age cutoff, living within reasonable commuting distance of the program service providers, legally able to work in the United States (that is, was a citizen or had the proper immigration status), or able to converse in English to be included in the program.

V. Referring Eligible NCPs for PFS

This section covers three topics: the issues encountered in defining PFS eligibility, the process used to refer NCPs to the program, and the procedures for making downward modifications of child support orders for PFS participants. All of these topics, while seemingly fairly technical and detail-oriented, involve program planners in important policy choices central to the case-sorting problem outlined earlier in this report.

A. Defining PFS Eligibility

The basic eligibility framework for the PFS Demonstration was set by the Family Support Act, which authorized a test of employment services for the NCPs of children receiving AFDC who were unable to pay child support because of unemployment. But as MDRC and sites grappled with the details of program design, questions arose — some small and some posing larger issues. A few were decided based on the logistics of program operation: for example, sites uniformly imposed a rule that NCPs had to live close enough to program services to participate. For most sites, this translated into a requirement that NCPs live within the county, though in a few cases NCPs with nearby addresses just over county borders were allowed into the demonstration.

Exhibit 9 presents some eligibility criteria that posed larger issues. The first two involve choices about how early in the NCPs’ interaction with the CSE system the PFS intervention is available. The second two arose because of a desire to recognize the economic instability in the lives of CPs and NCPs living below or near poverty.

The first row of Exhibit 9 shows how nonpayment of support was defined. Site practices reflected the belief that a major response such as PFS was best targeted at NCPs with more serious and longer-lasting nonpayment problems. As such, sites either developed specific definitions for the demonstration (for example, no payments for a set period, usually a month or more) or relied on the working rules-of-thumb that frontline CSE staff use as they review their caseloads and consider whether to schedule a contempt hearing.

One of the more complex issues was how to handle newly established paternities where a support order is about to be set. Even if the NCPs have serious employment problems, there is not yet a "nonpayment problem." The second row of Exhibit 9 shows how sites resolved this eligibility issue. Some sites (Hampden and Kent counties) chose to make this group a priority. Other sites found it difficult to introduce the PFS Demonstration research procedures into the paternity and order establishment process and thus excluded these cases. But even in these sites, staff recognized the advantage of helping people before large arrearages mounted.

Exhibit 9

Defining PFS Eligibility

Treatment of Key Issues	Duval County, FL (Jacksonville)	Hampden County, MA (Springfield)	Kent County, MI (Grand Rapids)	Los Angeles County, CA (Los Angeles)	Mercer County, NJ (Trenton)	Montgomery County, OH (Dayton)	Shelby County, TN (Memphis)
Nonpayment	No special definition for PFS; identified potential referrals off regular court docket	Most referrals from newly established paternities where NCP listed no or limited income on his financial statement	NCP has not paid support within last 4 weeks, or has paid less than 75 percent of what was owed within last 60 days	NCP has not paid support in 30 to 60 days	Standard definition used by Probation Department enforcement staff	NCP has not paid child support in at least 5 weeks	NCP has not paid in past five months
Newly established paternity and NCP reports being unable to pay support	Eligible	Eligible	Eligible under special exemption from nonpayment rule if NCP acknowledges paternity at hospital and special expedited procedures are followed	Eligible	Eligible	Not eligible	Initially not eligible, but restriction eliminated near end of sample intake
Underemployed	Eligible if earning less than $5.30 per hour for 40-hour week	Eligible if earning $200 or less per week or receiving unemployment benefits	Eligible if working at a job that does not provide enough income to meet his obligations	Eligible if working less than 30 hours per week	Eligible if working at less than $7 per hour and less than 20 hours per week	Eligible if working at a job that does not pay enough to support himself as well as required child support or if job is temporary	Eligible if working at temporary job, job paying less than $5 per hour or working less than 40 hours per week
CP not currently receiving AFDC, but NCP owes arrearages to state and continuing child support obligation	Eligible	Eligible	Eligible	Initially not eligible, but starting in early 1996 was eligible	Eligible	Eligible if CP in one case linked to NCP is receiving AFDC or NCP owes AFDC arrears of more than $150	Initially, CP in all NCP child support cases had to be receiving AFDC or have AFDC arrearage, but this restriction was later eliminated
Special site exclusions	None	None	None	Originally, NCPs had to live in areas within county targeted for aid following 1992 civil unrest; later expanded to entire county; NCP must be literate, legally able to work in the United States, and (early in program) able to speak English	None	Must be age 18 or over	Initially, NCP had to be age 45 or under, but restriction was eliminated near end of sample intake

Two other issues reflected the instability of NCPs’ and CPs’ lives, especially movements by NCPs into and out of jobs and by CPs onto and off welfare. As to job instability, low-income NCPs often moved back and forth between unemployment and low-wage/low-hour jobs, which — though technically not unemployment — clearly reflected underemployment. Sites also recognized that PFS could be an appropriate option for NCPs who were working at a short-term, temporary job at the time of the review of child support status. Thus, sites gradually moved from a strict eligibility requirement of unemployment toward a definition of underemployment that reflected the realities of their labor market, as shown in the third row of Exhibit 9. Some imposed maximum wage and/or hour cutoffs below which an NCP could be referred to the program as underemployed.

Other sites decided that NCPs could be referred to the program if their earnings were not sufficient to allow them to support themselves and pay the required child support or did not allow them to meet their "obligations." This approach might seem circular since child support obligations are — in theory — to be set as a percentage of income using the guidelines, but in practice the approach can make sense for several reasons. Child support obligations are not adjusted downward in most jurisdictions unless NCPs petition for such a change in a proceeding separate from a review for nonpayment, and the process is often complicated and time-consuming, leading few NCPs to go this route. There are often rules allowing or requiring courts to impute earnings when NCPs with a work history are unemployed. Further, most jurisdictions require that a change in NCP circumstances be substantial and in some sense permanent, rules that lessen the chance downward modifications will be made. Finally, as mentioned earlier in this report, several versions of state guidelines for child support can leave NCPs with very low incomes and insufficient remaining income to cover the necessities of life.

The frequency of movements of CPs onto and off public assistance also complicated the definition of PFS eligibility. Most of those working in child support are conscious of movement of some CSE cases from public-assistance to non-public-assistance status over time. The neat distinction drawn by PFS’s authorizing legislation (cases in which the children and CP are receiving AFDC at some point in time) does not reflect the reality of the world of either public assistance or child support. Thus, the PFS eligibility rules (shown in the fourth row of Exhibit 9) were expanded to include NCPs in child support cases in which the CPs were not receiving AFDC at the time of the PFS referral but (1) there were minor children for whom current support was owed and (2) arrearages owed the state had built up because of past nonpayment of support while the CPs received welfare and/or medical assistance.

B. Referring NCPs to the Program

During the demonstration, all sites developed a procedure to make the mandate to participate in PFS part of a court order. Thus, NCPs who did not meet program participation requirements could be held in contempt of court or the local equivalent. Even in Los Angeles County, where NCPs who appeared at the review hearing were asked to agree to participate (that is, sign a stipulation that they would participate), once this agreement was made participation became mandatory.

The sites differed in whether the court issuing the order retained jurisdiction of the case after it ordered participation in PFS and/or scheduled any future review hearings at the time the referral was made. If the court did not retain jurisdiction or did not set a follow-up review hearing, NCPs who later failed to meet program requirements once again had to be located and served legal notice of a new hearing. If at referral to PFS the court retained jurisdiction or set a follow-up review hearing, then such legal notice was not required prior to the court taking action to enforce its order. The variety of site practices suggests that the resolution of this issue was heavily influenced by state and local statutes on civil law procedures and customary practices. But those sites which found a way to avoid a second "locate and serve" effort clearly had a more effective way to enforce the participation mandate.

C. Making Downward Modifications in the Child Support Obligation

One of the most unusual elements of PFS is the downward modification of the support obligation while NCPs participated in the program. PFS asked for a commitment of time from the NCPs to participate in peer support and employment and training services; in exchange for this investment in the program, local staff could reduce the child support obligation to a minimum or even to zero. This approach differed from standard practice in CSE in two ways: normally, earnings could be imputed by courts when NCPs were unemployed and rarely would orders be modified without a formal request by an NCP and consideration of the issue in a special hearing on the motion to modify. (Importantly, modification normally could not be considered in contempt hearings.) Because this PFS change was such a break with normal practice, it was not uniformly implemented in all sites throughout the entire period of the demonstration.

Exhibit 10 describes the standard practices for modifying child support orders in each site, circumstances in which earnings are imputed, and special procedures during PFS. As the first column shows, in most of the sites modifications were usually made at the request of a parent. Periodic reviews of all cases are not yet a part of the administrative routine in most of the sites. Generally, there is a requirement that the change in circumstances be substantial (if defined, typically a change on the order of at least 10–15 percent), not temporary, and not voluntary (for example, quitting a job). In general, the modification requires some type of special hearing before a judge or hearing officer and cannot be considered as part of a contempt hearing. One reason for this separation of modifications and contempt actions is that since the notice of the contempt hearing does not alert the CPs that a change in the order is under consideration they might not attend or be prepared to address the issue.

Exhibit 10

Making Downward Adjustments in Child Support Orders for PFS Participants

Site	Usual Practice for Modifying Orders	Usual Practice for Imputing Income if NCP Unemployed	Special PFS Procedures
Duval County, FL (Jacksonville)	Modifications had to be made in hearing called for this purpose and could not be made as part of a contempt hearing. Modifications were made when in best interests of child (as defined in regulations) or when applications of guidelines would produce a change of at least 15 percent or $50, whichever was greater. Most modifications were made at a parent’s request.	Court imputed income if unemployment was voluntary and factors considered were physical and mental capacity to work and other circumstances over which NCP had no control. Court determined employment potential and probable earnings by reviewing work history, occupational qualifications, and prevailing earnings in community.	In practice, orders not consistently modified downward for PFS. Under the PFS plan, if NCP was unemployed and CP was currently receiving AFDC, NCP signed a consent agreement suspending child support obligation, and judge signed the consent order. If NCP was underemployed, his obligation was computed based on actual income rather than what would normally be imputed. When child support order was suspended, prior court order setting child support obligation still existed, but NCP did not have to pay on order and arrears did not accumulate. Court retained jurisdiction over the case and could reinstate old order if NCP did not follow through and participate in PFS.
Hampden County, MA (Springfield)	Modifications were made at request of a parent who had to file seeking order modification with court. Sometimes multiple modifications were scheduled in a block of time when there was room on docket. In a modification, if CP did not appear some judges would not go forward.	State guidelines called for using potential earnings if parent was earning substantially less than he could through reasonable effort. County practice was to consider what NCP was currently doing, and how and when he left his last job. If he was fired or voluntarily left last job, but not if laid off, income would be imputed.	Most NCPs referred to PFS involved new paternity cases and initial child support order was set at $50 per month. When a referral arose out of a modification request, order was reduced to amount NCP had requested or to an amount judged appropriate under guidelines by court. When an NCP was referred to PFS from a complaint hearing, judges were encouraged to reduce order to $50 per month, but this was not always done.
Kent County, MI (Grand Rapids)	Under Michigan law, Friend of Court (FOC) reviewed and recommended modifications (1) upon request of either party not more than every two years in non-AFDC cases, (2) every two years in AFDC cases, and (3) whenever it was in best interests of child. If one party wanted a modification more often than every two years, he or she could ask court but there was no recommendation from FOC.	Court could impute income when parent voluntarily reduced income. In deciding whether to recommend imputation, FOC considered prior work experience, education, disabilities, presence of children in home and its impact on ability to earn, availability of employment, prevailing wages, special skills and training, and other evidence parent could earn imputed income.	For NCPs in PFS, court suspended enforcement of child support order and any payments on arrearages of state debt, rather than adjusting them downward. Technically, arrearages built during participation, but if NCP complied with program requirements these added arrearages were canceled. Orders in PFS cases were reviewed on request by either parent or PFS staff regardless of usual law on frequency of modifications.
Los Angeles County, CA (Los Angeles)	Court action was required to modify a child support order. For upward modifications, the district attorney issued an order to show cause to NCP for a hearing.	District attorney sought to impute earnings based on working 40 hours per week at minimum wage for unemployed NCPs, unless NCP was on disability or public assistance. Judges did not always follow these recommendations.	Stipulation an NCP signed on entering the program states that “zero amount” child support orders would remain in place during program participation and other orders will be reduced to $50 per month.
Mercer County, NJ (Trenton)	Either the agency or the NCP could file a motion for modification. If arrearages existed, agency also would file a motion for enforcement with a statement of arrearages. Parties appeared before a hearing officer and NCP ordinarily had to show long-term, involuntary change in circumstances to get a reduction. If either party objected to hearing officer's recommendation, case was referred to a judge for a same-day hearing.	State guidelines allow for imputation of earnings for unemployed NCPs, based on past work history and earnings and area wage surveys. In case of NCPs with low skills or little work history, courts can impute earnings based on full-time work at the minimum wage. In some cases, courts also impute income based on in-kind contributions NCP receives, such as free room and board.	The order requiring participation in PFS also reduced NCP’s child support obligation to a minimal amount ($15 per week for one child, $5 per week for each additional child) and payments on arrearages were also reduced (to $5 per week). In practice, payments on state arrearages were not enforced while someone participated in PFS. If NCP failed to participate, order stated that original (higher) order would be reinstated retroactively.
Montgomery County, OH (Dayton)	There was an administrative process to review requests for modifications when one of parents requested a review or when required 3-year review period was reached. Agency made a recommendation to court for its approval or rejection. Only changes in circumstances producing at least a 10 percent change in order were considered substantial enough to require a modification.	Past practice was automatic imputation of earnings of unemployed parents for whom court had no financial information. New state legislation shifted to case-by-case imputation, based on ability to earn, availability of employment appropriate for parent in community, and unemployment rate. But Montgomery County still imputed income (usually based on work at minimum wage) and set $50 per month minimum order unless parent was incarcerated or receiving SSI.	In Domestic Relations Court, order was temporarily reduced to $50 per month for 6 months, during period of PFS participation, or until NCP found a job. In Juvenile Court, prior order was suspended for a similar length of time, meaning that there was no current obligation to pay and arrearages did not build. In practice, these changes in order were not always made.
Shelby County, TN (Memphis)	If there was a change in circumstance that would produce at least a 15 percent change in order, case could be placed on court docket at request of a parent. A prehearing conference was held and agency staff encouraged parents to reach an agreement that could be put in form of a consent decree. If an agreement was reached, judge signed new order and hearing was not held.	Income was imputed on a case-by-case basis after questioning parent about prior employment and current job search efforts, and making a determination if he had ability to pay. Often, standard practice was to impute earnings based on full-time, minimum wage work. Sometimes court would consider education and impute higher earnings.	Family Court reduced child support order to $50 per month and then credited an NCP $50 per month for participating in PFS program. This approach, which differed from usual rule against in-kind payments, had been approved by HHS on condition court made a finding that applying guidelines would be inappropriate and a provision for credit toward support obligation in exchange for participation in PFS was included in order requiring participation in program. Orders did not consistently include this language.

In all of the PFS sites a procedure to impute earnings when NCPs were unemployed (shown in the second column of Exhibit 10) was routinely performed if the court found the unemployment "voluntary." Factors typically considered include the employability of the parent, the availability of work, and current job search efforts. One jurisdiction weighs heavily the reason for the loss of a prior job and does not impute income in the case of layoffs or plant shutdowns. Generally, imputed earnings are at least those produced by full-time work at the minimum wage.

PFS sites were to make changes in these modification and imputation practices by not imputing earnings, reducing the order, and handling all of these issues in a single review. The final column of Exhibit 10 shows special PFS procedures in each site. Generally, site staff developed a standard court order directing NCPs to participate in PFS and reducing their orders. Sites made the order reduction in one of two ways: (1) four sites made an actual reduction in the amount of the support obligation, issuing a new order that replaced the prior support order; (2) three sites suspended the current order while the parent participated in PFS (the existing order remained in place, but the obligation was not enforced and arrearages did not build).

It appears that the choice between these two methods was largely determined by local court practice and civil court procedures. Sites typically developed a plan to accomplish the reduction in the obligation in a way that required the smallest departure from past practice.

One important program implementation issue was embedded in the choice, however. Under either method, it was important for a site to develop a way to quickly put in place a normal support obligation if NCPs did not follow through and participate in PFS. In two sites where orders were suspended (Duval and Kent counties), the order to participate in PFS explicitly stated that the court retained jurisdiction and the old obligation would automatically be put in place if the NCPs did not comply with program requirements. In one site in which the order was actually reduced (Mercer County), the new order stated that the original (higher) order would be reinstated retroactively if the NCPs did not comply with program requirements. Either of these approaches provided sites a way to avoid lengthy procedures to reimpose an obligation on NCPs who fail to meet program requirements.

NCPs with multiple CSE cases posed special problems for sites seeking to reduce the support obligation during PFS. When one of an NCP’s CSE cases was not a "public-assistance-related" case, the support obligation could not be reduced without the agreement of the CP. No site attempted to put in place procedures to secure this agreement. However, complications could also arise when an NCP had multiple public-assistance-related cases unless all these cases were "consolidated" into a single hearing and the obligation reduced for each case. Further, if the site chose to reduce the obligation to some minimum amount (say $50 per month), NCPs with several cases could still find themselves facing an obligation of $100 per month or more. Thus, it is important to consider an NCP’s total support obligation and work to develop an adjustment that is as comprehensive as possible.

VI. Revising Local Enforcement Procedures to Make PFS Work Better: A Case Study of Montgomery County

When Montgomery County began the PFS Demonstration, local staff anticipated that many NCPs would be appropriate for the program and could be referred to its services. Early estimates of the pool of potential referrals exceeded 6,000 cases. However, as implementation of PFS began, it became apparent that there were serious problems with early projections. Efforts to draw random samples of possible referrals to PFS from the CSE caseload quickly exhausted the pool of identified cases and few NCPs were reaching the hearings at which their appropriateness for PFS could be assessed. In the ensuing two years of the demonstration, local staff found new ways to identify more NCPs who potentially could benefit from PFS and shifted to a different type of hearing that lessened delays at this stage. This brief case study of the experience in Montgomery County first presents initial plans for identifying NCPs appropriate for PFS and then describes how local staff responded to the slow pace of program participant buildup.

A. Initial Plans for Identifying NCPs

During the planning stage of the demonstration, local staff estimated that approximately 10,000 NCPs appeared on default lists (NCPs defaulting on their child support obligation) over the course of a year and about 6,000 of these were potential referrals to PFS. (These 6,000 cases involved CPs who were receiving or previously had received public assistance.) Thus, local staff expected that it would be possible to randomly select about 250 NCPs each month and make special efforts to review these cases and refer appropriate parents to PFS.

As the demonstration began, this plan quickly proved infeasible. The pool of potential referrals produced by existing practices turned out to be approximately 2,000 NCPs, many fewer than anticipated. Further, the initial number of referrals to the program was much less than even this smaller-than-expected pool should have produced. Three months into the planned period of sample intake and program operation no NCPs had been referred to the program. Local PFS programs, staffed in anticipation of large numbers of program referrals, were left waiting for participants.

This shortfall occurred for three main reasons, which site staff worked to address over time.

1. Existing enforcement practices excluded many NCPs from the pool of cases from which referrals were identified.
2. The process of getting cases on the court docket for a hearing was complex and long lags occurred when more cases were added to the flow.
3. A low percentage of NCPs called for a child support hearing attended the review or otherwise contacted the agency.

B. Redefining the Pool of Potential PFS Referrals

Before the PFS Demonstration, when local staff identified an NCP on the default list with a residential address but no address for employment or other evidence of income they referred the case to the Parent Locator service (PLS), which served as the primary enforcement response for these cases. Staff would scan various databases to locate a more recent residential address or information on employment. Further, credit bureaus would be informed that the NCP had an outstanding debt. The line enforcement staff would wait for the PLS to provide further information on the case before taking any additional action.

With PFS, there was a new programmatic option for these cases. As staff sought to understand why there was such a dramatic shortfall in NCPs referred to PFS, they realized that referrals to the PLS accounted for the gap between the initial estimate of the pool of NCPs (6,000) and the 2,000 identified as potential referrals at the beginning of the demonstration. Starting approximately four months into sample intake, local staff returned to the normal CSE caseload approximately 4,800 NCPs whose cases had previously been assigned to the PLS because they lacked information on employment. As enforcement staff worked these cases to determine whether they were appropriate for PFS, many proved to have usable residential addresses and substantial numbers provided previously unreported information on employment or were referred to the PFS program.

Other refinements made in the eligibility rules did shrink the pool of potential referrals somewhat. Originally, nonpayment of support was defined as not having paid the full amount owed during the last six weeks. Soon after sample intake began, this was changed to not having paid at least 75 percent of what was owed on at least one case, and further changed over time to not having paid at least 75 percent on all cases. These restrictions on eligibility were made to avoid including NCPs with multiple cases who were meeting their obligations on some of their cases and falling behind on some. In these circumstances, local site staff thought PFS was not the best enforcement response.

C. Speeding Up the Hearing Process

Under the initial plan for the demonstration, local staff instituted a contempt of court hearing for NCPs identified as potential PFS referrals. In April 1994, MDRC and site staff began to draw samples of potential PFS referrals from the default lists (NCPs who were not paying and were linked to an AFDC case) and by July over 600 had been identified. However, through the end of July none of these NCPs had reached the stage of a contempt hearing and the courts had made no referrals to the PFS program. Local staff identified two sources of this problem: delays in relaying information on the large number of NCPs to be called for hearings to the prosecuting attorney’s office (which put cases on the court docket) and the three-month minimum time needed to schedule and hold a contempt hearing.

Responding to the first source of delay, local CSE staff developed a way to make mass referrals of scores of NCPs to the prosecuting attorney through an automated data file rather than in the traditional way of individual referrals on paper. This plan was complicated by the different computer systems used by the CSE and prosecuting attorney’s offices, but after several weeks of work it was put in place and lessened this source of delay.

The process of holding contempt hearings was lengthy for several reasons. First, before CSE staff could refer a case to the prosecuting attorney they had to have a postmaster’s verification within the last 30 days that an NCP was receiving mail at a specific address so there was a likelihood he could be personally served. Once a case was referred for a hearing, it was assigned to an attorney who scheduled an office appointment for the NCP to appear and explain why he was not making payments. (For example, the NCP might have reported that he was receiving SSI or incarcerated, which would stop the process of scheduling a hearing.) However, only about 15 percent of NCPs appeared for these hearings.

If the NCP did not appear for the office appointment or failed to provide a satisfactory explanation for nonpayment, the attorney proceeded to schedule a hearing and then schedule an appointment for the CP to come in and sign a formal complaint. In public-assistance cases, even if the CP did not appear the case could proceed without her signature on the complaint because she had assigned her rights for support to the state as a condition of AFDC receipt.

Next, unless the NCP had signed a waiver of notice, he had to be personally served and proof of service provided the court. In practice, the NCP was normally given four-to-six-weeks’ notice (usually served by the sheriff’s office). Law enforcement officials would appear at the address listed on the notice and try to serve the NCP. If he was not at that address, they did not have the resources to follow up on leads, even if there was information on his whereabouts. The notice ordered the NCP to appear on a specified date before a magistrate of the Juvenile or Domestic Relations Court.

At the contempt hearing, poor NCPs often requested appointed counsel, which could result in a 30-day adjournment of the case. Finally, contempt hearings were relatively time-consuming, with courts typically able to hear approximately two cases per hour.

This protracted process for contempt hearings led the local staff to shift to a different type of hearing three months into PFS sample intake: a review hearing for the purpose of determining eligibility for PFS. Since the magistrates presiding over this type of hearing did not have the authority to find NCPs in contempt of court, the notice requirements were substantially less — only seven days and personal service was not required. Notice was typically mailed to NCPs 10 to 15 days before the hearing. Staff also made other changes in standard procedures to shorten delays in holding a hearing. These included consolidating cases under a single attorney, eliminating the prehearing office conference, and dropping the practice (not legally required in public-assistance cases) of calling the CP into the prosecuting attorney’s office to sign a complaint.

Although these changes did speed up the process of holding hearings, much unavoidable work remained, indicated by the fact that 30 different forms and documents were required to get a case on the docket and before a magistrate. Further, the shift to a less formal hearing had implications for the percentage of NCPs who attended, as discussed below.

D. Increasing the Appearance Rate at Reviews

Local staff report that approximately 50 percent of NCPs ordered to appear for a contempt hearing typically appear. After the shift to a less formal review hearing, the percentage of potential PFS referrals appearing as ordered dropped substantially below this rate, in the range of 30 to 40 percent depending on the month. The drop occurred because of the shorter and less certain method of serving notice, the "less serious-sounding" purpose of the hearing, and the initial practice of dismissing the hearings of NCPs who did not appear at the review. The dismissal occurred because of the presumption of local staff that NCPs who did not appear at the review had not received notice of the hearing, which implied that the address in CSE records was incorrect and the case should be dismissed and referred to the Parent Locator Service to seek a current address. Over the course of the demonstration, local staff found ways to address these problems.

Starting in mid-1995, staff made a major break with the national CSE tradition of office-bound enforcement and instituted home visits prior to scheduled hearings. The change provided a way to address all of the problems cited above that were producing a low appearance rate at the hearings; over time they produced an appearance rate of approximately 70 percent. This was substantially higher than during the initial period when contempt hearings were used, even though in those hearings NCPs could face sanctions for nonappearance.

In August 1995, CSE managers agreed to let local PFS staff follow up on the mailed notice of the review with an in-person visit to NCPs at their reported address to remind them of the hearing. CSE staff provided PFS staff with docket lists of NCPs approximately one week in advance of the scheduled hearings. Home visits were made toward the end of the week prior to the reviews, which would provide enough time for the reminder to get to the NCPs through friends and family if they could not be personally located, but not so long that it would be forgotten in the press of other events.

Because there was some initial concern about the safety of the staff making these home visits, this duty was assigned to males who were experienced in making home visits for other social service programs; they were often interracial teams, and they carried a cellular phone. Before beginning the home visits, staff would call each CP to ask her if she had information on the residence or employment of the NCP. This produced more current or more accurate information on his location and activities than was previously available to the agency. On days when home visits were scheduled, two PFS staff persons would collect a county car at approximately 8:30 a.m. and begin a series of visits. In early 1996, for example, approximately 20 to 25 NCPs were scheduled for each review docket, a number that usually could be visited in a day. Staff making the visits observed that the reported residences of NCPs tended to cluster in a few neighborhoods within downtown Dayton.

The goal of the home visit was to leave a written reminder of the hearing with the NCP or a person who agreed to give the notice to the NCP, or — if this was not possible — in the door of a residence where it was likely the NCP lived. Unlike sheriffs doing personal service for contempt hearings, staff making the home visits would not stop their search if an initial address turned out to be invalid. They sought information on the whereabouts of the NCP from residents at the initial address, neighbors, and rental property owners or real estate agents. This questioning might turn up a lead on a new residence, information on the NCP’s place of employment, or evidence that the NCP was in jail, in the hospital, in a treatment facility, or even deceased.

Exhibit 11 shows what happened on home visits during the first six months of 1996. Staff were able to find a place to leave the reminder notice in 80 percent of the cases. In 7 percent of the cases, staff were able to leave the notice with the NCP himself and personally remind him of the hearing and its importance. In another 31 percent of the cases, staff left the notice with a person who agreed to give the notice to the NCP. In 24 percent of the cases, this person reported that the NCP lived at the same address while in 7 percent of the cases the NCP lived elsewhere. In 42 percent of the cases, staff were not able to leave the notice with a person but did leave it at the door.

Exhibit 11
Results of Home Visits in Montgomery County
(January to June 1996)
Result of Visit			Percentage or Number
Notice left at a residence (%)			80
	Left with NCP		7
Notice left with family member or friend who reports that NCP (%)
	Lives at the address		24
	Lives elsewhere		7
Notice left at door of residence (%)			42
Notice not left (%)			20
	NCP moved and address unknown		5
	NCP in jail or prison		2
	Other reason could not be delivered		13
Total number of NCPs on the hearing dockets			351

SOURCE: MDRC calculations from logs maintained by site staff.

Usually this occurred after verifying that the housing unit was occupied and obtaining corroborating evidence (from a name on the mailbox, neighbors, or some other source) that the NCP actually lived there.

Two other statistics support the importance of home visits. Most strikingly, the percentage of NCPs appearing at the review hearings rose from 41 percent in the first six months of 1995 (before home visits) to 69 percent in the first six months of 1996 (after home visits) and the break in the trend was immediate and sustained. Although the causal link is less clear, 73 percent of those NCPs for whom a notice could be left appeared at their hearing, as opposed to only 19 percent of those for whom a notice could not be left.

Beyond these numbers, the staff experience suggests that the home visits had a major effect. When local staff left the notice with a household member, it was most commonly the mother of the NCP and she often expressed an interest in helping her son resolve his problems with child support and promised to urge him to attend. Even in cases in which the notice was left at the door, messages from friends and neighbors could be an important impetus to attend. One NCP told PFS staff he came to the hearing because his neighbors told him "two guys were asking for him" in a serious way. Obviously, the impact of a personal visit ("we know where you live") can be substantial.

The home visit approach also allowed local staff to make other changes that reinforced the importance of attending the hearing. Prior to the home visits, the hearings for those NCPs who did not appear were dismissed on the presumption that service never occurred because the addresses were inaccurate. With the home visits, CSE and PFS staff could often bring current information on residence and evidence the NCP did know of the hearing before the magistrate handling the review.

Typically, the magistrate first conducts reviews for all NCPs who appear as ordered. Following this, the magistrate asks CSE and PFS staff whether they have evidence that NCPs who did not appear knew of the hearing and whether they wish to recommend referring their case to the prosecuting attorney for a contempt hearing. In cases in which staff making the home visit gave the notice directly to the NCPs or a close relative or friend who agreed to give the notice to the NCPs, the magistrates generally accept the staff’s recommendation that a contempt proceeding be instituted. In cases in which the notice was left at the door, the staff and the magistrate assess the strength of the corroborating evidence that the NCPs actually live at the address where the notice was left and reach a decision on whether to refer their case to the prosecuting attorney for a contempt hearing.

With this change in the way the CSE system responded to nonappearance, NCPs who had notice of the hearing and chose not to appear ran a much higher risk of sanctions, including being found in contempt of court and facing a possible jail sentence. As word of this change filtered into the community, NCPs who received notice of a hearing were more likely to take it seriously and appear and those who did not were more likely to be pursued further.

5. Summary of Lessons and Policy Implications

Many recent reforms of the child support system are based on the assumption that nonpayment is primarily an enforcement problem that can be addressed by enhancing the IV-D system’s ability to track down NCPs and to compel payment from those unwilling to pay voluntarily. And, as has been noted in this report, most CSE reforms are geared toward NCPs whose employment, financial, and residential circumstances are relatively stable.

The PFS experience suggests that among the low-income NCP population, nonpayment of child support is a much more complex problem. For some poor NCPs, the problem is clearly lack of a commitment to pay, at least through the formal child support system in which much — or, in the future, conceivably all — of the support payment for CPs and children receiving welfare never reaches the family. But other poor NCPs lack the financial wherewithal to pay, and still others — because of frequent changes in their lives — no longer should face a current support obligation. And the lives of many low-income noncustodial parents do not fit a middle-class profile, so many standard location and enforcement measures are of limited use and courts are often hard-pressed to determine the "real" circumstances of the NCPs’ lives.

PFS was developed as a response to this complexity, an attempt to fashion an alternative to a policy approach that — while appropriate in many families — seems like an overly standardized, "one-size-fits-all" strategy when applied to families receiving welfare. In doing this, it is clear that PFS goes against the trend in child support toward more standardization, more use of administrative records, more intensive efforts to impose sanctions for nonpayment, and efforts to lower the cost of enforcement. One of the crucial questions in PFS is whether the initial steps in the program to identify and refer appropriate NCPs to services are so difficult, time-consuming, and therefore expensive and the initial payoff is so small that the effort is not worthwhile from the very inception.

This report has described this initial stage in the program, showing how PFS sites developed or adapted administrative processes to serve as agents of the courts and CSE system. They worked to better tailor the enforcement response to the circumstances of the NCPs. The experience of the PFS sites so far suggests that they have succeeded in putting in place new procedures that can dramatically increase the accuracy of the sorting process inherent in CSE: the need to distinguish the unwilling from the unable. By increasing the monitoring of the status of nonpaying NCPs, PFS appears to be smoking out considerable previously unreported employment and helping local agencies to identify cases in which the circumstances of the NCPs have changed so substantially that they should no longer face a current obligation (though arrearages would need to be repaid). By having a new enforcement option (the PFS services) to which apparently unemployed or underemployed NCPs can be referred for mandatory participation, the courts and CSE agencies have a further method of sorting, which poses little risk. If an NCP is working and hiding that fact, the PFS participation mandate can force him to reveal it. If he is in need of services, they are available to him.

The final story on the accuracy of this sorting process, its costs relative to benefits, and the resulting impact on support payments and other aspects of NCPs’ involvement with their children is not yet known; it awaits future PFS reports estimating program impacts, costs, and benefits based on the random assignment, experimental research design put in place during the demonstration. But early indications at the sites are that the smokeout of previously unreported income has been substantial and produced child support payments, and the availability of PFS shows promise in helping poor, unemployed, or underemployed NCPs to participate more fully and effectively in the support and parenting of their children. As would be expected in an endeavor this difficult, the implementation of the program has had its problems and the program is clearly not able to help all NCPs. But this early experience suggests that PFS is in fact worth the effort to test rigorously.

Beyond these overall policy conclusions, this report suggests more concrete administrative lessons, primarily approaches that sites developed to implement PFS but that could offer insights for CSE agencies in general. These include identifying potential referrals to PFS, locating these NCPs, and gaining their cooperation in the review and referral process, and on ways that the PFS option allowed local CSE programs to tighten up the overall administration of their programs.

I. Identifying Potential Referrals to PFS

From the initial stages of the demonstration, sites participating in PFS recognized that change and instability were inherent in the lives of low-income custodial and noncustodial parents. This led to an expansion of the original definition of program eligibility to include NCPs who were underemployed and those whose affiliated CP was not currently receiving AFDC but had in the past. There was also a desire (not fulfilled in the demonstration) to expand eligibility beyond parents with a link to AFDC, to operate PFS as a preventive measure for families where the CP might otherwise need to rely on public assistance.

In seeking NCPs who were potential referrals to the program, sites also realized that the lack of a PFS-service option had led them to route some CSE cases to second-best options. The example from Montgomery County of cases routed to the PLS to determine employment was the most striking in this regard. With the new possibility of PFS, local CSE officials were able to identify many NCPs by retrieving cases from these alternates or reviewing the CSE caseload for nonpaying NCPs whose cases had received little prior enforcement priority.

II. Locating NCPs

While computerized information systems have undoubtedly improved the ability of IV-D agencies to locate absent parents overall, they are only as effective as the quality and timeliness of the information put into them. Therefore, they are most likely to help in locating NCPs who hold a regular job, reside in one place, and own assets. Information systems are less useful for locating low-income NCPs who are not connected to the mainstream economy or to government social programs.

Interviews with PFS participants and staff revealed that substantial numbers of these NCPs had sporadic work histories, punctuated by long periods of unemployment. Often when they did work, the PFS participants reported that they did so off the books. For these men, unemployment insurance records, IRS files, and credit bureau histories are of little use. The problem is exacerbated by the fact that information systems lag behind the individual’s current circumstances, and low-income NCPs tend to hold onto jobs for shorter periods of time than their middle-class counterparts.

Other databases typically relied upon by CSE agencies are designed to locate assets — department of motor vehicle records, for instance, and bank accounts. Interviews with PFS participants revealed that many of them did not have valid driver’s licenses, let alone cars, and that they did not have any credit or bank accounts. And low-income men, unlike their female counterparts, are less likely to participate in the welfare programs that bind recipients to the government.

The unstable living arrangements of many low-income NCPs also frustrate CSE agencies’ attempts to reach them. Addresses in the system are not always current, and often are those of relatives or friends with whom the NCPs periodically stay. Service of process rules in most of the PFS sites require that, at the very least, notices of impending hearings be sent by certified mail (although in some sites regular mail is sufficient); certainly, some of these notices never reached the intended recipients. On the other hand, relatively few letters that were sent were returned as undeliverable by the postal service — far fewer than the number of NCPs who did not show up. This suggests that many notices were delivered, but to relatives or friends who may or may not have encouraged the NCPs to comply.

The experience in Montgomery County, where CSE staff home visits led to an increase in the appearance rate at hearings, attests to the potential payoff to moving away from the usual practice of office-based investigation to a more community-based approach. During their outreach efforts, staff first contacted the CPs for leads and then had the staff follow up by visiting addresses and talking with other family members, neighbors, and landlords. This presence in the neighborhood led to new leads on addresses or employment; information on illness, death, or incarceration; and "filtering" of the word that NCPs should contact the CSE agency through informal networks of family and friends. As the sharp rise in appearances at hearings indicates, there is a payoff in increased success in locating NCPs.

III. Gaining NCPs’ Cooperation

While some portion of the low appearance rate at CSE hearings can be attributed to a lack of notice, an equal or possibly bigger issue appears to be noncooperation by NCPs. As revealed through interviews with PFS participants and staff, the failure of NCPs to respond may be due to a variety of attitudinal factors, including the perception that there is little risk that they will be caught, or, conversely, the fear of going to jail. Among PFS participants there is a pervasive sense that the child support system is unfair and insensitive to the plight of low-income fathers, more interested in wringing money out of them than in fostering stronger family ties. In the case of African-American men, this is compounded by a general sense of grievance against the criminal justice system, of which CSE is seen to be very much a part. Added to these attitudinal factors are the logistical problems caused by lack of money and transportation to get to the hearing site, a particular issue in geographically large jurisdictions with only a central office or courthouse. The PFS experience suggests that developing the right balance of sanction threat and message of opportunity, coupled with efforts to make appearance somewhat easier logistically, could increase cooperation.

A. Probability of Sanctions

While NCPs who fail to appear in response to a notice or fail to meet their existing child support obligations are theoretically subject to serious penalties, including the possibility of arrest and incarceration, few of the CSE agencies in PFS consistently applied these sanctions. This is probably due to several factors, including (1) the cumbersome legal procedures necessary to support such action — that is, official service of process, contempt hearing before a judge, and issuance of a bench warrant; (2) shortage of jail space for those not paying their child support; and (3) the fact that CSE agencies usually rely for arrest actions on other agencies such as sheriff’s offices, which often accord child support violations a low priority. Most sites in PFS also indicated that the majority of NCPs who are arrested on outstanding bench warrants for child support violations have been picked up by police on unrelated charges such as traffic violations. This feeds into the perception that the CSE system is in reality a part of the criminal justice system, further alienating some parents from cooperating and raising special issues for people of color.

B. Fine-Tuning the Message

In several sites, program staff tried to raise appearance rates by more fully explaining in the letters sent to NCPs that PFS offered services and not just punishment, but appearance rates were only slightly raised. In some sites, local staff discovered that more opportunity-oriented notices were viewed skeptically by NCPs as a possible sting operation, until word reached the street that PFS was in fact as described. Similarly, when sites toughened the letters to include threats of arrest and incarceration, appearance rates improved only slightly.

This suggests that the CSE agency’s ability, both actual and perceived, to effectively follow through with sanctions against noncompliant NCPs — and not the fine points of the message about the PFS opportunity initially presented in a hearing notice — is a critical factor influencing compliance. It is clear that many low-income NCPs express great fear of being jailed for child support violations, despite the resource constraints on enforcement faced by most CSE agencies. The perception of risk is not congruent with the reality. The perceived risk of incarceration also appeared to increase when home visits were a part of the notice process. NCPs sensed that the CSE agency might be very likely to be able to find them and impose a serious sanction for noncooperation.

The implications for enforcement may be that a few well-placed arrests will spread the word that noncompliance is likely to be sanctioned and thus stimulate greater cooperation, especially when the disposition of cases for unemployed NCPs is a referral to PFS. Further, offering more convenient ways to comply — by dispersing hearing sites throughout the jurisdiction or by offering a variety of ways for NCPs to contact the CSE agency prior to the hearing — may well serve to increase cooperation.

Issues of compliance and cooperation continue with the referral to PFS. Thus, the procedures that local PFS sites developed to reduce the current support obligation on referral to PFS and retain a means to reinstate a higher obligation if participation was not satisfactory were also important.

IV. The Effects on IV-D Agencies of Running PFS with an Enhanced CSE Component

Ordinarily, CSE staff assign a low priority to working those cases that require great effort and do not promise a big payoff. By participating in PFS, the sites essentially had to work a group of cases that would have otherwise not received much attention, a major commitment by agency staff already burdened with large caseloads. The experience with running this enhanced CSE suggests that simplifying procedures and using PFS program staff to help move cases through the system were critical to the agencies’ capacity to handle increased workloads. Furthermore, the intensive working of the low-income caseload paid dividends in that it enabled staff to identify more quickly than they otherwise might have those NCPs who were employed and those who were inappropriate targets for enforcement. Finally, the PFS program boosted CSE staff morale because it gave a sense of political legitimacy to efforts to enforce the child support obligations of low-income NCPs by offering an alternative to incarceration or ineffectual seek-work orders.

Enhanced CSE, because it brings cases into the system that would otherwise be accorded low priority, does make more work for CSE staff, but the PFS program was able to work with the CSE agencies and the courts to mitigate some of those burdens. This cooperation was probably crucial to the sites’ success in implementing enhanced CSE. For instance, in most sites the courts agreed to block scheduling and group hearings that minimized the delay of processing more cases. In some sites, the courts agreed to less formal notice procedures that allowed hearings to be scheduled more quickly and that did not require the courts to respond with formal sanctions against those who did not respond. And most courts developed standard forms for ordering NCPs into the PFS program to minimize delays associated with paperwork.

In many ways, PFS program staff acted as an extra arm of the court to make the hearing process more efficient — for example, making home visits prior to the hearing date in order to boost show-up rates and conducting orientations for NCPs before the formal hearings so that they were prepared for what would happen, thus allowing judges and hearing officers to spend less time making sure that each NCP understood the process. Also, PFS program staff monitored the NCPs’ compliance with the court orders, so that noncompliant obligors could be selectively followed up by the courts.

V. Conclusion

Recent reforms of the federal and state public-assistance and child support systems increase the need for creative and successful approaches to CSE. With the likely introduction of time limits in public assistance, low-income families will increasingly have to rely on parents’ income, and the child support system is under great pressure to improve its performance. Until recently, many IV-D systems spent little effort on cases involving low-income obligors, believing that they offered little prospect of financial return. Thus, meeting the challenge of the new reforms will require careful thought as to the best ways to allocate the already strained resources of state and local CSE agencies. The operational experience of PFS suggests that current enforcement remedies — based largely on a view of the world more appropriate for higher-income families — are inadequate to address the underlying problem of ensuring that children living in poverty get as much support as they can from both parents.

The early lessons from PFS provide insights into ways to improve CSE administration. Future research results from the demonstration will address the issue of the longer-term effectiveness of the program in increasing the employment, earnings, and child support of the NCPs and their ability to participate consistently and effectively

References

Bloom, Dan, and Kay Sherwood. 1994. Matching Opportunities to Obligations: Lessons for Child Support Reform from the Parents' Fair Share Pilot Phase. New York: MDRC.

Brien, Michael J., and Robert J. Willis. 1996. “Nonresident Fathers: Their Employment and Child Support Potential.” Unpublished paper.

Committee on Ways and Means, U.S. House of Representatives. 1994. Overview of Entitlement Programs: 1994 Green Book. Washington, D.C.: U.S. Government Printing Office.

Committee on Ways and Means, U.S. House of Representatives. 1996. 1996 Green Book. Washington, D.C.: U.S. Government Printing Office, U.S. Department of House and Human Services, Office of Child Support Enforcement.

Dodson, Diane, and Joan Entmacher. 1994. Report Card on State Child Support Guidelines. Washington, D.C.: Women's Legal Defense Fund.

Furstenberg, Frank F., Jr., Kay E. Sherwood, and Mercer L. Sullivan. 1992. Caring and Paying: What Fathers and Mothers Say About Child Support. New York: MDRC.

Garfinkel, Irwin. 1994. “The Child-Support Revolution.” American Economic Review 84 (May): 81–85.

Garfinkel, Irwin, and Patrick Wong. 1987. “Child Support and Public Policy.” Institute for Research on Poverty (IRP) Discussion Paper No. 854-87, prepared for DECD conference, Paris, December 1987.

Johnson, Earl S., with Anne Levine and Fred Doolittle. 1998. Participants in the Parents’ Fair Share Demonstration: A Qualitative Study. New York: MDRC.

Johnson, Earl S., and Fred Doolittle. 1996. Low-Income Parents and the Parents’ Fair Share Demonstration: An Early Qualitative Look at Low-Income Noncustodial Parents (NCPs) and How One Policy Initiative Has Attempted to Improve Their Ability to Pay Child Support. New York: MDRC.

Legler, Paul K. 1996. “The Coming Revolution in Child Support Policy: Implications of the 1996 Welfare Act.” Family Law Quarterly 30, no. 3 (Fall): 527–547.

Locker, Barbara B. 1976. “Enforcement of Child Support Obligations of Absent Parents — Social Services Amendments of 1974.” Southwestern Law Journal 30: 625–627.

Pirog-Good, Maureen A. 1993. Child Support Guidelines and the Economic Well-Being of Our Nation's Children. Madison, Wisc.: Institute for Research on Poverty.

Sorenson, Elaine. 1994. Noncustodial Fathers: Can They Afford to Pay More Child Support? Washington, D.C.: Urban Institute.

Turetsky, Vicki. 1996. Child Support Computers: A Summary of Current and Proposed Federal Requirements. Washington, D.C.: Center for Law and Social Policy.

Turetsky, Vicki. 1997. Child Support Administrative Processes: A Summary of Requirements in the Personal Responsibility and Work Opportunity Reconciliation Act. Washington, D.C.: Center for Law and Social Policy. January.

U.S. Bureau of the Census. 1991. “Child Support for Custodial Mothers and Fathers.” Current Population Reports, Series P60-187. Washington, D.C.: Department of Commerce.

U.S. Department of Health and Human Services, Administration for Children and Families. 1992. Characteristics and Financial Circumstances of AFDC Recipients. Washington, D.C.

U.S. General Accounting Office. 1991. Mother-Only Families. GAO/HRD-91-62. April. Washington, D.C.

U.S. General Accounting Office. 1992a. Interstate Child Support: Wage Withholding Not Fulfilling Expectations. GAO/HRD-92-65BR. Washington, D.C.

U.S. General Accounting Office. 1992b. Poverty Trends 1980–88: Changes in Family Composition and Income Sources Among the Poor. GAO/PEMD 92-34. September. Washington, D.C.

U.S. General Accounting Office. 1995. CSE: Opportunity to Reduce Federal and State Costs. GAO/T-HEHS-95-181. Washington, D.C.

Williams, Robert G., Mark A. Levy, J. Bruce Kilmer, David A. Price, and Victoria S. Schwartz. 1988. Literature Review and Analysis of Issues: Child Support Enforcement Amendments of 1984. Washington, D.C.: U.S. Department of Health and Human Services, Office of Child Support Enforcement.

Wilson, William. 1996. When Work Disappears: The World of the New Urban Poor. New York: Knopf. Random House, Inc.

Winston, Marion P., and Truce Forcher. 1971. “Nonsupport of Legitimate Children by Affluent Fathers as a Cause of Poverty and Welfare Dependence.” Santa Monica, Calif.: The Rand Corporation.