OCSE Home . Program Information . News . Publications . Policy . State Links . OCSE Search . Help

1999 REPORT TO CONGRESS:

Analysis of the Impact on Welfare Recidivism

of PRWORA Child Support Arrears Distribution Policy Changes

U.S. Department of Health and Human Services

Administration for Children and Families

Office of Child Support Enforcement

Acknowledgments

We take this opportunity to thank the Child Support Enforcement staff from the States that participated in this study analyzing the impact on welfare recidivism of PRWORA child support arrears distribution policy changes. Without their cooperation and assistance in providing both the data and access necessary to perform the computer simulations, this study would not have been possible. Their extensive knowledge and expertise in Child Support Enforcement provided valuable guidance and insight to the authors. In addition, a number of States that have conducted special studies of families leaving welfare have been very willing to share useful information.

This report provides the results of the study conducted by BETAH Associates with Joanne Barnhart as Principal Investigator and Deborah Chassman as Investigator and TRW Inc., to assist OCSE in responding to this statutory requirement.

Table of Contents

List of Tables

List of Figures

Executive Summary

Results in Brief

Have PRWORA Policy Changes Regarding the Distribution of Child Support Arrears Increased the Amount of Child Support Paid to Families That Formerly Received Public Assistance?

Overall, the financial impact of the PRWORA changes in distribution policy has been positive for former-assistance families. However, the magnitude of this impact is not consistent across all States nationwide.

Computer models employing historical financial data taken from State child-support computer systems reveal distribution outcomes that differ substantially based upon past State distribution policy. In States that, prior to PRWORA, sought to maximize State reimbursement for past assistance ahead of Custodial Parent-Owed Arrears, families will experience a 25 percent increase in child support income following PRWORA implementation. Further, for States that transitioned to PRWORA using a two-step approach, the bulk of this increase occurred during the first step that began on October 1, 1997. Conversely, in States that historically prioritized Custodial Parent-Owed Arrears first, family income initially will remain unchanged during the first step and then later decrease slightly under full PRWORA implementation. This later decrease occurs because a portion of the Custodial-Parent debt can be satisfied only after all State-Owed Arrears have been fully paid off. Overall, the financial gains for families outweighed the losses.

Does This Increase in Child-Support Income Reduce the Likelihood of Welfare Reentry?

Based on research findings, even small amounts of child-support payments reduce welfare recidivism. Thus, insofar as the new distribution policies provide higher child-support distributions to families, the changes will prevent some welfare recidivism. Further, the importance of child support for families who had exited welfare was reinforced anecdotally by interviews where mothers indicated they depended on child-support payments to make ends meet. This also suggests that distribution policies that increase payments to post-welfare families will contribute to a lower rate of welfare recidivism.

As the residual Temporary Assistance for Needy Families (TANF) caseload decreases over time, it increasingly comprises hard-to-place individuals who face substantial barriers to employment. Based on the anticipated low earning power of these low-skilled women, the relative value of even incremental increases in child support will be greater for the hard to place. Additionally, if these individuals reside in a low-benefit State, the relative replacement value for the TANF grant will be greater. This combination of factors suggests that PRWORA distribution policies that increase child-support payments to these families may have an even greater effect on welfare exits.

Does Increasing the Amount of Child Support That Is Passed Through to Families on Public Assistance and Disregarded for the Purposes of TANF Benefits Also Result in Increased Child-Support Collections?

Custodial-Parent surveys indicate that as long as a family remains on welfare and welfare benefits are reduced by the amount of child support received, many Non Custodial Parents will continue to pay8212and many Custodial Parents will continue to take8212child support 8220under the table.8221 In a controlled State experiment parents receiving a full pass-through received more in child support per month than those who did not. Anecdotal information also suggests that Non Custodial Parents may be more likely to pay child support when they know that the payments are going directly to benefit the family. Assuming this is correct, child support pass-through may increase the amount of child support collected.

Does Receipt of Child Support Enhance the Likelihood of Welfare Exits?

Based on State research results, the potential for child support to increase welfare exits depends not only on the amount of child support received but also on the percentage of welfare benefit that this child support replaces. This means that the same child-support benefit would be more likely to produce welfare exit in a low-benefit State than in a high-benefit State.

Implications for Future Study

Study outcomes indicate that similar8212perhaps even slightly more positive8212gains could be achieved for former-assistance families through a child support arrears distribution policy that simply pays all Custodial Parent-Owed Arrears ahead of State-Owed Arrears. The model further indicates that loss in State child support arrears-related income would be less than 6 percent (or $8.03 per month for each case) overall. This approach would accomplish the family-friendly outcome that Congress intended and, at the same time, could ease implementation difficulties for States.

We recommend further study of the implications surrounding PRWORA8217s distribution policy. Follow-up activities could include establishing the positive effect on Custodial-Parent behavior, the cost and difficulty of implementing PRWORA changes within State child support enforcement systems, and a more precise cost estimate for this recommended policy simplification.

The Study in Brief

Background

Congressional passage of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA) marked the transformation of welfare from a program that issues benefits to a program that offers self-sufficiency. PRWORA created the Temporary Assistance for Needy Families program (TANF) and mandated numerous and substantial changes in the Child Support Enforcement (CSE) program. Taken together, these changes provided for sweeping reform of the nation8217s approach to welfare. TANF emphasizes the temporary nature of cash public assistance. Congress recognized that the enforcement, collection, and payment of child support would be critically important in helping single-parent families exit the welfare rolls and maintain self-sufficiency. PRWORA continues the requirement that States operate a CSE program meeting Federal requirements in order to be eligible to receive TANF funds.

The CSE provisions in PRWORA require States to adopt of procedures to expedite both the establishment of paternity and the establishment, modification, and enforcement of support orders, and provisions revising the distribution of past-due child support payments to former welfare recipients.

PRWORA required the Office of Child Support Enforcement (OCSE), Administration for Children and Families (ACF), U.S. Department of Health and Human Services (HHS), to conduct a study and submit a report to Congress analyzing the effects of the changes in distribution policy. Specifically, the statute required OCSE to separately analyze the impact on welfare recidivism of pre-public assistance and post-public assistance arrears changes, and to provide an assessment of the overall impact of PRWORA 8220on moving people off welfare and keeping them off welfare.8221

Pre-PRWORA Child-Support Distribution Methods

Before PRWORA, when families applied for public assistance, the Custodial Parent was required to assign to the State the right to collect (1) any child support obligations that had accumulated before the family joined welfare and (2) any such support coming due while the family received welfare benefits. During the time the family was receiving welfare, most child-support collections from the Non Custodial Parent generally were distributed between the State and the Federal Government.

Once the family left the welfare rolls, the State had the flexibility to determine whether the Custodial Parent or the State received payment resulting from arrearage collections first. A State often would keep the full amount of arrears collections until the State and Federal Governments had been reimbursed for all welfare payments to the family.

However, Federal law also contained several incentive and protection provisions for families receiving public assistance, including a requirement that the State 8220pass through8221 to the Custodial Parent the first $50 of each monthly current-support collection while the family was on welfare assistance and to pay to the family the difference between the welfare payment for the month and the support obligation collected for the month (referred to as the "monthly excess support payment").

Post-PRWORA Child-Support Distribution Methods

PRWORA provides for several significant changes in child-support distribution methods.

• PRWORA eliminates the monthly excess support payment requirement and the $50 pass-through.

• PRWORA requires the immediate unassignment of any portion of arrears designated as owed to the State (State-Owed Arrears) that is in excess of the amount needed to reimburse the State for past public assistance to the family.

• Under PRWORA, Custodial Parent-Owed Arrears and State-Owed Arrears have been expanded from two to six debt sub-categories.

• PRWORA limits State flexibility with regard to distribution practices. The intent of these changes is to improve debt collection for families63719in particular, families of former welfare recipients. The new distribution rules reclassify State-Owed and Custodial Parent-Owed debts at a more detailed level and establish a more defined payoff hierarchy that prevents the State from prioritizing State-Owed debt ahead of debts owed to families. In addition, the statute offers the State the flexibility to select the timing of PRWORA implementation that best meets both State interests and available development resources.

• PRWORA establishes the following timeframes relative to implementing the PRWORA changes in distribution policy. These timeframes have the greatest impact on cases where the Custodial Parent formerly received assistance from the State (former-assistance cases).

• Pre-PRWORA (before PRWORA became applicable)63719Prior to October 1, 1997

• Original Two-Step PRWORA Implementation Mandate (later called "Plan A"):

Step 1 (Resulting in partial PRWORA implementation)8722

160October 1, 19978722September 30, 2000

Step 2 (Resulting in full PRWORA implementation)8722

160October 1, 2000, and beyond.

• Plan B (Single-Step PRWORA Implementation Alternative)63719October 1, 1998, and beyond.

Because many States foresaw difficulty in complying with the original PRWORA two-step implementation plan beginning October 1, 1997, an alternative was established by the Balanced Budget Act of 1997 (BBA). "Plan B" offered States the option to defer distribution changes until October 1, 1998, and then to implement distribution changes for former-assistance cases in a single step. With Plan B's one-step implementation, States must fully implement all provisions under PRWORA two full years in advance of the original (now called "Plan A") October 1, 2000, deadline.

The goal of the PRWORA child-support distribution provisions is to create a distribution priority that favors families once they leave the assistance rolls. With States required to adhere to a federally determined distribution hierarchy for former assistance cases, a larger portion of child-support payments should become available to the family. This additional family income should then reduce dependence on public assistance by both promoting exit from TANF and preventing entry and re-entry to TANF.

Status of State PRWORA Policy Elections

States were found to be at varying stages of completion in implementing PRWORA changes. Twenty-eight jurisdictions have elected PRWORA implementation under Plan A; and twenty-six have elected PRWORA implementation under Plan B.

Key Factors Affecting the Study

Several key factors affected the conduct of this study:

• Delayed State PRWORA Implementation63719Few States have fully implemented the required PRWORA programming changes in their child support enforcement (CSE) automated systems and, among those that have, full compliance has not yet been verified.

• Requirement for Extensive Historical Data63719The requirement for an extensive amount of historical statewide data related to financial transactions and debt balances within State CSE automated systems presented a further complicating factor.

• Absence of a Control Group63719Although the study was able to produce some quantifiable totals regarding an increase in Custodial-Parent income caused by PRWORA distribution changes, it is limited in its ability to definitively produce conclusions regarding the effects this increased income has had on Custodial-Parent welfare-exit/-reentry behavior. In the absence of a controlled experiment, other studies examining the effects of child support receipt on welfare exit and recidivism were reviewed.

Methodology

Through computer-based modeling, a series of scenarios was established to mirror State behavior before and after PRWORA implementation for each scenario analyzed. The total outstanding arrears balances, the total funds distributed on behalf of these arrears, and the total grant payments issued under TANF were established. These outcomes were contrasted with results prior to the implementation of the PRWORA distribution changes.

Within all scenarios, current-support obligations were always satisfied ahead of arrears debts.

Four distribution policy scenarios were established to distribute arrears:

• Scenario 1: Pre-PRWORA State-First Distribution Policy

  Under this policy, States satisfied all State-Owed Arrears ahead of any Custodial Parent-Owed Arrears.

• Scenario 2: Pre-PRWORA Custodial Parent First Distribution Policy or

  Plan A First Step Distribution Policy, Where State Elects To Pay Custodial Parent First

  Under this policy, States satisfy all Custodial Parent-Owed Arrears ahead of any State-Owed Arrears.

• Scenario 3: Plan-A First-Step Distribution Policy, Where State Elects To Pay State First

  Under this policy, States must first satisfy any Never-Assigned Arrears owed to the Custodial Parent, in conformance with PRWORA. If any funds remain, State policy then determines the subsequent payoff sequence.

• Scenario 4: Full PRWORA Implementation (Plan-A Second-Step/Plan-B Single-Step) Distribution Policy

  Under this scenario, States pay arrears according to a PRWORA-prescribed sequence.

The study began in October 1998 and the collection and analysis of State data was conducted from January through June 1999.

Data for the study were obtained from automated databases of the participating States and from relevant research reports, surveys, and other sources.

1999 REPORT TO CONGRESS:

Analysis of the Impact on Welfare Recidivism

of PRWORA Child Support Arrears Distribution Policy Changes

I.160Introduction

Congressional passage of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA) marked the transformation of welfare from a program that issues benefits to a program that offers self-sufficiency. PRWORA created the Temporary Assistance for Needy Families program (TANF) and mandated numerous and substantial changes in the Child Support Enforcement (CSE) program. Taken together, these changes provided for sweeping reform of the nation8217s approach to welfare. TANF emphasizes the temporary nature of cash public assistance. Congress recognized that the enforcement, collection, and payment of child support would be critically important in helping single-parent families exit the welfare rolls and maintain self-sufficiency. PRWORA continues to require States to operate a CSE program meeting Federal requirements in order to be eligible to receive TANF funds.

The CSE provisions in PRWORA require States to adopt a series of procedures to expedite both the establishment of paternity and the establishment, modification, and enforcement of support orders. These requirements include specified actions State CSE agencies must take in processing cases63719including computer matches with newly established State and Federal databases. Other provisions include the establishment of an automated network to link States to information about parental assets and to assist in locating parents; the introduction of additional enforcement techniques; elimination of the requirement that States pass through the first $50 of child support paid directly to the recipient; and revision of distribution of past-due child support payments to former welfare recipients.

Section 457 of the Social Security Act required the Office of Child Support Enforcement (OCSE), Administration for Children and Families (ACF), U.S. Department of Health and Human Services (HHS), to conduct a study and submit a report to Congress analyzing the effects of the changes in PRWORA distribution policy. Specifically, the statute required OCSE to separately analyze the impact on welfare recidivism of pre-public assistance and post-public assistance arrears changes, and to provide an assessment of the overall impact of PRWORA 8220on moving people off welfare and keeping them off welfare.8221 In addition, the statute directed OCSE to offer any further policy recommendations regarding the implementation of PRWORA distribution-related requirements.

This report provides the results of the study conducted by BETAH Associates, with Joanne Barnhart as Principal Investigator and Deborah Chassman as Investigator, and TRW Inc. to assist OCSE in responding to this statutory requirement.

• 
  
  In Part II, Background, of this report, we provide the information for understanding the results of the study. We furnish information about the distribution of arrears procedures prior to the changes made by PRWORA, describe the PRWORA arrears distribution procedures as modified by the Balanced Budget Act of 1997 changes in implementation dates, and highlight the differences in the distribution hierarchy.

• In Part III, Status of State PRWORA Policy Elections, we provide a table that summarizes State policy decisions related to PRWORA implementation.

• In Part IV, Underlying Study Assumptions, the underlying assumptions of the study regarding the anticipated effects of the distribution changes are provided.

• In Part V, Key Factors Affecting the Study, we discuss various factors that limit the methodologies available to produce definitive study findings.

• In Part VI, Methodology, we describe the methodologies that were selected to develop data for the study, and provide the detailed steps used to capture the monetary impact of the arrears changes.

• In Part VII, Study Findings Based on Model Outcomes, we present the actual monetary impacts as determined using computer-based modeling of State child support data.

• In Part VIII, Data From Other Research and Evaluation Findings, we discuss findings from other research evaluations, and studies concerning the effects of child support payments on welfare entry, exit, and recidivism.

• In Part IX, Data From Custodial-Parent Surveys, we describe the results from Custodial-Parent surveys related to this study.

• Finally, in Part X, Research and Study Findings, we present our conclusions about both the impact of the distribution of arrears changes on the amounts of child support distributed to families and how these changes in distribution are likely to affect welfare exit and welfare recidivism.

II.160Background

Essential to design, conduct, and interpretation of the study was an understanding of the differences between pre-PRWORA and post-PRWORA (PRWORA-mandated) child-support arrears distribution methods utilized by the States.

A.160Pre-PRWORA Child-Support Distribution Methods

To properly grasp how welfare reform financially impacts families, the distribution rules that State Child Support Programs previously operated under must first be understood. A useful summary of past practice is contained within the following excerpt from the 1998 Green Book.¹

8220Two parties have claims on child support collections made by the State. The children and Custodial Parent on behalf of whom the payments are made, of course, have a claim on payments by the Non Custodial Parent. However, in the case of families that have received public aid, taxpayers who paid to support the destitute family by providing a host of welfare benefits also have a legitimate claim on the money.

Thus, over the years a series of somewhat complex rules has developed to determine who actually gets the money. It is helpful to think of these rules in two categories. First, there are rules in both Federal and State law that stipulate who has a legal claim on the payments owed by the Non Custodial Parent. These are called assignment rules. Second, there are rules that determine the order in which child support collections are paid in accord with the assignment rules. These are called distribution rules.

As long as families remain on welfare, the distribution of child support is straightforward. When families apply for the Aid to Families with Dependent Children (AFDC), the Custodial Parent must assign to the State the right to collect any child support obligations that accumulated before the family joined welfare as well as support that comes due while the family is receiving welfare benefits. As long as the family remains on welfare, child support collections are generally kept by the State and split with the Federal Government.

Consider a simple example. Suppose that when a given mother signed up for welfare, the child support agency was successful in locating the father, establishing a support order for $200 per month, and collecting the payments. Each month, the State would retain the $200, which in turn would be split with the Federal Government. In addition, the amount of welfare reimbursement owed to the State by the Non Custodial Parent would be reduced by $200 each month. If the AFDC benefit were $300 per month the amount owed to the State by the Non Custodial Parent would increase by only $100 each month rather than the full $300.

Once families leave welfare, the amount of support assigned to the State is the amount that equals total AFDC payments to the family minus any child support paid by the Non Custodial Parent while the family was on welfare. At the moment the family leaves welfare then, the Non Custodial Parent usually owes child support to both the government and the family. The amount owed the family is the amount of payments that accumulated before the family went on welfare plus any amount that accumulated because of non-payment after the family leaves welfare.

The real issue, of course is the order in which child support will be paid against these debts once the family leaves AFDC. The first rule is straightforward: Payments against current support collections always go to the family. In the case above, no matter how long the mother was on welfare, the first $200 of monthly payments is assigned to and distributed to the mother once the family leaves welfare. If the father never pays against arrearages, the government never gets repaid for the AFDC benefits it provided and the mother never gets repaid for arrearages that accrued before or after the family was on welfare.

Now assume that the father begins to make payments in excess of the current support amount of $200. The issue arises of whether the State can keep the amount above the current support as repayment for AFDC benefits or whether the State must give the arrears payments to the family. Here we see that distribution law trumps assignment law under some circumstances; namely, whenever two or more parties have been assigned child support that is past due. Both parties have legal claims; the issue is which one is paid first.

Before the 1996 welfare reform law was enacted, Federal law allowed States to design their own distribution rules to determine who got arrearage collections. States could even keep the entire amount of AFDC benefits and not share any of it with the family. Only when the State and Federal Governments had been repaid the entire amount of AFDC benefits provided to the family were States required to pay arrearage collections to the family.8221

With this flexibility to design their own distribution rules, States split fairly evenly between those that retained arrears collections to reimburse the State for past AFDC benefits and those that elected to pay any arrears collections first to the family.

Prior to the 1996 welfare reform enactment, Federal law also contained several incentives and protection provisions for families. Perhaps the most notable of these provisions was the requirement to 8220pass through8221 to the family the first $50 of each monthly Current Support collection while the Custodial Parent was receiving welfare assistance. This requirement was codified in former Federal Regulations at 45 CFR Section 302.51(b)(1). This incentive encouraged Custodial Parents to cooperate with the State in obtaining and enforcing child support obligations on welfare cases. In situations where a Custodial Parent had children by more than one Non Custodial Parent, resulting in more than one child support case, collections were tallied together for all cases to ensure that only a single $50 monthly maximum pass through was distributed to the Custodial Parent.² This $50 pass-through to families was disregarded for purposes of income testing by the State AFDC agency.

Another provision within the law prevented States from retaining more in current support collections received in any month than the State in turn paid to the family as welfare assistance during the same month. Instead, the State was required to reimburse the family for any portion of the monthly current support collection that exceeded the monthly grant.³ Because this provision was described within paragraph (b)(3) in Section 302.51 of Title 45 of the Code of Federal Regulations, these monthly excess support payments are often referred to as "(b)(3) payments."

Finally, the law contained a provision to prevent States from retaining more in welfare assistance reimbursement from child support collections than the States had actually paid to families as AFDC grants over the entire life of the welfare case.⁴ Because this provision was described within paragraph (b)(5) in Section 302.51 of Title 45 of the Code of Federal Regulations, these monthly excess support payments are often referred to as "(b)(5) payments."

B.160Post-PRWORA Child-Support Distribution Methods

PRWORA provides for several significant changes. First, PRWORA requires the immediate unassignment of any portion of arrears designated as owed to the State (State-Owed Arrears) that is in excess of what was needed to reimburse the State for past public assistance. Next, PRWORA eliminated the Section 302.51 (b)(1) $50 pass-through and (b)(3) monthly excess support payment requirements, as described earlier, in public assistance cases. These incentives were replaced with the option for States to establish State-specific incentive policies, provided that only State funds were used. Finally, under PRWORA, arrears owed to the Custodial Parent and the State have been subdivided. PRWORA limits State distribution flexibility and establishes an arrears payoff hierarchy. These provisions are further described below.⁵

1. Requirement To Immediately Unassign Arrears63719When a family exits TANF, States may retain arrears to ensure full reimbursement for past welfare assistance. However, when the amount of State-Owed Arrears exceeds the Unreimbursed Assistance Balance (i.e., when the amount of State-Owed Arrears is larger than the amount needed to reimburse the State for past public assistance), under PRWORA, this excess must be immediately unassigned to the family.

The requirement to immediately unassign arrears is intended to improve the financial outcomes of Custodial Parents. With arrears returned to Custodial Parents immediately upon their exit from TANF, Custodial Parents no longer need to wait until the State has been fully reimbursed for its grant payments by the Non Custodial Parent before receiving any distributions on these arrears.

State-Owed Arrears can exist in two forms63719Temporarily Assigned and Permanently Assigned. Temporarily Assigned Arrears represent the portion of State-Owed Arrears that a Custodial Parent was required to assign to the State as a condition of obtaining public assistance. Permanently Assigned Arrears represent the portion of State-Owed Arrears that accrued while the Custodial Parent was receiving public assistance.

Under PRWORA, when both types of State-Owed Arrears exist in the same case, State policy determines which category should first be reduced by the unassignment. This choice is significant because it can impact both the amount of arrears that the family receives and the distribution priority that these arrears are assigned in the payoff hierarchy.

In situations where a Custodial Parent is active in more than one case, State policy (rather than Federal policy) defines the specific methodology used to apportion these unassigned arrears to each case.

1. Elimination of Section 45 CFR 302.51 (b)(1) and (b)(3) Payments63719PRWORA eliminates the incentives contained within Section 45 CFR 302.51(b)(1) $50 pass-through and (b)(3) monthly excess support payments to TANF families. However, PRWORA preserves the distribution provisions formerly referenced in 45 CFR 302.51(b)(5) to ensure that States never retain more in child support reimbursement than was paid to families under public assistance. The elimination of (b)(1) and (b)(3) payments has, in effect, shifted the authority for setting Custodial Parent incentives and associated costs directly to States, whereas the continuation of (b)(5) maintains Custodial Parent protections.

Under PRWORA, States are still permitted to pass through a portion of the State share of arrears collected to TANF families. However, the Federal share of the collections no longer can be used to help pay for these pass-through distributions. Although no Federal financial participation is provided for these pass-throughs, States may count as TANF Maintenance of Effort (MOE) funds the State share of arrears passed through to TANF families and disregarded in determining the amount of TANF benefits.

According to a report released by the Center for Law and Social Policy (CLASP) ⁶, 24 States have adopted a child support pass-through of some amount. Although the approaches vary from State to State, the effect in States that have opted to continue providing a pass-through is to maintain or increase the amount of money available to the TANF families that receive child support payments. Fourteen (14) States provide for a $50 pass-through, and disregard the amount of the pass-through for purposes of determining TANF eligibility and benefits. In addition, Kansas and Nevada provide for pass-throughs of up to $40 and $75, respectively.

Although Connecticut, Vermont, and Wisconsin are the only States that pass through all child support, Delaware, Georgia, Maine, South Carolina, and Tennessee provide for fill-the-gap budgeting that has the effect of passing through significant amounts of child support. However, TANF families who receive child support payments in the 26 States that have eliminated the pass-through have suffered a decrease in available money. It should be noted that because Vermont and Wisconsin are conducting experimental design evaluations, control-group recipients receive only up to $50 of the child support pass-through. Table 1, prepared by CLASP, identifies State practices regarding pass-through provisions as of January 1999.

Table 163719State Policy Re: Pass-Through and Disregard of Current Month8217s Child Support Collected for Families Receiving TANF-Funded Cash Assistance as of January 1, 1999

Information updated based on State survey conducted by Lewin Group.

1. 
   
   Change in Distribution Processing

With the intent of improving debt collection for families (in particular, families that are former welfare recipients), PRWORA limits prior State flexibility with regard to distribution practices. The new rules reclassify State-Owed and Custodial Parent-Owed debts at a more detailed level and establish a more defined payoff hierarchy that prevents States from prioritizing State-Owed debt ahead of debt owed to families. Finally, PRWORA rules allow States to select the timing of PRWORA implementation that best meets both State interests and available development resources.

a.160Arrears subclassification. States are now required to subclassify Custodial Parent-Owed and State-Owed Arrears into new subclassification categories. The new subclassifications are designed to help distinguish both the circumstances under which each arrears component was accrued or collected and which component takes priority in the distribution hierarchy. The new subclassifications are as follows:

• Permanently Assigned Arrears: The portion of State-Owed Arrears that accrues while the Custodial Parent is receiving public assistance and all pre-October 1, 1997, assigned arrears.

• Temporarily Assigned Arrears: The portion of the total State-Owed Arrears remaining after Permanently Assigned Arrears have been removed.

• Never-Assigned Arrears: The portion of Custodial Parent-Owed Arrears that accrues following the last period of public assistance.

• Unassigned Pre-Assistance Arrears: The portion of Custodial Parent-Owed Arrears that accrues prior to the Custodial Parent8217s last entry into public assistance and that is later unassigned upon exit from public assistance.

• Unassigned During-Assistance Arrears:160The portion of Custodial Parent-Owed Arrears that accrues while the Custodial Parent is receiving public assistance and that is unassigned after public assistance ends.

• Conditionally Assigned Arrears: Beginning October 1, 2000, for Plan-A (two-step implementation) States, and October 1, 1998, for Plan-B (single-step implementation) States⁸, States must convert any Temporarily Assigned Arrears to Conditionally Assigned Arrears for all former-assistance cases. Conditionally Assigned Arrears refers to a debt that is owed to the family unless it is collected from the Non Custodial Parent through the Federal Income Tax Refund Offset.

1. New Subclasses Determine Distribution Hierarchy. A description of the distribution hierarchy for the arrears subclassifications follows in subsection d, Implementation Timeframes.

c.160Requirement To Pay Never-Assigned Arrears First. One of the most visible changes in PRWORA is the requirement that States pay all Never-Assigned Arrears ahead of any State-Owed Arrears. In a former-assistance case, the term 8220Never-Assigned Arrears8221 refers to all arrears that accrue after the end of a family8217s most recent period of assistance.

1. Implementation Timeframes. As Table 2 indicates, PRWORA established specific timeframes for implementing the changes in distribution policy. These timeframes have the greatest impact on cases where the Custodial Parent formerly received assistance from the State. For these former-assistance cases, PRWORA established three timeframes that impact the distribution of arrears. These timeframes are:

• Pre-PRWORA (before PRWORA became applicable)8212
  
  Prior to October 1, 1997

• Original two-step PRWORA Implementation Mandate (later called 8220Plan A8221):

Step 1 (resulting in partial PRWORA implementation)8212

October 1, 1997, through September 30, 2000

Step 2 (resulting in full PRWORA implementation)8212

October 1, 2000, and beyond

• Plan B (Single-Step PRWORA Implementation Alternative)8212
  
  October 1, 1998, and beyond.

Table 263719Regular (Non-IRS Interrupt) Distribution Payoff-Hierarchy Rules

on Former-Assistance Cases

States are subject to increasingly detailed guidelines regarding the prioritizing of Custodial Parent-Owed and State-Owed Arrears distribution within each of these timeframes. A correspondingly greater emphasis is placed on satisfying Custodial Parent-Owed Arrears ahead of State-Owed Arrears, as the arrears payoff hierarchy becomes more regimented. States were encouraged to have certain enhanced-processing capabilities in place by October 1, 1997, to comply with the requirements. These enhanced features included the capability of distinguishing between Permanently Assigned and Temporarily Assigned Arrears and between Never-Assigned and Unassigned Arrears.

Because many States found compliance with the PRWORA Plan-A (two-step implementation) deadline to be problematic, an alternative was established on August 5, 1997, in the Balanced Budget Act of 1997 (BBA), PL 105-33. This alternative was labeled "Plan B," with the original PRWORA directive dubbed "Plan A." Section 5532 of the BBA revised assignment section also added Section 457(b)(6) to the Social Security Act, offering States the option of deferring distribution changes until October 1, 1998, and then implementing distribution changes for former-assistance cases in a single step. This implementation extension was a welcome reprieve; however, it posed a significant disadvantage for some States. With Plan B's one-step implementation, States must fully implement all provisions under PRWORA two full years in advance of the original (Plan A's) October 1, 2000, deadline.

Section 454(24)(B) of the Social Security Act specifies that State automated systems must meet PRWORA requirements by October 1, 2000.⁹ However, as described above, distribution and arrears-assignment processing requirements under PRWORA have earlier implementation deadlines. Policy guidance from OCSE specifies that States must meet the specific effective date for each PRWORA requirement.¹⁰

e.160Summary of Hierarchy Changes. Table 2 above sets forth the distribution hierarchy required for former assistance cases under PRWORA implementation Plans A and B during each implementation period. The table also illustrates areas where States continue to have flexibility in prioritizing arrears payoffs within the distribution hierarchy.

1. Implications for Pre/Post Welfare Exit63719The elimination of the 45 CFR 302.51(b)(1) and (b)(3) payments described in Section 2 above affect individuals receiving welfare. The PRWORA arrears reclassification and distribution hierarchy changes described in Section 3 above affect individuals who have exited welfare.

Therefore, for the purposes of this study analyzing the impact on welfare recidivism of PRWORA distribution policy changes, the primary focus will be on the effects of arrears reclassification and distribution. Because the continuance or termination of (b)(1) and (b)(3) payments may affect a family8217s overall financial well-being and, therefore, its ability to exit welfare, we believe it is important to consider the impact of State practices on Custodial Parent behavior following the elimination of the pass-through requirement. Although under PRWORA States may provide pass-throughs in excess of the $50 previously required, 26 States have eliminated the pass-through.

The goal of the PRWORA child-support distribution provisions is to create a distribution priority that favors families. By eliminating some State discretion in the distribution hierarchy and requiring that States adhere to the federally determined distribution hierarchy, a larger portion of child-support payments should become available to the family. Providing additional income to the family furthers the PRWORA goal of reducing dependence on public assistance by both hopefully promoting exit from TANF and preventing entry and re-entry to TANF.

III.160Status of State PRWORA Policy Elections

The implementation of PRWORA impacted a wide range of modules within each State8217s child support automated system. This impact stemmed from changes that required States to establish and track new debt types, to adjust distribution-processing hierarchies, and to display and report revised financial activity. Because each State automated system differs in the physical implementation of these logical processes and the programming resources available to complete them, States were found to be at varying stages of completion in implementing PRWORA changes. Twenty-eight jurisdictions have elected PRWORA implementation under Plan A and twenty-six have elected PRWORA implementation under Plan B.

Table 3¹¹ shows pre-PRWORA and post-PRWORA State distribution practices, as described above. The chart indicates which States have elected Plan A8217s two-step implementation approach and which have elected Plan B8217s single-step approach. The chart also indicates former pre-PRWORA distribution practices for most States.

Table 363719State Status as of May 1999

IV.160Underlying Study Assumptions

The fundamental assumption underlying this study is that PRWORA-related distribution changes would increase child support payments to former-assistance families and serve as a partial income replacement for the TANF grant. As a result, these PRWORA-related changes would assist former-assistance families in continuing to remain off public assistance.

PRWORA distribution changes are expected to result in the establishment of more uniform and clearly defined distribution procedures across all States. These distribution procedures include prioritizing newly defined categories of Custodial Parent-Owed Arrears ahead of any State-Owed debts and the immediate unassignment of arrears to families when they leave public assistance. Because former distribution policies varied significantly across States, the degree of income increase resulting from PRWORA changes is also expected to vary.

An additional assumption is that the replacement of mandatory incentives to families (formerly paid as (b)(1) and (b)(3) payments) with a State voluntary incentive program is expected to have a measurable influence on family behavior related to exiting from public assistance. Because the voluntary program offers States complete freedom in establishing any level of incentive amount, but without the benefit of Federal reimbursement, specific State policies will exert varying positive or negative influences, depending on the specific incentive levels selected.

V.160Key Factors Affecting the Study

Three key factors are expected to affect the design, conduct, and interpretation of this study: delayed State PRWORA implementation, requirement for extensive historical data, and absence of a control group.

A.160Delayed State PRWORA Implementation

As of the date of this study, all States have now elected Plan A (two-step implementation) or Plan B (single-step implementation) as their approach to PRWORA implementation. However, few have reported full implementation of the changes in their child support enforcement (CSE) automated system. Further, even among States that have completed the necessary programming changes, full compliance has not yet been verified with the newly revised test deck.¹³ Evaluations using the new PRWORA test deck are planned as part of the Year 2000 certification process for States.

Because the study was required statutorily to be completed by October 1, 1999, full implementation of distribution changes had not occurred prior to gathering data for the study. This factor affected the ability of the study team to assess Pre-Assistance and During-Assistance Arrears distribution rule changes.

B.160Requirement for Extensive Historical Data

The requirement for an extensive amount of historical statewide data related to financial transactions and debt balances within State CSE automated systems presented a further complicating factor for the study. States that were excellent candidates for inclusion in the study because of their PRWORA-compliant State child-support policies lacked a statewide-certified system upon which analysis could be performed.

C.160Absence of a Control Group

In an ideal study environment, a control group would be utilized that comprised Custodial Parents with income levels and family circumstances similar to those of the treatment group, but that were not impacted by the PRWORA-mandated distribution changes. The use of such a control group would permit statistically significant inferences to be made regarding the impact of distribution changes on the former-assistance, Custodial-Parent population at large.

8220Many social welfare programs look successful63719to their own staffs as well as to outsiders63719because their Custodial Parents seem to be doing so well. For example, a substantial portion of trainees may have found jobs after having gone through a particular program. The question is: did they get their jobs because of the program, or would they have done so anyway? Answering this question is the central task in evaluating the impact of a program or policy. In other words, what would have happened to the Custodial Parents if they had not been in the program or subject to the new policy?8221¹⁴

Researchers believe that the best way to measure the impact of a program or policy change is through an experimental design in which individuals or families are randomly assigned either to a treatment group that is subject to the new program or policy or to a control group that does not receive the treatment. However, because States are applying the arrearage changes to all eligible Custodial Parents, such an experimental design was not possible.

Furthermore, other techniques for measuring program effectiveness could not be used to measure the effectiveness of the arrearage policy changes. For example, although not as generally accepted by professional evaluators, with proper statistical adjustments, various quasi-experimental designs that select a comparison group whose members are not subject to the intervention but nevertheless are thought to be similar to the treatment group, can be used to measure the impact of a policy change. However, the usual methodologies for selecting comparison groups are not available for the arrearages measures; because all recipients are by definition 8220participants,8221 we could not select a group of non-participants with similar characteristics. Nor could we compare the results with any comparison sites utilizing different rules, since the distribution changes are both statewide and nationwide.

Although 8220pre-post8221 comparison or 8220time series and cross-sectional analysis8221 techniques sometimes can be used to compare outcomes over time, we believe that major changes created by other sections of the PRWORA legislation preclude distinguishing which policy changes account for any differences. For example, these techniques would have required adjustment for the other major welfare-reform changes63719such as the 60-month time limit and the 24-month work requirements63719which have taken place over the same time period and affect the same populations. Finally, it clearly would have been impractical to treat any Custodial Parent in a less advantageous manner with regard to fund distribution, simply to further the interests of this study.

As a result, although the study was able to produce some quantifiable totals regarding the increase in Custodial Parent funds, it is limited in its ability to definitively draw conclusions regarding the effects these increased funds have had on the welfare-exit/re-entry behavior of Custodial Parents. We have, however, reviewed other studies examining the effects of child support receipt on welfare exit and recidivism and reviewed surveys of former welfare recipients. In the absence of a controlled experiment, we utilized the quantified results of the changes and these other data to determine whether the PRWORA distribution changes are likely to increase welfare exit and/or decrease welfare recidivism.

VI.160Methodology

The study team established a series of scenarios mirroring State behavior both before and after PRWORA implementation for each scenario analyzed. The total outstanding arrears balances, the total funds distributed on behalf of these arrears, and the total grant payments issued by TANF were established. These outcomes were contrasted with results prior to the implementation of PRWORA distribution changes.

As described in Section V above, because assessment of the impact of distribution changes on Custodial-Parent behavior could not be precisely measured, the quantitative data was supplemented with qualitative information from existing relevant research evaluations and reports.

A.160Study Team

The study team comprised a former Welfare Commissioner, several senior-level system developers with extensive experience in State CSE systems, former CSE and welfare policy specialists, and a statistician. In addition, numerous State and Federal CSE staff provided invaluable support.

B.160Timeframe During Which the Study Was Conducted

The study began in October 1998 and the collection and analysis of State data was conducted from January through June 1999.

C.160Data Sources

Data for the study were obtained from the following sources:

1. Automated System Databases of the Participating States63719The quantitative portion of this study relied upon application and report data from several of the study States. Data to populate the distribution model was extracted from shadow production databases established by State child support automated systems to support reporting and testing activities. These databases contained copies of actual 8220live8221 data from State systems, which were current to within a few weeks of the 8220live8221 data from which they were extracted. The data types gathered included monthly current-support starting balances, start-of-month arrears balances, monthly collection amounts, and distribution amounts. The information was gathered for periods of several months during 1998 and 1999. To ensure confidentiality, person and case identifiers were removed from the data prior to its use.

Taken together, the States included in the analysis do not constitute a statistically significant sample adequate to project a precise nationwide estimate of Custodial Parent economic outcomes. A substantial level of effort was required to extract, process, and analyze the data. Unfortunately, resource and time constraints precluded significantly increasing the number of States selected. However, the analysis results obtained do accurately forecast the relative impacts that differing distribution policies have on Custodial Parent distribution outcomes. The results also provide an indication of the types of distribution increases that Custodial Parents could be expected to receive.

1. Research Reports, Evaluations, and Studies63719Due to the unavailability of actual experiential data specifically for purposes of this study, a literature search was conducted to identify both previous and ongoing research examining the dynamics of child support payments and welfare entry, receipt, exit, and re-entry. Although substantial national research and analysis of the dynamics of welfare movement has been conducted, few of these efforts explore and analyze how child support payments affect such recipient behavior.

Fortunately, a number of States currently are conducting evaluations that examine factors related to the study. Thus, identified sources included: State-specific studies focused on the interaction between child support and welfare exit; and national reports and studies focused on employment and demographic information for individuals exiting welfare to enter employment.

A complete bibliography is provided as Attachment 2.

Four States63719Wisconsin, Washington, Texas, and Virginia63719have research projects underway to examine the following factors related to this study:

a.160Custodial Parent behavior after implementation of PRWORA distribution policy changes in child support payments.

b.160The role of child support in moving individuals into employment and off welfare.

c.160The role of child support in helping former welfare families stay off welfare.

d.160The kinds of support required to help women on welfare make the transition to employment, and the employment path of such women.

e.160How former TANF families are faring after their exit from welfare.

f.160The design of State welfare programs in the post-PRWORA environment.

Although none of the State evaluations have been completed, interim reports and

findings were used to prepare this report.

3.160Other Reference Sources63719Other reference sources include CSE programmatic and statistical information; and informational policy memoranda describing State pass-through policies.

4.160Custodial Parent Surveys63719The following sources were used to obtain information on Custodial-Parent views regarding the importance of child support payments in exiting and remaining off welfare: A survey of former welfare recipients conducted by the State of Wisconsin¹⁵; and the report Making Ends Meet¹⁶, which catalogues and summarizes interviews with 379 low-income single mothers from a four-city survey of families who left welfare.

D.160Four Scenarios: Approach to Analysis of the Monetary Impact of the Arrears Changes

A necessary first step to evaluate what effect PRWORA distribution changes have on welfare recidivism is to quantify the resulting level of increase in child support payments that families now receive. Because prior to PRWORA, States had substantial flexibility in setting distribution their child support distribution policies, the analysis must explore the full range of effects that PRWORA has had across States with differing former policies. That is, the analysis must contrast the impact that these changes had on States that formerly emphasized State-Owed debts ahead of Custodial Parent-Owed debts and those States that behaved in the reverse. Additionally, this evaluation should also permit Plan-A (two-step implementation) States to be evaluated separately from Plan-B (single-step implementation) States.

Finally, the analysis should distinguish between the influence that Pre-Assistance and Post-Assistance Arrears rule changes have on distribution outcomes. That is, the study should distinguish between the effect that distributions associated with Unassigned Pre-Assistance and Conditionally Assigned Arrears have on family income and the effect that distributions associated with Unassigned During-Assistance Arrears have on family income.

Ideally, this analysis would be conducted by examining CSE systems within States to measure the actual distribution outcomes that occurred during the months before and after PRWORA was implemented. However, this approach poses several difficulties. First, many States are still in the process of implementing the welfare reform changes and lack adequate post implementation results. Even in States that have reported completion of PRWORA programming changes, certification reviews employing the financial test deck have not yet been performed. A more troublesome problem with this approach involves our inability to isolate the impact of PRWORA distribution policies. Other influences stemming from State policy changes and program-specific factors also impact caseload, collection, and distribution outcomes. Examples of these factors include the recent New Hire Reporting and Financial Institution Data Matching initiatives that enhance a State8217s ability to improve collections.

To avoid these difficulties, a computer-based modeling approach was selected. The model8217s goal was to permit researchers to simulate typical pre-PRWORA and post PRWORA State behavior by altering the model8217s distribution rules to correspond with each desired policy scenario. Actual case starting month arrears balances and monthly collections data were obtained from several States to populate the model. Outcomes were then generated for each scenario. The model was also exercised with data from several different months to ensure that results obtained were consistent over time.

Within all scenarios, current-support obligations were always satisfied ahead of arrears debts. Four distribution policy scenarios were established to distribute arrears, as follows:

Scenario 163719Pre-PRWORA State-First Distribution Policy

Under this policy, States satisfied all State-Owed Arrears ahead of any Custodial Parent-Owed Arrears. This policy was only allowed prior to PRWORA implementation.

Scenario 263719Pre-PRWORA Custodial Parent-First Distribution Policy or

Plan-A First-Step Distribution Policy, Where State Elects To Pay Custodial Parent First

Under this policy, States satisfy all Custodial Parent-Owed Arrears ahead of any State-Owed Arrears. States have employed this scenario prior to PRWORA implementation and can continue to implement this policy until October 1, 2000, by electing to be a Plan-A (two-step implementation) State. This approach is compliant with Plan A, because the first step in Plan A requires only the portion of Custodial Parent-Owed Arrears categorized as Never Assigned to be paid first ahead of State-Owed Arrears. Under this scenario, a State would elect to pay all remaining Custodial Parent-Owed Arrears ahead of any State-Owed Arrears.

Scenario 363719Plan-A First-Step, Distribution Policy, Where State Elects To Pay State First

Under this policy, States must first satisfy any Never-Assigned Arrears owed to the Custodial Parent, in conformance with PRWORA. If any funds remain, State policy then determines the subsequent payoff sequence. Under this scenario, a State would elect to pay State-Owed Arrears ahead of any remaining Custodial Parent-Owed Arrears.

Scenario 463719Full PRWORA Implementation (Plan-A, Second-Step/Plan-B, Single-Step) Distribution Policy

Under this scenario, States pay arrears according to a PRWORA-prescribed sequence, as follows:

• Never-Assigned Arrears

• Unassigned Pre-Assistance Arrears

• Conditionally Assigned Arrears

• Permanently Assigned Arrears

• Unassigned During-Assistance Arrears.

  A brief summary of the steps used to conduct the analysis is presented below:

  1. Former-assistance cases that had collections during a study month were selected for analysis.

  1. For each selected case, existing State CSE balance history and financial adjustment records were used to recalculate the starting monthly balances for current-support obligations and all related arrears debts for each study month.

  1. The arrears starting month balances for Custodial Parents and the State also were reclassified into PRWORA-required arrears subcategories (i.e., Never Assigned, Unassigned Pre-Assistance, Conditionally Assigned, Permanently Assigned, and Unassigned During-Assistance Arrears).

  1. Adjustments were made to correct any State-Owed Arrears that should have been unassigned to the family under PRWORA guidelines (i.e., State-Owed Arrears in excess of the Unreimbursed Assistance balance).

  1. The model8217s distribution hierarchy was set in accordance with the requirements for each scenario.

  1. The model was exercised.

  It is important to note that the study focused solely on non-IRS refund offset collections. IRS refund offset collections are processed in accordance with a different distribution rule, which was not changed under PRWORA. Therefore, the processing of IRS collections fell outside the scope of the model.

VII.160Study Findings Based on Model Outcomes

This section covers results evaluation, impact of immediate arrears unassignment, model findings, and model implications for future study.

A.160Results Evaluation

Figure 1 summarizes the change in Custodial-Parent child support income for a State whose pre-PRWORA policy was to maximize State-Owed Arrears ahead of Custodial Parent-Owed debts. The chart illustrates the State8217s progress initially from a State-first pre-PRWORA policy to a State-first arrears payoff policy under the first-step implementation rules mandated by Plan A and finally to the more detailed set of distribution policy rules prescribed in full PRWORA implementation (Plan A, Second Step or Plan B, Single Step). The distribution results presented reflect the weighted average of several months of model outcomes for all States. The $26.48 average increase represents a 25 percent increase in child-support income, with the bulk of the increase occurring in the initial transition to the first step under Plan A, where the State elects to pay the State first. Although the range of values differs somewhat between States, the results offer some insight into the consequences for Custodial Parents as State PRWORA distribution policies evolve.

Figure 1- Average Custodial-Parent Monthly Distribution Outcomes Associated With Pre-PRWORA State-First Distribution Policy

Figure 2 displays average distribution outcomes to Custodial Parents for a State whose pre-PRWORA policy was to maximize Custodial Parent-Owed Arrears. Because a State that pays all Custodial Parent-Owed Arrears first already meets the distribution requirements specified under Plan A's first implementation step, where the State elects to pay the Custodial Parent first, the chart indicates no change in distribution results. However, when the State transitions to full PRWORA implementation, (Plan A, Second Step or Plan B, Single Step), the chart illustrates an unexpected result. Model outcomes indicate that Custodial Parents in these States may actually experience a decrease in collections. This decrease of $8.03 per month represents a 5.8 percent decrease in the average Custodial Parent monthly distribution. This decrease occurs despite the reassignment of State-Owed Arrears to the Conditionally Assigned Arrears category. The decrease results from the reclassification of a portion of Custodial Parent-Owed Arrears into Unassigned During-Assistance Arrears. Within the distribution hierarchy required by full PRWORA implementation, because these arrears are paid only after all State-Owed Arrears are satisfied, Custodial Parent distributions are reduced.

Figure 2 8211 Average Custodial-Parent Monthly Distribution Outcomes Associated With Pre-PRWORA Custodial Parent-First Policy

Figure 3 plots the distribution outcome trends for Custodial Parents and the State in States that progress from either a State-first or a Custodial Parent-first pre-PRWORA arrears policy. The chart summarizes the range of outcomes that could be expected over time. As expected, because the additional income given to the Custodial Parents formerly was paid to the State, State income trends related to Custodial Parent-first or State-first policy are the inverse of Custodial Parent income trends.

Figure 3 8211 Comparison of State-First and Custodial Parent-First Monthly Distribution Policies

Figure 4 provides a breakout of the average Custodial Parent-Owed Arrears balance per case for each category of PRWORA debt. Although current support is displayed as a component of the total debt owed to the Custodial Parent, this balance represents a monthly recurring obligation rather than a point-in-time outstanding balance. Further, although Conditionally Assigned Arrears are also displayed, these arrears would exist only after PRWORA rules are fully implemented in States and only if the arrears are not collected through tax offset. Under the pre-PRWORA and Plan-A First-Step scenarios, Conditionally Assigned Arrears are still owed to the State as Temporarily Assigned Arrears.

Figure 4 8211 Average Custodial Parent-Owed Debt Per Case Grouped by PRWORA Debt Type

The Unassigned Pre-Assistance and Unassigned During-Assistance Arrears amounts shown include arrears that were immediately unassigned to conform to PRWORA policies.

It is notable that Never-Assigned Arrears comprise the largest portion of Custodial Parent-Owed debt in all States. Because the first step under Plan A prioritizes this debt type ahead of State-Owed Arrears, the first step has a significantly greater impact on Custodial-Parent distributions than the subsequent transition to full implementation. Additionally, the chart illustrates that the average Unassigned During-Assistance Arrears balance is substantially greater than the average Conditionally Assigned Arrears balance. This factor may explain the reduction in Custodial-Parent distributions that occurs when a State that previously elected to pay all Custodial Parent-Owed Arrears first is required to transition to full implementation of PRWORA rules. The loss of Unassigned During-Assistance Arrears-related distributions that can be satisfied only when no State-Owed Arrears remain outweighs the gain in distributions associated with Conditionally Assigned Arrears.

Finally, the chart shows that Conditionally Assigned Arrears, which exist only after PRWORA is fully implemented, represent an additional increase in Custodial Parent-Owed Arrears of 4.83 percent if they are not collected through tax offset.¹⁷

Figure 5 displays a corresponding breakout of the average State-Owed Arrears balance per case for Permanently Assigned and Temporarily Assigned Arrears. As the chart indicates, Permanently Assigned Arrears comprise, on average, 97 percent of the total State-Owed Arrears available. The limited amount of Temporarily Assigned Arrears available underscores the limited effect that these arrears have on distributions when they become Conditionally Assigned Arrears upon full PRWORA implementation.

Figure 5 8211 Average State-Owed Debt Per Case Grouped by PRWORA Debt Type

Figures 6, 7, 8, and 9 provide distribution breakdowns by debt type for each of the four arrears distribution policy scenarios described in Section VI. The data represent weighted averages for all States and months involved in the analysis.

Because in all scenarios current-support obligations must be satisfied first, the portion of distribution associated with current-support debt remains unchanged in each case. When viewed together, the charts clearly show that prior distribution policies within States have a significant effect on how Custodial Parents are impacted by PRWORA distribution policies.

Figure 6 8211 Breakout of Average Monthly Collections Distribution for Pre-PRWORA State-First Distribution Policy (Scenario 1)

Figure 7 8211 Breakout of Average Monthly Collections Distribution for Pre-PRWORA Custodial Parent-First Distribution Policy OR Plan A First Step Distribution Policy, Where State Elects To Pay Custodial Parent First (Scen. 2)

Figure 8 8211 Breakout of Average Monthly Collections Distribution for Plan A First Step Distribution Policy, Where State Elects to Pay State First (Scenario 3)

Figure 9 8211 Breakout of Average Monthly Collections Distribution for Full PRWORA Implementation (Plan A Second Step/Plan B Single Step) Distribution Policy (Scenario 4)

Within Figure 9, the portion of the average payment applied to Custodial Parent-Owed Arrears is further exploded to show the impact of Conditionally Assigned Arrears. As the chart illustrates, Conditionally Assigned Arrears contribute an average increase of only $0.58 to the family8217s total payment, but only if not through tax offset. This amount constitutes less than 1 percent of the child support payment received by the family.

B.160Impact of Immediate Unassignment

As described previously, PRWORA requires States to immediately unassign arrears when a family exits TANF, if the total amount of State-Owed Arrears debt exceeds the amount required to reimburse the State for its past assistance to the family. In addition, States have flexibility in establishing the specific unassignment policy that they will implement. Although unassignment increases the amount of arrears owed to Custodial Parents, it does not, as the model shows, necessarily produce an immediate impact on distribution outcomes. However, the growth in arrears and their eventual collection should improve family income over the long term.

Figure 10 illustrates the percent increase in Custodial Parent-Owed and State-Owed Arrears and current support balances that results from unassignment. Balances from each of three unassignment scenarios displayed represent weighted averages for the start of the month prior to any distributions being processed against the cases. It is important to note that Conditionally Assigned Arrears are classified as Custodial Parent-Owed, even though IRS offset collections applied against this debt would be payable to the State. As the Figure illustrates, the amount of Custodial Parent-Owed Arrears increases substantially (i.e., by 42 percent) when unassignment is performed. Further, election of a family-favored versus a State-favored unassignment policy results in an additional small increase63719$6.70, or 2.2 percent63719in the amount of Custodial Parent-Owed Arrears for regular (non-IRS offset) collections.

Figure 10 8211 Impact of Immediate Unassignment on Arrears and Current Support Balances

C.160Model Findings

Despite the limited number of States analyzed, the model outcomes provide a reasonable indication of the range of outcomes that States can expect under each distribution-policy scenario presented. Perhaps the clearest model finding is that the distribution policy previously elected by a State, given pre-PRWORA flexibility in prioritizing arrears, appears to play a significant role in the degree of income change a family will experience as PRWORA distribution changes are implemented. States that formerly favored State-Owed Arrears in their distribution policy will show a substantial growth in family income under PRWORA, whereas States that favored a family-first policy will experience little income change (and perhaps even a slightly negative change). Further, when States that previously favored State-Owed Arrears elect a two-step approach to PRWORA implementation (Plan-A States), families initially will experience a greater-percentage income increase in the transition to the first step (effective October 1, 1997) than in the subsequent transition to the full implementation rules (effective October 1, 2000). Conversely, Plan-A States that favored the family prior to PRWORA will experience no distribution changes in converting to the first step and only a small possibly negative change in migrating to full PRWORA implementation.

Regardless of prior State distribution policy, one factor that appears to improve Custodial Parent outcomes is the practice of immediately unassigning arrears. Our analysis of baseline data reveals that Custodial Parents will experience a significant increase in the portion of case-related debt to which families can lay claim. This occurred even in States that favored Custodial Parent-Owed Arrears. Although this increase in available arrears to families did not show an immediate corresponding family-income increase within the model, given the limited number of payments that the model actually processed (over a period of a single month), family income would be expected to grow over time as support collections of these arrears increases.

Additionally, when unassignment was performed, the portion of unassigned arrears classified as Unassigned During-Assistance Arrears was also unexpectedly high. Unassigned During-Assistance Arrears only occur when the Custodial Parent8217s current-support obligation exceeds the monthly grant amount while the Custodial Parent is receiving assistance. A possible explanation for this outcome is that the Unreimbursed Assistance (URA) balances for older cases may be artificially low because, during conversion, many States that were required to estimate the initial URA value did so conservatively.

Along with the high level of Unassigned During-Assistance Arrears, Conditionally Assigned Arrears represented only a small portion of a case8217s total arrears following unassignment. Conditionally Assigned Arrears8217 low balances, coupled with the limited amount of associated distributions, appear to underscore the limited role this debt type plays in improving family income.

A final conclusion arising from this study is the very substantial level of programming effort required to reclassify State-Owed and Custodial Parent-Owed Arrears into the appropriate PRWORA subcategories and to implement immediate State-Owed Arrears unassignment. The programming requires an intimate knowledge of the financial structures within each State and completion of an extensive amount of complex logic. For States in the process of completing PRWORA changes, the complexity is further increased by the need to retroactively apply the logic as a result of delays or mistakes in PRWORA implementation.

D.160Model Implications for Future Study/Recommendations for Changes to PRWORA Policy

An analysis of the distribution hierarchy required under full PRWORA implementation does reflect a reasonable theoretical approach to assigning payoff priorities based upon the circumstances under which each subcategory of State-Owed and Custodial Parent-Owed Arrears types accrued. However, the complexity of both the arrears classification scheme and the distribution and unassignment rules poses significant practical challenges.

First, the difficulty in developing distribution models for this study clearly underscores the problems that States are experiencing in implementing PRWORA guidelines within already highly complex financial systems. Additionally, courts, Non Custodial Parents, and Custodial Parents all will require significant education to understand the payment outcomes that they experience. Study outcomes indicate that similar8212perhaps even slightly more positive8212gains could be achieved for former-assistance families through a child support arrears distribution policy that simply pays all Custodial Parent-Owed Arrears ahead of State-Owed Arrears. The model further indicates that loss in State child support arrears-related income would be less than 6 percent (or $8.03) per month for each case) overall. [This simple, more straightforward approach would likely accomplish a more family-friendly outcome; and, at the same time, ease existing implementation difficulties for States.]

We recommend further study of the implications surrounding PRWORA8217s distribution policy. Follow-up activities could include establishing the positive effect on Custodial-Parent behavior, the cost and difficulty of implementing PRWORA changes within State child support enforcement systems, and a more precise cost estimate for this recommended policy simplification.

VIII.160Data From Other Research and Evaluation Findings

This section provides our analysis of findings from other relevant research and analyses.

A.160Employment Patterns and the Earning Power of Women

To examine the effect of receipt of child support63719specifically, increases in child support payments resulting from distribution policies63719in assisting welfare recipients to exit and remain off cash public assistance by retaining employment, the role of child support payments should be considered in relationship to the earning power and employment patterns of welfare recipients. Quantitatively, the lower the individual8217s earning capacity, the greater the relative value of child support payments.

Prior to the passage of PRWORA, studies of welfare dynamics¹⁸ indicated that at any given point in time, about half of the individuals on the AFDC rolls were in the middle of a two-year stay, while approximately half were in the middle of an eight-year stay. Thus, half of the AFDC adult head-of-households were not long-term welfare recipients. Rather, they were individuals who were using the welfare system as a temporary means of support to assist in a time of crisis8212such as death of a spouse, divorce, desertion, or birth of an out-of-wedlock child. The remaining half were, in fact, long-term welfare recipients, with little or no recent attachment to the work force. Recipients in the second category were considered harder to employ because of the length of time they had spent on welfare.

As States have successfully reduced the welfare rolls since the passage and enactment of PRWORA, with the most recent data indicating a national TANF-caseload reduction of 45 percent from January 1993 to December 1998¹⁹ and of 38 percent from August 1996 to December 1998²⁰, the TANF population increasingly comprises hard-to-place individuals. Such a dynamic is not surprising; the expectation was that individuals able to find employment would do so, leaving those with substantial barriers to employment dependent on public assistance. Thus, there are essentially two groups for consideration: (1) those who have exited the welfare rolls to accept employment and (2) those who remain on assistance because they face substantial barriers to employment.

Thus, the role that changes in the distribution of child support play, by increasing child support payments to families, may be different in assisting individuals who have obtained employment to leave the rolls, in assisting employed former recipients from returning to the welfare rolls, and in assisting the hard-to-place residual caseload to exit welfare.

Typically, welfare families are single-parent families headed by women. Recent statistics for families on assistance indicate that 87.9 percent are headed by single parents, the vast majority of whom are female. Only 25.9 percent of welfare recipients are high-school graduates.²¹ Similarly, most Custodial Parents who are owed child support are women. Therefore, an examination of the employment patterns and earnings capacity of low-skilled women is most instructive for purposes of this study.

LaDonna Pavetti, Senior Research Associate, Urban Institute, has done substantial analysis of the employment patterns of women63719especially low-skilled women. Her report, Against the Odds: Steady Employment Among Low-Skilled Women,²² shows that most women take one of two paths to steady employment: They either are steadily employed from the time they reach early adulthood; or they move in and out of the labor force, making a more gradual transition to stable employment. Although a substantial portion of women with moderately low skills make an immediate transition to employment (40 percent), very few extremely low-skilled women do so (17 percent). Nearly half of extremely low-skilled women who make the transition to steady employment initially spend more time not working than working, but then gradually increase their employment levels until they are working steadily.

To the extent that welfare recipients8217 employment prospects mirror those of all women with similar skills, these data suggest the following: It is reasonable to expect that most recipients with moderately low skills will be able to work steadily, although most will work primarily in low-paying and/or part-time jobs and may need access to ongoing work- or income-supports such as child care, transportation, or the Earned Income Tax Credit (EITC). In contrast, the challenge of helping the lowest-skilled recipients make the transition to steady employment is likely to be quite formidable.²³

According to Pavetti8217s report, poor mastery of basic skills such as reading is, perhaps, the key characteristic that distinguishes welfare recipients from other mothers. This conclusion is based on an analysis of the Armed Forces Qualifying Test (AFQT) results of welfare recipients. Almost two-thirds of women between the ages of 26 and 33 who received welfare in 1991 had AFQT scores that fell in the bottom quartile of the AFQT distribution, as compared to just 22 percent of women of the same age who did not receive welfare. Research shows that poor mastery of basic skills is associated with longer stays on welfare; and Pavetti found that women with higher-than-average AFQT scores were more likely than women with average or lower-than-average skills to leave initial spells of welfare for work, less likely to return to welfare, and more likely to leave welfare again if they did return.

In addition, women with low-level basic skills are more likely than women with higher-level skills to have children by age 27, are more likely to be a member of a racial or ethnic minority, and are more likely to experience poor social outcomes that are associated with long-term welfare receipt.

Table 4, 8220Proportion of Women With Selected Characteristics Working Steadily by Age 26/27, by Skill Level,8221 based on data over a two-year period from the National Longitudinal Survey of Youth shows the substantial differences in steady employment among women with 8220extremely low,8221 8220moderately low,8221 and 8220higher8221 skills.²⁴

Pavetti8217s report also emphasizes the importance of income supplements for women transitioning from welfare. She concludes that:

8220Some [recipients] will require a more gradual transition to steady employment. Recipients who make a more gradual transition to steady employment are likely to continue to need government support during periods of joblessness, but the need for such assistance should decline over time. Recipients with moderately low skills primarily will hold relatively low-paying and/or part-time jobs. Thus for this group, work supports, such as child care and transportation assistance and income supplements such as the EITC, may be especially important and may be needed for an extended period of time.8221²⁵

In another report, Moving Up, Moving Out or Going Nowhere? A Study of the Employment Patterns of Young Women,²⁶ Pavetti specifically references the importance of child support: 8220...[P]olicies that provide additional income to working families, such as the Earned Income Tax Credit and child support, are likely to be especially important to women who are unable to make the transition from a bad job to a good job to work steadily in a good job.8221²⁷

Table 48212Proportion of Women With Selected Characteristics Working Steadily by Age 26/27 by Skill Level

B.160Child Support and Income

Important in analyzing the behavioral effect of increases in child support payments to families is obtaining an understanding of the effects of receiving child support (1) as a replacement for the TANF grant, (2) as reflected in the percentage of a family8217s income, and (3) as a regular income supplement.

A theoretical model developed by the Commonwealth of Virginia provides instruction regarding the replacement effect of child support on families entering the workforce and exiting TANF.²⁸ The model utilized the following variables: (1) the maximum TANF benefit levels in the family8217s locality; (2) whether or not the family received child support payments; and (3) whether or not the family has housing costs.

The model shows that depending on the geographic location of the TANF case and whether or not the case had earnings, child support, and/or housing costs, total monthly family-income resources during receipt of TANF ranges from $546 to $719. During TANF, but after employment, total family income resources ranged from $1,247 to $1,430. After TANF, with continued employment, total monthly family-income resources ranges from $1,134 to $1,316.8221

The Virginia model further shows that loss of the TANF benefit does not result in a net loss equal to the full value of the TANF benefit. Food Stamps, child support payments, and increased earnings all contributed directly to a reduction in the net loss of income. Cases that leave TANF with average earnings can have up to half of their lost TANF benefits offset by increased Food Stamp benefits and receipt of the full child support payment (not just the $50 passed through to the family while they were a TANF recipient). Further, expected increases in average earnings can offset the TANF benefit loss by another 20 percent.8221 The model also 8220highlights the importance of promoting continuous employment and increased child support collections as well as the use of the EITC to sustain TANF case income resources.

Similar conclusions are found in a General Accounting Office (GAO) study released in August 1998.²⁹ In response to a Congressional request, GAO provided information on the likely success of States in obtaining child support for families whose TANF benefits are subject to time limits. The GAO study focused on: (1) the success rate of States that experimented with time-limited benefits before the passage of PRWORA in obtaining child support for families that reach their limits; (2) how successful States have been in obtaining child support for families within a five-year period (the maximum time a family may receive TANF); and (3) the implications time limits have for States and families.

Among its findings, GAO noted that many families may not be able to count on child support as a steady source of income when their time-limited benefits expire. Although the issue of time limits, per se, is not a part of this study, the GAO findings provide some relevant and instructive information.

In Connecticut, Florida, and Virginia63719the first three States to implement time-limited welfare63719only about 20 percent to 40 percent of families received any child support in the 12 months preceding time limit-related termination of TANF benefits. In a sample of cases from Washington and Minnesota, about two-thirds of AFDC/CSE cases received some child support payments in the 12 months preceding termination of TANF benefits during the six year period of 1992 through 1997.³⁰ The most relevant portion of this GAO report was the finding that child support can be an important supplement to post-public assistance family income.

Based on findings from another GAO report, Welfare Reform: States Are Restructuring Programs to Reduce Welfare Dependence³¹, GAO noted that many families who leave welfare are employed in low-wage jobs, and that to the extent that earnings of families who leave welfare do not increase over time, 8220based on limited employment-based fringe benefits, their ability to maintain themselves may depend to a great extent on the availability of income supports such as subsidized medical and child care and the EITC.8221

In the welfare reform study, which focused on TANF implementation in seven States, GAO found that in States that tracked wages for individuals after leaving TANF, the mean wages for employed recipients ranged from $5.60 to $6.60 per hour. The GAO report on child support³², discussed previously, referenced this GAO welfare reform report³³, and noted that while such families may be receiving the EITC as well as Food Stamp benefits and Medicaid, their employment may result in significant work-related expenses. Specifically, GAO noted, 8220for such families, child support payments could further enhance family incomes or reduce their need for public assistance.8221 However, for families with no earnings, GAO concluded, 8220child support is unlikely to replace families8217 lost cash assistance.8221 This conclusion is based on the fact that in the three States that were first to impose time limits, 8220the mean monthly child support collected ranged from 22 percent to 60 percent of the mean grant received in the month before termination.8221³⁴

C.160Child Support and Welfare Dynamics

One purpose of child support reforms in PRWORA and predecessor legislation was to increase child support payments to prevent welfare dependency. The reforms are based on the belief that, by increasing available income to low-income households, entries to welfare will be decreased and exits from welfare will be increased. As discussed previously, research examining the specific correlation between receipt of child support and the effect on welfare recipient behavior is not extensive. This section of the report examines the limited research exploring this dynamic.

Wisconsin

A 1993 report, Child Support and Welfare Dynamics: Evidence from Wisconsin,³⁵ by Dan Meyer, Institute for Research on Poverty, examined the effects of child support payments on AFDC exits and re-entries for divorced women in Wisconsin from 1980 through 1989.

Meyer described the labor8722leisure theoretical model construct, in which individuals are assumed to consider the amount of income and leisure they would receive from welfare and from all possible hours of work. Then individuals make choices based on the amount of work that maximizes their well-being, given budget and time constraints. Meyer explained how the model is applied for child support purposes:

8220The standard model predicts the effect of child support on welfare participation. Because child support provides a family with additional income, it clearly makes life off welfare more attractive. Even if the amount of child support is smaller than the amount of AFDC, some women may be able to combine that support with earnings and thus exit (or not enter) AFDC. On the other hand, child support makes staying on welfare somewhat more attractive for women already receiving AFDC, because since 1984³⁶ women have been allowed to keep the first $50 per month of child support without any decrease in their AFDC grant. (Any amount over $40 per month, however, does not benefit the AFDC recipient.) Therefore, small amounts of child support are predicted to have ambiguous effects on AFDC use, but large amounts are predicted to decrease AFDC participation.8221³⁷

Meyer points out that the model has not addressed the possibility that the stigmatic effects of receiving AFDC may result in a devaluation of the AFDC income. Thus, a woman may prefer receiving a smaller child support payment rather than a larger AFDC grant.

In 1990, 39 percent of Wisconsin AFDC families received some child support63719a payment record significantly above the then national average of 16 percent. The average amount collected was about $160, a little more than one-third of the average AFDC grant. In one sample of divorced female Custodial Parents in Wisconsin, about 90 percent received a child support award.

Prior simulation models using Wisconsin data found that the AFDC caseload would be decreased by no more than 19 percent63719even if every case received an award based on the Wisconsin standard and all due payments were collected. This may be explained in part by the fact that most Non Custodial Parents are poor themselves. Thus, the potential amount of child support collected remains small compared to the AFDC grant, especially in a high-benefit State such as Wisconsin. These factors suggest that child support would have a small effect in causing exits from AFDC63719particularly in a high-AFDC benefit State63719although it is possible that child support could play a larger role in preventing re-entry to AFDC.

Meyer8217s analysis was conducted by using administrative data from the Institute for Research on Poverty (IRP) Court Record Data (CRD), which contains a sample of divorce, separation, and paternity cases in which at least one child potentially is eligible to receive child support. Using Social Security numbers and names, this file was matched to three other administrative databases to obtain information on AFDC receipt for every month from 1980 though 1989 and on taxable income in 1980, as well as other demographic data. Meyer used this data in two models, one for entering welfare and the other for reentering welfare.

Meyer found that the amount of child support received by a welfare recipient does have a positive effect on exits from welfare. 8220As child support increases, women become more and more likely to exit, but a woman is not significantly more likely to exit welfare unless she is receiving more than $300 per month.8221³⁸ Further, 8220women are more likely to exit in the first three years than after they have been on AFDC for more than four years.8221³⁹

Welfare re-entry also is positively affected by the receipt of child support: Meyer found that 8220women who had received child support for three or more of the previous six months are significantly less likely to reenter.8221 He noted that 8220[t]his finding may suggest that regularity of child support is effective in keeping women off welfare.8221⁴⁰

Meyer concludes that large amounts of child support would be needed to increase the likelihood of leaving welfare in Wisconsin, but that modest amounts of child support may have a stronger effect in States with lower AFDC benefits. Regarding reentry, the results of the analysis show that the receipt of any amount of child support has a significant effect on decreasing welfare reentry.

Washington

Preliminary findings in a study funded by HHS support the conclusions of the IRP report. Washington is one of three States participating in this HHS study. In January 1999, as part of the study, the Washington State Division of Child Support issued an interim report that focused on cost offsets for the CSE program and welfare use between 1993 and 1997.⁴¹ Specifically, the report focuses on the reduction in welfare costs associated with collection of child support payments.

In an effort to evaluate the impact of child support payments, these payments were isolated from other factors, such as Custodial Parent characteristics, history, and geographic location, as well as Custodial Parent access to State programs that promote self-sufficiency, any of which can cause a reduction in welfare use. Due to the timing of the report, only the pre-TANF Job Opportunity and Basic Skills program (JOBS) could be included as a State program promoting self-sufficiency.

The Washington State study found that receipt of child support payments and participation in JOBS appeared to be jointly beneficial. That is, JOBS helped Custodial Parents exit welfare and good child support payments helped Custodial Parents remain off welfare.⁴² However, although the JOBS program substantially reduced the time spent on welfare, it had little or no effect on recidivism.

The major findings of the Washington State study that are of particular relevance to our study were as follows: Good child support payments were associated with little or no effect on welfare exit rates; and good child support payments after exit from welfare were associated with lower recidivism rates, which substantially increased time off welfare. These findings substantiate the conclusions of the earlier IRP report with regard to the importance of child support in reducing welfare recidivism.

Figures 11, 12 and 13 from the Washington State report illustrate that the primary effect of good child support payments is to lengthen, on average, an individual8217s time off welfare.

Key to Washington State Figures:

CPJN - Poor CSE Collections without JOBS

CGJN - Good CSE Collections⁴³ without JOBS

CPJY - Poor CSE Collections with JOBS

CGJY - Good CSE Collections with JOBS

Figure 118212Survival Times; Spells On Welfare

Figure 128212Survival Times; Spells Off Welfare

The following Figure from the Washington State report shows that both child support payments and the JOBS program may have small effects on the length of working spells. Individuals receiving good child-support payments and participating in JOBS appear to have the longest working spells, whereas those with poor child support payments and JOBS appear to have the shortest spells of work.

Figure 138212Survival Times; Spells Working

Additional data from the Washington State report show that reduced recidivism is associated with good child support payments, regardless of work status. When a Custodial Parent has good collections, as defined in the report, child support payments have been flowing into the parent8217s case account and arrears are small. Thus it can be expected that support payments are more likely to accrue to the Custodial Parent after exit from welfare. That income could then be expected to help the Custodial Parent remain off welfare. These findings indicate that child support payments decrease the time spent on welfare, increase length of employment, and reduce recidivism. However, because arrears amounts are small when a Custodial Parent has good collections, the PRWORA changes in distribution are likely to have a smaller effect on such cases.

Texas

The Role of Child Support in Texas Welfare Dynamics, a report on a study conducted by the Lyndon B. Johnson School of Public Affairs⁴⁴, indicates that findings in Texas reinforce the conclusions from both the Wisconsin (IRP) and Washington State (HHS) studies.

The Texas study focused on the following questions: Which factors have the greatest influence on the award and collection of child support? To what extent does child support influence AFDC exits and recidivism? What is the combined influence of child support and earnings on removing families from poverty?

Administrative data records for a sample of AFDC families were used to perform the analysis.⁴⁵

Findings from the study indicate that 11.7 percent of the sampled caretakers exited AFDC during each quarter of the study. 8220Sixteen percent of total AFDC exits occurred as a result of child support collections. Of the total child support collected, 47 percent was retained by the State to recoup the costs of AFDC, with the remaining 53 percent distributed to families.8221⁴⁶

Significant for the purposes of this study, child support received by the Custodial Parent contributed strongly to the probability that a caretaker would exit AFDC. 8220A $100 increase in quarterly child support collections induced a 2.5 percentage point increase in the probability of exit, raising the quarterly probability of exit from 11.7 percent to 14.2 percent. The amount of child support received by an AFDC caretaker had approximately three times as much power as an equal amount of caretaker earnings in inducing exits from AFDC.8221⁴⁷

It is important to point out that the earnings data used in the Texas study measured the before-tax earnings of the Custodial Parent, whereas the data on the amount of child support measured the amount of money actually received by the Custodial Parent. Also, an individual incurs work-related expenses, such as transportation and childcare, to earn wages. This may provide a partial explanation for the result that a given amount of child support has more exit-inducing power than a like amount of the Custodial Parent8217s earnings.

Other economic and demographic factors that influenced the probability of AFDC exits in the Texas study reflect Pavetti8217s analysis of the employment patterns of women, as described above. Specifically, caretakers who have completed high school are more likely to exit than those with lower educational attainment. Individuals with a history of long-time AFDC receipt and many younger children are less likely to exit AFDC.

In contrast to results from the other studies referenced, findings from this study indicate that child support has a strong positive influence on AFDC exits. Although the effect on exits appears contradictory, that is not necessarily the case. As Meyer pointed out in the IRP study, the effect on exits may vary depending on whether the State is a high-benefit or low-benefit State. Both Wisconsin and Washington State are high-benefit States, whereas Texas is one of the lowest-benefit States in the nation.⁴⁸ Therefore, receipt of child support as a relative percentage of income has a higher value in Texas than in high-benefit States. Further, the combined receipt of child support and earnings would make more Texas recipients ineligible, based on income limitations for AFDC/TANF63719thus increasing exits from the rolls of families receiving child support.

With regard to recidivism, the Texas study found that approximately 35 percent of all caretakers who exited AFDC returned to the rolls within one year. 8220For every $100 of child support received per quarter, the probability of recidivism in that quarter was reduced by 1 percentage point. As with exits, the receipt of child support had over a three times negative relationship with/reductive effect on recidivism than an equivalent dollar amount of the caretaker8217s own earnings.8221⁴⁹ Thus, Texas findings with respect to the impact of receipt of child support on recidivism are consistent with the other studies.

Virginia

A report on the Virginia experience, What Happened to Virginia8217s First 94 Two-Year Limit Cases?⁵⁰, provides additional information regarding the role of child support in welfare dynamics. Virginia obtained a Federal waiver to implement its welfare reform program, VIEW, in July 1995. VIEW provides for a mandatory work requirement for able-bodied TANF parent recipients, a two-year time limit on receipt of benefits, and disregard up to 100 percent of earned income for the TANF benefit calculation. Implementation began in 1995 in two districts, Culpepper and Lynchburg. In January 1998, an analysis of 500 cases that enrolled in VIEW between July and December 1995 was conducted, because these cases would have faced termination due to the two-year limit.⁵¹

As of January 1998, 80 cases were still mandatory⁵², 94 cases reached their 24-month time limit, and 326 cases left TANF before their 24-month time limit. Of the latter 326 cases, 63.6 percent were employed; and 27 percent returned to TANF before January 1998.

The cases that actually reached their 24-month time limit had a variety of income resources available to them after their TANF cases closed. Resources identifiable through the administrative databases included earnings, Food Stamps, Social Security Income (SSI) benefits, and child support. For these 94 families, a higher-than-average number6371948 percent63719received child support, compared to 34 percent for all TANF cases. Child support payments averaged $48 before the TANF case closure, when only up to $50 was passed through to the TANF recipient; and payments averaged higher-than-average $133 after TANF case closure when the Custodial Parent received the entire current child support payment. Finally, even with the loss of TANF benefits, 21 percent experienced an increase in total income resources, due largely to increased earnings and the receipt of their full child-support payment. On average, those employed had an $80 per-month increase in monthly earnings after exit. Prior to the TANF case closing, they had average TANF benefits of $258 per month. Figure 14 shows selected characteristics for closed cases:

Figure 148212Selected Characteristics for 24-Month Time Limit Cases When the Cast Closed

On average, increased Food Stamp, Child Support, and other benefits offset 40 percent of the TANF benefit loss. Cases that reduced their attachment to the labor force experienced the greatest offset to their TANF benefit from Food Stamps, child support payments, and other benefits, such as SSI. Figure 15 displays the TANF benefit offset:

Figure 158212Percent of TANF Benefits Loss Offset by Other Resources for Earning Groups

The data establish that 8220two-year time limit cases had income resources over and above their earnings. More than half had received child support in the past year. The group with decreased earnings had the highest percent of other resources, 58 percent had child support payments in the past year.8221⁵³

Table 5 displays sources of income after recipient exit from TANF.

Table 58212Total Post-TANF Sources of Income

*Between September 1997 and January 1998

**For those with income resource type

The study noted that, effective October 1998, Federal welfare reform also effectively will increase the collected child-support dollars that are passed on to the post-TANF family because of the PRWORA arrears distribution changes. Finally, the authors note 8220new Federal policy changes the priority in which child support collections are dispersed such that arrears owed to the post-TANF family will be paid before the State is reimbursed for past public assistance grants paid to the family. Based on current collection data, this could almost double the child support dollars passed on to the family.8221⁵⁴

Wisconsin

Wisconsin is currently examining the effects of child support on welfare exits and recidivism in its welfare program, W-2. The child support component of W-2 is operating as a waiver demonstration program, with an evaluation based on the random assignment of a subset of participants to experimental and control groups. Experimental group participants are allowed to keep all child support paid on their behalf. Members of the control group may keep only up to $50 a month or 41 percent, whichever is greater. This new policy is expected to lead to higher child-support receipt by Custodial Parents.

According to the most recent W-2 preliminary report issued in March 1999⁵⁵, across the sample as a whole, there are no significant differences between the experimental and control groups in the likelihood or amount of child support payment. However, there are significant differences among some subgroups. For example, among divorced resident parents, there is a 12.5 percent greater likelihood of payment between the experimental group and the control group, and the mean payment is higher. Among combined divorce/paternity cases, the mean (positive) payment is significantly higher among the experimental group. Among urban county cases, mean (positive) payments are $35 higher among the experimental group. Finally, Custodial Parents in the experimental group received significantly more child support than parents in the control group. The experimental group received an average $46 more per month than those in the control group. Further, Custodial Parents in the experimental group who were initially assigned to a lower W-2 tier were more likely to have left W-2 at the end of the fifth quarter than were control group parents assigned to a lower tier.⁵⁶ Twelve percent of the experimental group moved to one of the higher tiers and 52 percent moved off W-2 completely. Overall, experimental-group cases are more likely to move off W-2 than control group cases.

Cumulative totals further show that 44 percent of Custodial Parents in the experimental group had at least some child support paid on their behalf. This is higher than the amount for any given quarter as it covers a longer period. Among those cases in which support was paid, the average positive monthly payment was $235.⁵⁷

Forty percent of Custodial Parents in the experimental group received a child support payment during the first 15 months of the demonstration. Among those who received support, the mean positive monthly amount was $194. Note that for the distribution reasons cited above, the differential between paid and received is $41 per month. Custodial Parents in the control group receive an average of $46 less per month when a payment is received, and receive an average of $13 per month less overall⁵⁸.

For purposes of this study, these disparities are particularly important. The Wisconsin report notes that:

8220One possible reason that the proportion of Custodial Parents in the experimental group who receive support is lower than the proportion on whose behalf support is paid stems from the treatment of payments in cases with percent-expressed orders. Due to the method of handling payments for cases with percentage-expressed orders, support payments for cases in which arrears are owed are sometimes applied to reimburse the State for arrears before going to cover current orders.8221⁵⁹

The report goes on to say that:

8220Another case in which resident parents in the experimental group could have a payment made on their behalf without actually receiving a payment is when there is no current order for support and the payment was made for arrears owed to the State.8221

IX.160Data From Custodial-Parent Surveys

This section provides information about surveys of Custodial Parents.

Wisconsin conducted a survey of families who left the family welfare program63719either AFDC or W-263719between January and March 1998 and did not return during the next six to nine months.⁶⁰ The interviews were conducted between August and November 1998, with a random sample of cases that left welfare during the first quarter of the year. A total of 375 telephone and in-person interviews were conducted. Data on the leavers8217 characteristics were obtained from the Wisconsin Management Information System CARES database.

For purposes of this study, the most important finding was that 27 percent of leavers received child support. While 48 percent of respondents indicated they agree or agree strongly that 8220they have more money now than when on welfare⁶¹,8221 three significant events occurred more frequently after welfare than when on welfare, as depicted by Table 6:

Table 663719Comparison of Welfare-/Post Welfare-Recipient Financial Status

When asked, 8220What do you say are the reasons you no longer get welfare?8221, 54 percent of respondents indicated 8220employment-related8221 and only 2 percent of respondents indicated 8220Assets/Child Support Exceed Limits.8221⁶²

In the report Making Ends Meet, the authors interviewed 379 low-income single mothers from four cities: Boston, Chicago, San Antonio, and Charleston, South Carolina.⁶³ One-third of the welfare mothers interviewed reported receiving child support from non custodial fathers during the prior year. This is nearly three times what national child support data indicate. The discrepancy was due largely to the fact that only a third of the contributions made were made through the child support enforcement system. The recipients indicated they preferred the 8220covert support8221 due to the fact that Federal rules reduced the amount of payments going to the family.⁶⁴

Although during the time period the interviews were conducted the $50 pass-through requirement was in effect, the benefits of the pass-through were mitigated substantially because Food Stamps and Federal housing programs counted the child support income when determining benefits. According to the authors, a $50 child support pass-through caused a $15 a month loss in Food Stamps and, for families living in subsidized housing, an additional cost of $15 in rent. Thus, even though a father might have paid hundreds of dollars in child support, the family8217s net income rose by only $20 per month.⁶⁵

Through the interviews, the authors learned that overall, mothers on welfare who received child support received payments averaging $121 a month. Those who received payments through the formal system averaged only $46 a month while those who received covert support received an average of $141 a month.⁶⁶

Those fathers aware that only $50 was passed through to their families 8220often felt that any participation in the formal system was foolish.8221 Conversely, it is possible they believed that 8220covert contributions over $50 were generous.8221 When mothers left the welfare rolls to employment, the financial disincentives no longer applied, and 8220absent father may have felt their contributions counted more.8221⁶⁷ This view was reinforced by the fact that some mothers indicated the fathers were more willing to contribute if the mothers were working. In fact, 42 percent of the working mothers interviewed indicated they had received contributions from the fathers during the previous year compared to only one-third of the mothers on welfare. The authors posit that, in fact, it may have worked the other way. 8220That is, if the children8217s father provided regular support, it paid for the mother to work because she could keep all of the money he gave them.8221⁶⁸

The financial disadvantages of payment through the formal child support system were compounded by the fact that some mothers believed that direct payment from the father to the family enhanced the father/child relationship. 8220For example, a father who maintained some contact would know when his child needed new shoes or clothing or money for a school trip. According to mothers, arranging child support through the courts diminished the absent father8217s of parental pride and obliterated any feeling of gratification for doing his duty.8221⁶⁹

For many of the mothers interviewed, child support was the single most important factor that allowed them to work at a low paying job. They indicated that child support was necessary for them to make ends meet.⁷⁰

X.160Research and Study Findings

PRWORA Child Support Payment Distribution Policy Changes Have Increased the Amount of Child Support Paid to Families That Formerly Received Public Assistance.

Overall, the financial impact of the PRWORA changes in distribution policy has been positive for former-assistance families. However, the magnitude of this impact is not consistent across all States nationwide. Computer models employing historical financial data from State child-support computer systems reveal distribution outcomes that differ substantially based upon past State distribution policy. In States that, prior to PRWORA, sought to maximize State reimbursement for past assistance ahead of Custodial Parent-Owed Arrears, families will experience a 25 percent increase in child-support income following PRWORA implementation. Further, for those States that transitioned to PRWORA using a two-step approach, the bulk of this increase occurs during the first step that began on October 1, 1997. Conversely, in States that historically prioritized Custodial Parent-Owed Arrears first, family income initially will remain unchanged during the first step and then later decrease slightly under full PRWORA implementation. This later decrease occurs because a portion of the Custodial-Parent debt can be satisfied only after all State-Owed Arrears have been fully paid off. Overall, the financial gains for families outweighed the losses.

This Increase in Child-Support Income Reduces the Likelihood of Welfare Reentry.

The 1994 Wisconsin study did find that receipt of any amount of child-support has a large effect on reducing the likelihood of reentry into welfare. This finding is substantiated by data from Washington State, that child support payments are associated with lower recidivism rates. However, the Washington State data separates child-support payments into good and poor payments, and found only good payments acted to prevent reentry.

Texas also found that child support strongly reduced welfare recidivism. While 35 percent of welfare leavers returned to the AFDC rolls within a year, every $100 of child support received in a quarter reduced the probability of welfare reentry by 1 percentage point. However, the fact that Texas is a low-benefit State makes it difficult to discern whether the reduced reentry effects were due to failure to meet AFDC eligibility income tests, and the study did not address the issue of those who attempted reentry but were ineligible due to low income thresholds.

Virginia data show that a significant percentage of families that remain off welfare receive child support. Forty-eight percent of all post-TANF families received child-support payments in the last year of the study.

Based on the research findings, even small amounts of child-support payments reduce welfare recidivism. Thus, insofar as the new distribution policies provide higher child-support distributions to families, the changes will prevent some welfare recidivism. Further, the importance of child support for families who had exited welfare was reinforced anecdotally by interviews where mothers indicated they depended on child-support payments to make ends meet. This also suggests that distribution policies that increase payments to post-welfare families will contribute to a lower rate of welfare recidivism.

In addition, as the residual TANF caseload decreases over time, it increasingly comprises hard-to-place individuals who face substantial barriers to employment. Based on the anticipated low-earning power of low-skilled women, the relative value of even incremental increases in child support will be greater for the hard to place. In addition, if these individuals reside in a low-benefit State, as explained above, the relative replacement value for the TANF grant will be greater. This combination of factors suggests that PRWORA distribution policies that increase child-support payments to these families may have an even greater effect on welfare exits.

Increasing the Amount of Child Support That Is Passed Through to Families on Public Assistance and Disregarded for the Purposes of TANF Benefits Also Results in Increased Child-Support Collections.

Based on the In Making Ends Meet Custodial-Parent surveys, it is clear that, as long as a family remains on welfare and welfare benefits are reduced by the amount of child support received, many Non Custodial Parents will continue to pay, and many Custodial Parents will continue to take, child support 8220under the table.8221 To some extent, the increase in paternity establishments, the New Hire Directory, tax offsets, license revocations, and other new CSE tools will mitigate this fact, but not completely eliminate 8220informal8221 child support payments.

For this reason, it is difficult to accurately measure the effect of child support payments on welfare exit. Studies that rely on administrative matches of government records of welfare leavers with records of child- support payments will fail to measure the effect of the informal child-support market on welfare exits.

Because of informal child-support payments, survey data are needed to capture the full effect of child-support payments on keeping families from returning to welfare. We believe that, when informal child support is included, child support has an even greater effect on preventing welfare reentry. Further, the anecdotal information provided in Making Ends Meets suggests that Non Custodial Parents may be more likely to pay child support when they know that the payments are going directly to benefit the family.

Assuming this is correct, child support pass-throughs may increase the amount of child support collected. However, it is important to point out that even if pass-throughs do increase child-support collection, this benefit should be considered in light of the cost of providing a full pass-through as well as other factors. Specifically, the potential for a full pass-through keeping families eligible for welfare longer may delay some exiting. At the same time, TANF is time limited, which ensures that even a full pass-through will only delay and not eliminate such exits.

In Wisconsin, resident parents receiving a full pass-through received $34 more per month than those in the control group. Although this difference is small, it is important to note that in the Wisconsin demonstration, while the experimental group receives a 100 percent pass-through, the control group received a 40 percent pass-through. Thus the disparity could be expected to increase if comparing a 100 percent pass-through to no pass-through.

Receipt of Child Support Enhances the Likelihood of Welfare Exits.

Based on the 1994 Wisconsin IRP study, which established that only families receiving large child support benefits of $300 or more a month in Wisconsin are more likely to exit welfare, we know that the child-support payment must represent a significant portion of the welfare benefit to affect welfare exits. However, this does not mean that $300 a month in child-support is needed in every State. We can speculate that the higher the child support received by the Custodial Parent as a proportion of the TANF benefit, the more likely it will produce welfare exit. This means that the potential for child support to increase welfare exits depends not only on the amount of child support received but also on the percentage of welfare benefit that this child support replaces. This means that the same child-support benefit would be more likely to produce welfare exit in a low-benefit State than in a high-benefit State.

In low-benefit States, additional factors also play a role in connecting receipt of even small amounts of child support with welfare exit. TANF eligibility income thresholds are lower in low benefit States. Therefore, receipt of even a low amount of child-support is more likely to make a family ineligible for TANF relative to families in higher benefit States. This has the effect of decreasing entries due to failure to meet income eligibility tests and increasing exits from welfare where child support payments cause the family to exceed income limitations.

Data from Washington State, a high TANF benefit State, supports the finding that child-support payments did not appear to produce welfare exits. The Washington State study did not provide data broken down by amounts of child support, so it cannot be used to test the Wisconsin theory that in high benefit States, only child support in large amounts would produce welfare exits.

The results of the Texas study are consistent with the speculation that child support payments will have a greater effect on welfare exits in low benefit States. In Texas, receipt of child support accounted for 16 percent of the total exits from AFDC. Notably, receipt of child support influenced exits from welfare more than receipt of earnings, although this finding is mitigated to some extent by the fact that gross, not net, earnings were used in this measure.

Implications for Future Study

Study outcomes indicate that similar8212perhaps even slightly more positive8212gains could be achieved for former-assistance families through a child support arrears distribution policy that simply pays all Custodial Parent-Owed Arrears ahead of State-Owed Arrears. The model further indicates that loss in State child support arrears-related income would be less than 6 percent (or $8.03 per month for each case) overall. This simple, more straightforward approach would accomplish a more family-friendly outcome that Congress intended and, at the same time, ease existing implementation difficulties for the States.

We recommend further study of the implications surrounding PRWORA8217s distribution policy. Follow-up activities could include establishing the positive effect on Custodial-Parent behavior, the cost and difficulty of implementing PRWORA changes within State child support enforcement systems, and a more precise cost estimate for this recommended policy simplification.

Attachment 163719OCSE AT 97-17

ACTION TRANSMITTAL

OCSE-AT-97-17

October 21, 1997

David Gray Ross/s/

Deputy Director

Office of Child Support Enforcement

TABLE OF CONTENTS

Distribution Requirements

1. Determination of the required monthly child support obligation1601601606

160a.160Child support obligation document

160b.160Frequencies other than monthly

160c.160Voluntary payments

160d.160Rounding to whole dollar amounts

160e.160Date of collection

160f.160Future payments

160g.160Gap payments

160h.160Assigned medical support collections

160

II.160Reporting collections to the IV-A agency in current assistance cases 1601609

III.160Definitions 1601601601601601601601601601609

IV. 160Assignment of support rights16016016016016016016016011

V. 160Distribution of collections through Federal income tax refund offset 16016013

VI. 160Distribution of collections (except for Federal income

160tax refund offsets) 16016016016016016016016016014

160160

VII. Hold harmless provision 16016016016016016016016016018

VIII.Distribution of collections in title IV-E foster care cases 16016016016018

Questions and Answers

Assignment of rights to support.16016016016016016016016019

Distribution in current assistance cases.16016016016016016016019

Federal share of collections.16016016016016016016016016024

Existing regulations at 45 CFR 302.51 and 302.52 former

section 457 of the Act.16016016016016016016016016024

Federal and State income tax refund offset and Federal

Administrative Offset.16016016016016016016016016016025

Distribution of Arrearages in former assistance cases.16016016016016027

Gap payments.16016016016016016016016016016016028

Future payments.16016016016016016016016016016030

Federal reporting requirements.16016016016016016016016031

Other questions.16016016016016016016016016016031

Systems-related questions.16016016016016016016016016032

Appendix: Case Scenarios

INSTRUCTIONS FOR THE DISTRIBUTION OF CHILD SUPPORT COLLECTIONS PURSUANT TO AN APPROVED IV-D STATE PLAN

I.160Determination of the required monthly child support obligation

160

160(a) 160Child Support Obligation Document Before any distribution of amounts of child support collected can be made, an amount which represents the required child support obligation for one month must be determined. In all cases, the amount must be ascertained from the document, which established the support obligation -- a court or administrative order or other legal process established under State law. See 45 CFR 302.50 and 302.56. In most cases, the amount is taken directly from the support order. For example, if a support order states that the Non Custodial Parent shall pay $500 per month in child support, $500 is the amount which represents the required monthly child support obligation.

160(b) 160Frequencies other than Monthly Child support payments are often ordered to be paid in frequencies other than monthly and must be converted to an amount which represents the required monthly child support obligation. If the State has an existing system for such conversions, such a system may be used or, as an alternative, the State may use one of the following conversion systems:

160160160160160Example 1

160160If the support obligation is to be paid weekly, such as $20 per week, multiply by 4.345 to get the required monthly child support obligation. Some States have existing systems, which use 4-1/3 as the multiplier; 4-1/3 is also acceptable, but slightly less accurate.

160160160

160160160$20 x 4.345 = $86.90 per month160

160160160160160Example 2

160160If the support obligation requires a weekly payment of $20 each Monday and there are 4 Mondays in a month, the current monthly obligation equals $80. However, in months in which there are 5 Mondays, the current monthly obligation equals $100.

160160160$20 x 4 Mondays = $80 for the month

160160160$20 x 5 Mondays = $100 for the month

160160

160160Similarly, if the support obligation requires bi-weekly (every 2 weeks) payments, there will be some months in which 3 payments are due to satisfy the current monthly obligation. If the order requires a bi-weekly payment of $100 and there are two charge days in the month, the current monthly obligation equals $200. However, if there are three charge days in the month, the current monthly obligation will equal $300.

160160160$100 x 2 = $200 for the month

160160160$100 x 3 = $300 for the month160

160160160160160Example 3

160160If the support obligation is to be paid every two weeks, such as $25 every two weeks, multiply by 2.173 (2 1/6 is also acceptable) to get the required monthly child support obligation.

160160160160$25 x 2.173 = $54.33 per month

160

160160160160160Example 4

160160If the amount is to be paid in some other frequency, such as $10 every 8 days, the conversion can be made by expressing the order or agreement in days, multiplying by 365, and dividing by 12.

160160160 160$10 x 365 days = $38.02 per month

160160160 8 days 12 months

160160The conversion can also be made by multiplying the amount required by the number of payments required in a year and dividing by 12.

160160160160

160(c)160Voluntary payments. In the absence of a support order, any amounts received as voluntary payments shall be treated as the obligation for the current month in which they were received. (see PIQ-81-9)

160(d)160Rounding to Whole Dollar Amounts Once the required monthly child support obligation has been determined, the amount may, at the discretion of the IV-D agency, be rounded to a whole dollar amount. The amount computed to be paid to the family, retained by the State, or reimbursed to the Federal Government may also be rounded. See 45 CFR 302.51(a)(2).

160(e) 160Date of Collection The date of collection for amounts collected and distributed under title IV-D of the Act is the date of receipt by the State disbursement unit established under section 454B of the Act, except that if current support is withheld by an employer in the month when due and is received by the State disbursement unit in a month other than the month when due, the date of withholding may be deemed to be the date of collection.

160(f)160 Future payments. Under 45 CFR 302.51(c), if, in a current assistance case, an amount collected as support represents payment on the required support obligation for future months, the amount shall be applied to such future months. However, no such amounts shall be applied to future months unless amounts have been collected which fully satisfy the support obligation assigned under section 408(a)(3) of the Act for the current month and all past months.

160160160160160160Example 5

160160Assume there are no arrearages in a current assistance case and the required monthly child support obligation is $400. If more than $400 is received and designated by the payer as a future payment, the excess should be applied to such future month. If there is no such designation, a State may determine if a payment represents a future payment.

160(g) 160Gap payments under section 457(e) of the Act and former 402(a)(28) of the Act.

160160Former section 402(a)(28) of the Act provided for the payment to the family of child support collected by the State as a protection against reduction in the total income available to a family in a month. This section only applied to those States whose State title IV-A plan, in July, 1975, permitted a portion of the monthly child support payment after application of appropriate disregards to be retained by a family receiving AFDC without causing a dollar-for-dollar reduction in the AFDC payment made to the family.

160160Under section 457(e) of the Act, at State option, section 457 of the Act does not apply to any amount collected on behalf of a family as support by the State (and paid to the family in addition to the amount of assistance otherwise payable to the family) if such amount would have been paid to the family by the State under section 402(a)(28) of the Act, as in effect and applied on August 21, 1996.

160

160(h) 160Assigned medical support collections.

160160Any amounts collected by the IV-D agency which represent specific dollar amounts designated in the support order for medical purposes that have been assigned to the State under 42 CFR 433.146 shall be forwarded to the Medicaid agency for distribution under 42 CFR 433.154. See 45 CFR 302.51(e)(1).

160160When a family ceases receiving assistance under the State's title XIX (Medicaid) plan, the assignment of medical support rights under section 1912 of the Act terminates, except for the amount of any unpaid medical support obligation that has accrued under such assignment. The IV-D agency shall attempt to collect any unpaid specific dollar amounts designated in the support order for medical support purposes. Under this requirement, any medical support collection made by the IV-D agency shall be forwarded to the Medicaid agency for distribution under 42 CFR 433.154. See 45 CFR 302.51(e)(2).

II.160Reporting collections to IV-A agency in current assistance cases.

160(a)160In accordance with 45 CFR 302.32(b), the IV-D agency must inform the State's IV-A agency of the amount of the collections which represents payment on the required support obligation for the month as determined in section 302.51(a) within 10 working days of the end of the month in which the support is received by the IV-D agency responsible for final distribution. This amount is not necessarily the total amount of the collection; it is only that portion of the collection, which represents payment on the support obligation for that month.

160(b)160Method of Informing. There is no mandated standard procedure for the IV-D agency to inform the State IV-A agency. The State may adopt any procedure it finds best fits both agencies' individual requirements so long as the IV-A agency can easily use the information and properly determine eligibility based upon the amounts collected which represent payment on the child support obligation for the current month.

III.160Definitions used in this Action Transmittal.

160As used in this document:

160(a) 160ASSISTANCE.--The term `assistance from the State' means assistance under the State program funded under title IV-A of the Act (see also TANF-ACF-PA-97-1, dated January 31, 1997) or under the State plan approved under title IV-A (as in effect on August 21, 1996).

160(b)160FEDERAL SHARE.--The term `Federal share' means that portion of the amount collected resulting from the application of the Federal medical assistance percentage in effect for the fiscal year in which the amount is distributed.

160

160(c) 160FEDERAL MEDICAL ASSISTANCE PERCENTAGE.--The term `Federal medical assistance percentage' means--

160

160(1) 16075 percent, in the case of Puerto Rico, the Virgin Islands, Guam, and American Samoa; or

160(2) 160the Federal medical assistance percentage (as defined in section 1905(b) of the Act, as such section was in effect on September 30, 1995) in the case of any other State.

160(d) 160STATE SHARE.--The term `State share' means 100 percent of the amount collected which does not exceed the cumulative unreimbursed assistance, minus the Federal share.

160(e) 160CURRENT ASSISTANCE CASE. -- The term "current assistance case" means any IV-D case, which is currently receiving assistance.

160(f) 160FORMER ASSISTANCE CASE. -- The term "former assistance case" means any IV-D case, which formerly received assistance. 160

160

160(g) 160NEVER-ASSISTANCE CASE. -- The term "never-assistance case" means any IV-D case, which has never received assistance.

160(h)160PERMANENTLY-ASSIGNED ARREARAGES.--The term "permanently-assigned arrearages" means those arrearages which do not exceed the cumulative amount of unreimbursed assistance paid to the family as of the date the family leaves the assistance rolls: (1) which are or were assigned under an assignment of support rights in effect on September 30, 1997, and (2) which accrued under an assignment entered into on or after October 1, 1997, while a family is receiving assistance.

160(i)160TEMPORARILY-ASSIGNED ARREARAGES -- The term "temporarily-assigned arrearages" means those arrearages which do not exceed the cumulative amount of unreimbursed assistance paid to the family as of the date the family leaves the assistance rolls, which accrued prior to the family receiving assistance and which were assigned to the State after September 30, 1997. These arrearages are not permanently assigned and the temporary assignment will expire when the family leaves the assistance program or on October 1, 2000, whichever date is later.

160(j)160CONDITIONALLY-ASSIGNED ARREARAGES -- The term "conditionally-assigned arrearages" means those arrearages which do not exceed the cumulative amount of unreimbursed assistance paid to the family as of the date the family leaves the assistance rolls and which are owed to the family unless they are collected through Federal income tax refund offset. They are arrearages, which were temporarily assigned to the State and became conditionally assigned to the State when the temporary assignment expired. If a conditionally-assigned arrearage is collected through a Federal income tax refund offset, the collection is retained by the State to reimburse the State and the Federal government up to the cumulative amount of unreimbursed assistance paid to the family. Collections of conditionally-assigned arrearages by any other enforcement mechanism are paid to the family.

160(k)160NEVER-ASSIGNED ARREARAGES -- The term "never-assigned arrearages" means all arrearages in never-assistance cases, and, in former assistance cases, means those arrearages that accrue after the family's most recent period of assistance ends.

160(l)160UNASSIGNED DURING-ASSISTANCE ARREARAGES -- The term "unassigned during-assistance arrearages" means all previously-assigned arrearages which exceed the cumulative amount of unreimbursed assistance when the family leaves the assistance program and which accrued during the receipt of assistance.

160(m) 160UNASSIGNED PRE-ASSISTANCE ARREARAGES: -- The term "unassigned pre-assistance arrearages" means all previously-assigned arrearages which exceed the cumulative amount of unreimbursed assistance when the family leaves the assistance program and which accrued prior to the receipt of assistance.

160(n)160UNREIMBURSED ASSISTANCE. -- The term "unreimbursed assistance" means the cumulative amount of assistance paid to a family for all months which has not been repaid by assigned support collections. The total amount of unreimbursed assistance a State may recover through the IV-D program is limited by the total amount of the assigned support obligation.

IV. Assignment of support rights under sec. 408(a)(3) of the Act

160(a) 160In General.--As a condition of eligibility for assistance under title IV-A of the Act, a member of a family must assign to the State any rights a family member may have (on behalf of the family member or of any other person for whom the family member has applied for or is receiving such assistance) to support from any other person, not exceeding the total amount of assistance paid to the family, which accrue (or have accrued) before the date the family leaves the program.

160160All support collected pursuant to such assignment while the family is currently receiving IV-A assistance will be retained to reimburse the cumulative amount of assistance which has been paid to the family, unless a State opts to pay all or a portion of the collection to the family.

160160After the family leaves the IV-A assistance program, accrued arrearages remain assigned in accordance with the following rules:

160

160(1)160For an assignment entered into prior to October 1, 1997, the applicant assigns to the State all rights to support which have previously accrued and which will accrue prior to the family leaving assistance.

160(2)160For an assignment entered into on or after October 1, 1997, and before October 1, 2000, --

160160(A)160with respect to collections by Federal income tax refund offset, the applicant assigns to the State all rights to support which have previously accrued and which will accrue prior to the family leaving assistance.

160160(B)160with respect to collections by other than Federal income tax refund offset,

160160160(I) the applicant assigns to the State all rights to support which will accrue while the family is receiving assistance; and

160160160(II) the applicant temporarily assigns to the State all rights to support, which accrued prior to the family receiving assistance, until October 1, 2000, or such time that the family leaves assistance, whichever date is later.

160(3)160For an assignment entered into on or after October 1, 2000, --

160160(A)160with respect to collections by Federal income tax refund offset, the applicant assigns to the State rights to support which have previously accrued and which will accrue prior to the family leaving assistance.

160160(B)160with respect to collection by other than Federal income tax refund offset,

160160160(I) the applicant assigns to the State all rights to support which will accrue while the family is receiving assistance; and

160160160(II) the applicant temporarily assigns to the State all rights to support, which accrued prior to the family receiving assistance, until the family leaves assistance.

160(4)160If a State elects to implement on October 1, 1998, distribution requirements for former assistance cases that would otherwise be effective October 1, 2000,

160160(A) 160For an assignment entered into prior to October 1, 1998, the assignment rules under section (IV)(a)(1) of the Action Transmittal apply.

160160(B) 160For an assignment entered into on or after October 1, 1998, the assignment rules under section (IV)(a)(3) of this Action Transmittal apply.

160(b) Limitations

160(1) 160A State may not require, as a condition of providing assistance to any family, that a member of the family assign to the State any rights to support described in paragraph (a) which will accrue after the date the family leaves the program.

160(2) 160The cumulative amount of assigned arrearages in former assistance cases may not exceed the cumulative amount of unreimbursed assistance paid to the family under all assignments.

160160160

V.160Distribution of collections through Federal income tax refund offset under section 457(a)(2)(B)(iv) of the Act

160(a)160General rule. Notwithstanding any other provision of section 457 of the Act, any amount of support collected pursuant to section 464 of the Act shall be retained by the State to the extent past-due support has been assigned to the State as a condition of receiving assistance from the State, up to the amount necessary to reimburse the State for cumulative amounts paid to the family as assistance by the State.

160160The State shall pay to the Federal Government the Federal share of the amounts so retained.

160160To the extent the amount collected pursuant to section 464 of the Act exceeds the amount required to be retained, the State shall pay the excess to the family.

160(b)160Current Assistance Cases. Support collections through Federal income tax refund offsets in current assistance cases are retained by the State up to the cumulative amount of unreimbursed assistance paid to the family. The Federal statute does not specify the order in which collections are applied to satisfy assigned arrearages. The State must have procedures, which specify the order in which assigned arrearages will be satisfied. Collections over and above the cumulative amount of unreimbursed assistance are paid to the family.

160(c)160Former Assistance Cases. Support collections made through Federal income tax refund offsets in former assistance cases must first be applied to assigned arrearages. This includes temporarily-assigned and conditionally-assigned arrearages, as defined. These collections must be retained by the State up to the cumulative amount of unreimbursed assistance paid to the family. The Federal statute does not specify the order in which collections are applied to satisfy assigned arrearages. The State must have procedures, which specify the order in which assigned arrearages will be satisfied. Collections over and above the cumulative amount of unreimbursed assistance are paid to the family.

160(d)160Never-assistance Cases. Support collections through Federal income tax refund offsets in never-assistance cases are paid to the family.

160

VI.160Distribution of collections (except for Federal income tax refund offsets)

160(a) In General.

160In accordance with 45 CFR 302.51(a)(1), for purposes of distribution in a IV-D case, amounts collected, except for amounts collected through Federal income tax refund offset, must be treated first as payment on the required support obligation for the month in which the support was collected and if any amounts are collected which are in excess of such amount, these excess amounts shall be treated as amounts which represent payment on the required support obligation for previous months.

160

160The requirement to apply collections first to satisfy the current support obligation is critical in all IV-D cases to ensure that payment records are consistent in interstate cases, regardless of whether the amount applied to current support is paid to the family (as in a former assistance case) or retained by the State to reimburse unreimbursed assistance in a current assistance case.

160(b) Current Assistance Cases.

160The State shall (not exceeding the cumulative amount of unreimbursed assistance paid to the family):

160(1) 160First, pay to the Federal Government the Federal share of the entire amount collected;

160160

160(2) 160Second, retain, or distribute to the family, the State share of the amount collected; and

160(3) 160Third, reduce the cumulative amount of unreimbursed assistance by the total amount collected and disbursed under (1) and (2), and distribute collections exceeding the cumulative amount of unreimbursed assistance to the family.

160The Federal statute does not specify the order in which collections are applied to satisfy assigned arrearages in current assistance cases. The State must have procedures, which specify the order in which assigned arrearages will be satisfied.

160

160(c) Former Assistance Cases.

160(1)160For collections made prior to October 1, 1997 (other than through Federal income tax refund offset), the State must:

160160(A) First, distribute the amount collected to satisfy the current monthly support obligation and pay that amount to the family.

160160(B) Second, distribute any amount above the current monthly support obligation to satisfy arrearages owed to the family or assigned to the State (see 45 CFR 302.32(f)(ii). The Federal statute does not specify the order in which collections are applied to satisfy arrearages. The State must have procedures, which specify the order in which assigned arrearages will be satisfied. If the State distributes any amount to assigned arrearages, the State must pay to the Federal Government the Federal share of the amount so collected and must retain the State share of the amount so collected. The State must retain the State share of the amount so collected, with one exception. The State may retain or pay to the family the State share of collections applied to arrearages which accrued while the family was receiving assistance after October 1, 1996, in accordance with section 457(a)(1) and (2)(B)(iii) of the Act and section VI(b)(2) of this Action Transmittal.

160(2)160For collections made on or after October 1, 1997 and before October 1, 2000, or earlier at State option, (other than collections through Federal income tax refund offset), the State must:

160160(A) First, distribute the amount collected to satisfy the current monthly support obligation and pay that amount to the family;

160160(B) Second, distribute any amount above the current monthly support obligation to satisfy never-assigned arrearages and pay that amount to the family;

160160(C) Third, distribute any amount above amounts distributed in (A) and (B) to satisfy arrearages owed to the family or assigned to the State (see 45 CFR 302.32(f)(ii). The Federal statute does not specify the order in which collections are applied to satisfy arrearages. The State must have procedures, which specify the order in which assigned arrearages will be satisfied. If the State distributes any amount to assigned arrearages, the State must pay to the Federal Government the Federal share of the amount so collected. The State must retain the State share of the amount so collected, with one exception. The State may retain or pay to the family the State share of collections applied to arrearages which accrued while the family was receiving assistance after October 1, 1996, in accordance with section 457(a)(1) and (2)(B)(iii) of the Act and section VI(b)(2) of this Action Transmittal.

160160

160(3)160For collections made on or after October 1, 2000, or earlier at State option (other than collections through Federal income tax refund offset), the State must:

160160160

160160(A) First, distribute the amount collected to satisfy the current monthly support obligation and pay that amount to the family;

160160(B) Second, distribute any amount above the current monthly support obligation to satisfy never-assigned arrearages and pay that amount to the family;

160160(C) Third, distribute any amount above amounts distributed in (A) and (B) to satisfy unassigned pre-assistance arrearages and conditionally-assigned arrearages and pay that amount to the family. The Federal statute does not specify the order in which collections are applied to satisfy these arrearages. The State must have procedures, which specify the order in which assigned arrearages will be satisfied. If there are unassigned, previously permanently-assigned arrearages which were assigned under former section 402(a)(26) of the Act and the State cannot determine whether they were pre-assistance or during-assistance arrearages, those unassigned arrearages must be paid to the family.

160160(D) Fourth, distribute any amount above amounts distributed in (A), (B) and (C) to satisfy permanently-assigned arrearages. The State must pay the Federal Government the Federal share of the amount so collected. The State must retain the State share of the amount so collected with one exception. The State may retain or pay to the family the State share of collections applied to arrearages which accrued while the family was receiving assistance after October 1, 1996, in accordance with section 457(a)(1) and (2)(B)(iii) of the Act, and section VI(b)(2) of this Action Transmittal.

160160(E) Fifth, reduce the cumulative amount of unreimbursed assistance by the total amount distributed under (D), distribute collections exceeding the cumulative amount of unreimbursed assistance to satisfy unassigned during-assistance arrearages and pay those amounts to the family.

160(d)160State option for October 1, 1998 effective date for distribution rules in former assistance cases

160(1) 160Notwithstanding any other requirement for distribution of collections in IV-D cases, a State may elect to apply on and after October 1, 1998, the distribution rules that would be effective in former assistance cases for collections made on or after October 1, 2000. These rules are set forth in section VI(c)(3) above.

160(2)160If a State elects the option described in section (d)(1) above, the requirements of section 457 (other than section 457(b)(1)) of the Act, as in effect and applied on August 21, 1996), shall apply to amounts collected before October 1, 1998. These rules are set forth in section VI(c)(1) above.

160(3)160A State must indicate which effective dates will apply in the State by submittal of the appropriate State plan preprint page.

160(e)160Never-assistance Cases. All support collections in never-assistance cases must be paid to the family.

VII.160Hold harmless provision under section 457(d) of the Act

160(a)160General rule. If the amounts collected which could be retained by the State in the fiscal year (to the extent necessary to reimburse the State for amounts paid to families as assistance by the State) are less than the State share of the amounts collected in fiscal year 1995 (determined in accordance with section 457 as in effect on August 21, 1996), the State share for the fiscal year shall be an amount equal to the State share in fiscal year 1995.

160160Amounts collected which could be retained by the State are those collections that would be retained by the State if the State retained the entire amount of the State share of collections (rather than paying any of the State share to the assistance family) and the State implements each provision of section 457 on the date specified in the statute (rather than earlier, as allowed by the statute.)

160(b)160Gap payment States. For purposes of (a), the State share of any gap payments to the family shall be considered amounts which could be retained by the State if such payments were reported by the State as part of the State share of amounts collected in fiscal year 1995.

VIII.Distribution of collections in title IV-E foster care cases

160Notwithstanding the preceding provisions in this Action Transmittal, amounts collected as support in a title IV-E foster care cases must be distributed in accordance with 45 CFR 302.52.

ASSIGNMENT OF RIGHTS TO SUPPORT

QUESTION 1: Does the assignment under section 408(a)(3) of the Act cover a child who is receiving Supplemental Security Income (SSI)?

ANSWER 1: Whether or not the assignment covers a child receiving SSI depends upon whether the child is included in the State's definition of an assistance unit.

DISTRIBUTION IN CURRENT ASSISTANCE CASES,

INCLUDING $50 PASS-THROUGH AND DISREGARD

QUESTION 2: Are States required to continue to pass through to families receiving assistance the first $50 of support collected after September 30, 1996?

ANSWER 2: Under title I of PRWORA, if a State chose not to implement Temporary Assistance to Needy Families sooner, it had to continue to disregard for AFDC eligibility purposes until July 1, 1997, the first $50 of any child support passed through to a family receiving assistance. However, under title III of the law, effective on or after October 1, 1996, States are not required to pass any child support collections through to families receiving assistance. Rather, States must pay the Federal government its share of the entire child support collection. The State may retain, or pass through to the family, the State's share of the collection. The Federal mandate to pass through to the family the first $50 of support collected was eliminated after September 30, 1996. If the State provides no pass-through, there is nothing to disregard. Any support passed-through, whether required by State law or by State option, must come entirely from the State's share of the collection.

QUESTION 3: Were States required, until June 30, 1997, to disregard the first $50 of any support passed-through to a family receiving assistance?

ANSWER 3: The disregard of the first $50 of any support passed through to a family receiving assistance was mandatory through June 30, 1997, in any State that did not implement TANF until July 1, 1997, in accordance with title IV-A of the Act, as in effect prior to enactment of PRWORA.

QUESTION 4: Do any IV-D regulations or requirements apply to a State which continues paying the former "$50 pass-through" to TANF recipients entirely out of State funds?

ANSWER 4: Yes. Any amount collected on behalf of a family as support by a State pursuant to a State IV-D plan must be distributed in accordance with section 457 of the Act, as revised by section 302 of PRWORA, and in accordance with any aspect of 45 CFR 302.51 which is not inconsistent with the revised section 457.

QUESTION 5: Is there a State option allowed under PRWORA for the State to continue passing the $50 (or even a higher amount at State option) through to the current TANF recipient and disregarding this amount in calculating the grant amount for the recipient?

ANSWER 5: Under section 457(a)(1) of the Act, as revised by section 302 of PRWORA, States may continue to distribute ("pass-through") any portion of the State share of the amount collected on behalf of a family receiving assistance under title IV-A of the Act, after first paying the Federal government the Federal share of the amount collected. For eligibility purposes, the State TANF program may choose to disregard all or a part of the child support distributed to the family under '457(a)(1)(B).

QUESTION 6: The $50 pass-through mandate was eliminated effective October 1, 1996. Sometimes employers do not send child support payments made through wage withholding in a timely manner. As a result, the collections must be distributed based on the date the wages were withheld rather than the date received by the IV-D agency. How should the IV-A and IV-D agencies handle retroactive pass-through payments?

ANSWER 6: States were required to pass-through, and disregard, in assistance cases, the first $50 of support collected through wage or income withholding in the month when due prior to October 1, 1996, in accordance with 45 CFR 302.51(a)(4), under which the date of collection for distribution purposes is the date the wages or other income are withheld to meet the support obligation. If an employer withheld child support in the month the support was due (in months prior to October 1996) and subsequently paid that support to the IV-D agency in a later month (even after October 1996), the $50 pass-through must be credited for the period in which the withholding actually occurred.

With respect to collections made before October 1, 1996, but received by the IV-D agency, and distributed after October 1, 1996, the Federal government will continue to share in the cost of the $50 pass-through payment.

(Please note that section 5549 of the BBA amends section 454B to define the date of collection as the date of receipt by the State disbursement unit (SDU), except that if current support is withheld by an employer in the month when due and is received by the SDU in a month other than the month when due, the date of withholding may be deemed to be the date of collection, at the option of the State. See Q38A 21.)

QUESTION 7: How long should States continue the $50 pass-through for support withheld or collected prior to October 1, 1996 but actually received or disbursed after October 1, 1996?

ANSWER 7: There is no Federal limit on the time period during which a State must continue the $50 pass-through for support withheld or collected prior to October 1, 1996, but actually received or disbursed after that date. States must distribute any amount received after October 1, 1996, but withheld or collected prior to October 1, 1996, under the requirements that applied to the timeframe in which the support was withheld or collected. For example, if, in an AFDC case, support was withheld in September of 1996 and sent by the employer to the State but received by the State in October, the State must pass-through to the family the first $50 of support collected. New distribution requirements effective October 1, 1996, only apply to support collected as of that date forward.

QUESTION 8: Will the Federal government share in the cost of necessary modifications to a State's automated system necessary to conform distribution to the new requirements under section 457 of the Act?

ANSWER 8: Yes.

QUESTION 9: States are no longer required to pay the $50 pass-through payment after October 1, 1996. However, if they choose to continue to do so, are they still required to pay the pass-through out of current collections only?

ANSWER 9: No, States may pass through any or all of the State share of collections in current assistance cases. Section 457(a) allows States to pay all or a portion of the State share of either current or arrearage collections to the family.

QUESTION 10: If the State does not retain the State share of the collection and opts to pay all or a portion of its share to the family, must the unreimbursed assistance be reduced by the amount paid to the family?

ANSWER 10: Yes. Under section VI(b)(3) of this Action Transmittal, in a current assistance case, any amount collected, including any amount paid to the family from the State share must reduce the cumulative amount of unreimbursed assistance.

QUESTION 11: If States use funds other than child support collections to make payments to families in current assistance cases, can the programming and processing of these payments be charged as a IV-D expenditure?

ANSWER 11: No.

QUESTION 12: In a current assistance case, if a State chooses to pay the family more than the State share of the collection, how are such payments to be treated for IV-D program purposes? For example, if the benefit for the month is $40, unreimbursed assistance is $500, and the collection is $200 with a 70% Federal Medical Assistance Percentage (FMAP), the Federal share of the collection would be $140 and the State share would be $60. If the State wanted to pay the family $160, how would they account for the $100 ($160-$60) that was more than the State share?

ANSWER 12: The $100 in the example which is not part of the State share cannot be considered a distribution or disbursement of child support under the title IV-D program.

QUESTION 13: In current assistance cases with permanently-assigned arrearages and temporarily-assigned arrears, may a State apply collections to satisfy temporarily-assigned arrearages before permanently-assigned arrearages?

ANSWER 13: The Federal statute does not specify the order in which assigned arrearages must be satisfied in a current assistance case. It is, therefore, a question of interpretation of applicable State law and procedures.

QUESTION 14: Are States allowed to perform distribution at the child level when the child is included in the family assistance unit, by splitting or pro-rating the family's grant or assistance payment amount on a per-child basis and applying the collections to the unreimbursed assistance balances attributable to each child's portion of the family's grant?

ANSWER 14: No. As a condition of eligibility for assistance under title IV-A of the Act, a member of a family must assign to the State any rights a family member may have (on behalf of the family member or of any other person for whom the family member has applied for or is receiving assistance) to support from any other person, which accrue (or have accrued) before the date the family leaves the program. Therefore, in current or former assistance cases, States may not use child-level accounting by splitting or pro-rating the family grant amount on a per-child basis when the child is (or was) included in the family unit and must continue to apply collections to the cumulative amount of unreimbursed assistance balances based on the total monthly family grant amount.

QUESTION 15: In a current assistance case, is the amount of support, which the State may retain under the assignment limited to the amount of the grant for the month? For example, in a current assistance case, the support obligation for the month is $400, the grant is only $250, but there is unreimbursed assistance of $500. If the State collects the full amount of current support owed for the month, is the amount of assigned support, which it should retain $250 or $400?

ANSWER 15: The assigned support retained by the State is $400. When an individual applies for or receives assistance under the TANF program, he or she assigns ANY rights to support he or she may have, and ANY rights to support on behalf of any other individual for whom assistance is being sought or paid. If the entire support obligation for the month is collected and it exceeds the current month's assistance payment, the State must retain the excess to apply toward unreimbursed assistance, unless it elects to pay to the family all or a portion of the State share of the collection. Former section 457(b)(3), which required that monthly support amounts in excess of the amount of assistance paid to the family for the month be paid to the family, is no longer in effect.

QUESTION 16: In a current assistance case, no collection is made in a month, the obligation for the month is $400, there is a $250 assistance payment for the month but there is no other unreimbursed assistance. Is the entire $400 obligation assigned?

ANSWER 16: The entire $400 is assigned. However, the amount of assigned support collected and retained may not exceed the cumulative amount of unreimbursed assistance. If the family leaves the assistance rolls at that point, $250 remains permanently assigned and the $150 becomes unassigned during- assistance arrearages.

QUESTION 17: In a current assistance case, may a State pursue only the amount of the obligation for each particular month? For example, if the Non Custodial Parent's support obligation for the month is $200, the grant for that same month is $450, and no payment is made, may a State calculate the unreimbursed assistance for the month as the amount of the support obligation for the month, and increase the cumulative unreimbursed assistance balance by only the $200 obligation?

ANSWER 17: No. The unreimbursed assistance for the month is the amount of assistance paid to the family, not the amount of the assigned monthly support obligation amount. The cumulative amount of unreimbursed assistance a State may recover under the IV-D program, however, is limited to the total amount of the assigned support obligation.

QUESTION 18: If, in a current assistance case, at the time a support order is established the order includes a lump sum amount of support for prior periods, is that amount considered to have accrued for distribution purposes during the prior periods or at the time the order is entered? The response to this question affects distribution of the lump sum amount at the time it is collected.

ANSWER 18: Section 457 does not directly address the question of date of accrual of retroactive support awards. Therefore, it is a matter for the State to determine pursuant to its law and the terms of the support order.

FEDERAL SHARE OF COLLECTIONS

QUESTION 19: Is the Federal share of a child support collection in a current assistance case limited by the amount of assistance paid to the family in the month of collection or by the cumulative amount of unreimbursed assistance?

ANSWER 19: Distribution requirements in current assistance cases under section 457(a)(1) of the Act do not distinguish between current support collections and arrearage collections. The Federal share of assigned support is calculated on the total payment collected in the month, limited by the cumulative amount of unreimbursed assistance.

Example: In a current assistance case, a collection is received in the amount of $500. The assistance payment is $300. There are assigned arrearages and cumulative unreimbursed assistance in excess of $500. The FMAP rate is 50%. The Federal share is $250, and the State share is $250. The State may retain, or pay to the family, at State option, the State's share of the collection.

QUESTION 20: Does the State have to maintain the amount of the Federal and State share of each collection and show which payments to the family were paid out of the State's share of the collection at the case level?

ANSWER 20: The State needs to keep a record of the amount of each collection that COULD have been retained to reimburse assistance. Since the FMAP is applied to this amount to determine the Federal share, it could easily be computed if needed. This amount could be summarized for reporting purposes to simplify the computation of the Federal share of total collections. The amount of payments to the family, as always, would need to be maintained at the case level.

EXISTING REGULATIONS AT 45 CFR 302.51 AND 302.52

AND FORMER SECTION 457

QUESTION 21: Are States required to distribute child support collected on or after October 1, 1996 according to the date of collection rules at 45 CFR 302.51(a)(4)?

ANSWER 21: States must continue to use the date of collection as defined in 45 CFR 302.51(a)(4) until there is a State Disbursement Unit in the State which meets the requirements of section 454B of the Act. Section 5549 of the BBA amended Federal requirements at section 454B(c) of the Act governing State disbursement units (SDUs) to redefine the date of collection as the date of receipt by the State disbursement unit. However, if current support is withheld by an employer in the month when due and is received by the SDU in a month other than the month when due, the date of withholding may be deemed to be the date of collection at the option of the State. SDU requirements are effective October 1, 1998, unless the State qualifies for the one-year delay to continue to process the receipt of child support payments through local courts.

QUESTION 22: Are States required to distribute future payments made on or after October 1, 1996, in accordance with 45 CFR 302.51(c)?

ANSWER 22: Yes. See section I(f) of this Action Transmittal.

QUESTION 23: Are States required to distribute collections in title IV-E foster care cases in accordance with 45 CFR 302.52?

ANSWER 23: Yes. Section 5547 of the BBA revised section 457 by reinstating the distribution rules for collections in title IV-E foster care cases, which were contained in former section 457. 45 CFR 302.52 implemented former section 457 requirements for distribution of collections in title IV-E foster care cases and therefore is consistent with the new section 457(f) of the Act.

QUESTION 24: In a current assistance case, when the support collected for the current month exceeds the amount of assistance paid to the family for the month, must the difference be paid to the family, as required by former section 457(b)(3) of the Act, or be retained by the State to satisfy the cumulative amount of unreimbursed assistance?

ANSWER 24: Former section 457(b)(3) of the Act no longer exists. Effective October 1, 1996, in the case of a family currently receiving assistance from the State, the State must, not exceeding the cumulative amount of assistance paid to the family, (1) pay to the Federal Government the Federal share of the total amount collected; and (2) retain, or distribute to the family, the State share of the total amount collected.

160160FEDERAL AND STATE INCOME TAX REFUND OFFSET

160160160AND FEDERAL ADMINISTRATIVE OFFSET

QUESTION 25: Section 457(a)(2)(B)(iv) of the Act requires that, to the extent an amount collected through Federal income tax refund offset exceeds the amount retained to satisfy assigned support, the State shall distribute the excess to the family. There are instances under which Federal income tax refund offset amounts must be returned to the obligor or his or her current spouse (e.g., if the amount was offset from a joint return or if the amount offset exceeded the amount owed at the time of offset.) Are States now precluded from distributing offset amounts to obligors or their current spouses?

ANSWER 25: No. Sections 464(a)(3)(C) and (D) of the Act, which govern Federal income tax refund offsets based on joint returns and the treatment of amounts collected by offset which exceed the amount owed at the time of offset, respectively, remain in effect.

QUESTION 26: Could a State certify the cumulative amount of unreimbursed assistance for Federal income tax refund offset?

ANSWER 26: No. Section 464 of the Act specifies the criteria for the collection of past-due support from Federal income tax refunds. Section 464 limits amounts which may be submitted for offset to past-due support, as defined in section 464(c) to "...the amount of a delinquency determined under a court order, or an order of an administrative process established under State law, for the support and maintenance of a child...".

QUESTION 27: Under the new distribution requirements in section 457, are collections through State tax refund offset distributed in the same manner as collections through Federal income tax refund offset under section 457(a)(2)(B)(iv) of the Act?

ANSWER 27: No. Only collections through Federal income tax refund offset are distributed under section 457(a)(2)(B)(iv). Collections received through State income tax refund offset must be distributed as other collections (other than Federal income tax refund offsets) are distributed under section 457 of the Act. (See section VI of this Action Transmittal.)

QUESTION 28: Under the new distribution requirements in section 457, are collections through the Department of the Treasury administrative offset process under the Debt Collection Improvement Act of 1996 distributed in the same manner as collections through Federal income tax refund intercept under section 457(a)(2)(B)(iv) of the Act?

ANSWER 28: No. Only collections through Federal income tax refund intercept are distributed first to satisfy arrearages under section 457(a)(2)(B)(iv). Collections received through Federal administrative offset of other Federal payments (other than Federal income tax refund intercepts) under the Debt Collection Improvement Act of 1996 must be distributed in accordance with the rules that apply to all other collections under section 457 of the Act. (See section VI of this Action Transmittal.)

QUESTION 29: May State income tax refund intercepts be distributed in the same manner as Federal income tax refund intercepts if provided under State law?

ANSWER 29: No.

160DISTRIBUTION OF ARREARAGES IN FORMER ASSISTANCE CASES

QUESTION 30: In former assistance cases, does the distribution scheme outlined under section 457(a)(2)(B)(ii)(II) of the Act only pertain to arrearages that accrue after October 1, 2000 OR do States need to distribute the arrearage balance owed as of September 30, 1999 in the manner required during the period the arrearages accrued?

ANSWER 30: After applying the collection to the current month's support obligation and the never-assigned arrearages, under section 457(a)(2)(B)(ii)(II) of the Act, as added by section 302 of PRWORA, States must distribute remaining collections made on or after October 1, 2000 to the family to satisfy:

(1) any unassigned pre-assistance arrearages and conditionally-assigned arrearages, which accrued after September 30, 1997, (and, any unassigned previously permanently-assigned arrearages that accrued under section 426(a)(26) of the Act which the State cannot distinguish between pre-assistance or during-assistance arrearages),

(2) then any remaining amount to permanently-assigned arrearages to the extent necessary to reimburse the cumulative amounts paid to the family as assistance by the State, and

(3) finally, any remaining amount to the family to satisfy any unassigned during-assistance arrearages.

This distribution scheme pertains to any arrearages collected on or after October 1, 2000, not just to arrearages that accrue on or after October 1, 2000. Therefore, States may not distribute the arrearage balance owed as of September 30, 1999 in the manner required during the period the arrearages accrued, unless such arrearages are collected prior to October 1, 2000.

QUESTION 31: Is a State permitted to implement some or all of the distribution changes earlier than the specified implementation dates noted in the legislation?

ANSWER 31: Section 457 of the Act allows States to implement some or all of the distribution changes earlier than mandated. However, if the State elects to implement any distribution change earlier than required by section 457 of the Act (for example, as permitted under section 457(a)(6) of the Act), any amount of support collected which "could have been retained" had the State not implemented the distribution provisions earlier than required, will be counted for purposes of the "hold harmless" provision in section 457(d) of the Act.

QUESTION 32: Do support obligations which accrue under an assignment in effect on September 30, 1997, remain permanently assigned to the State through September 30, 1997, or through the last date of assistance?

ANSWER 32: Any support obligation that accrues through the last date of assistance under an assignment in effect on September 30, 1997, remains permanently assigned to the State in an amount not exceeding the cumulative amount of unreimbursed assistance as of the last date of assistance paid under that assignment.

QUESTION 33: In a former assistance case, may a State apply collections to conditionally-assigned arrearages that accrued before the family went on assistance before applying collections to permanently-assigned arrearages?

ANSWER 33: States have the option to implement the October 1, 2000 distribution rules which apply to former assistance cases early and, therefore, a State may opt to apply collections to never-assigned and conditionally-assigned arrearages first if they choose to do so prior to October 1, 2000. However, effective October 1, 2000, in a former assistance case, after paying current support, States must apply collections to conditionally-assigned arrearages before applying any collections to satisfy permanently-assigned arrearages, unless the collection is the result of a Federal income tax refund offset.

GAP PAYMENTS

QUESTION 34: What is a "gap payment" State?

ANSWER 34: A "gap payment" State is a State that elected in 1975, to provide AFDC families protection against a decrease in grants because of the payment of support directly to the State. Former section 402(a)(28) of the Act provides for the payment to the family of child support collected by the State as a protection against a reduction in the total income available to the family in a month. This section applies to those States whose State title IV-A plan, in July, 1975, permitted a portion of the monthly child support payment after application of appropriate disregards to be retained by a family receiving AFDC without causing a dollar-for-dollar reduction in the AFDC payment made to the family.

QUESTION 35: For States making gap payments, will Federal funds continue to be available once a State implements TANF? What will be the source of the Federal funds, if available?

ANSWER 35: Section 457(e) of the Act, as added by section 302 of PRWORA, allows States to continue to make gap payments if they were making such payments to families receiving assistance prior to August 22, 1996. The Federal government will continue to share in the cost of such payments, made in addition to the amount of assistance otherwise payable to the family, once the State implements the TANF program. Since gap payments are made from support collections in assistance cases, the portion of the support collection paid to the family to fill the gap between the need standard and assistance payment level is the source of the Federal contribution to the gap payment. Any collections in such cases in excess of the amount of the gap payment must be distributed in accordance with section 457(a)(1), with the Federal government receiving its share of the collection.

QUESTION 36: Does section 457(e) allow a State to retain the Federal share of child support collections and pay it to a family that receives assistance under TANF?

ANSWER 36: Under section 457(e) of the Act, as added by section 302 of PRWORA, a State which made gap payments under former section 402(a)(28) of the Act before the date of enactment of PRWORA may continue to make such gap payments. The new distribution requirements of section 457 do not apply to the amount of the gap payment but do apply to any other amount collected. For example, assume a "gap State" elects to continue to "fill the gap" using child support collections as permitted by new section 457(e) of the Act. Assume that the child support collection is $250 and that the "gap" payment is calculated to be $100. In this case, $100 of the $250 collection is paid to the family as a "gap" payment. While the distribution rules are inapplicable to the $100 of the child support collection used to fill the gap (based on the section 457(e) exception), the State must follow the distribution rules with respect to the remaining sum of the collection ($150). As required by section 457(a)(1)(A), the Federal share is deducted first, and then the State may retain, or distribute to the family, the State share of the amount collected, as specified in section 457(a)(1)(B).

QUESTION 37: May the state make adjustments to the maximum amount of the "gap" payment?

ANSWER 37: While a State may make adjustments to the maximum amount of the gap payment, distribution rules are waived only with respect to a gap payment at the same level as a gap payment made under section 402(a)(28) as in effect on August 21, 1996. If a State raises the maximum amount of the gap payment, any additional payment to the family would be subject to section 457(a)(1) of the Act and may be made only from the State share of any collection in excess of the exempted gap payment level, after having paid the Federal share of any excess amount collected.

QUESTION 38: Does the last sentence in section 457(e) have any effect on the State's option to continue using child support collections to make the "gap" payment?

ANSWER 38: Under the last sentence in section 457(e) of the Act, as added by section 302 of PRWORA, for purposes of the hold harmless provision in section 457(d), the State share of the gap payment to the assistance family will be considered an amount that could be retained by the State if the gap payments were reported by the State as part of the State share of amounts collected in FY 1995. This requirement does not affect the State's option to continue to make gap payments, but would result in counting the State share of gap payments as part of the State share of collections for a given year if the State counted gap payments as part of the State share in FY 1995. The result could impact the amount of the State share of collections for the current fiscal year.

QUESTION 39: If my State has not been a "gap payment" State, can it become one?

ANSWER 39: No.

QUESTION 40: Are Federal income tax refund offset collections included in the gap payment calculation?

ANSWER 40: Under 45 CFR 232.21(a), Federal income tax refund offset collections are excluded from the definition of arrearages that may be used to fill the gap in gap payment States. This requirement remains in effect for States which elect, under section 457(e) of the Act, to continue gap payments under former 402(a)(28) of the Act, with one exception. In gap payment States in the 1st Circuit, Federal income tax refund offset collections may be used to fill the gap, as a result of the decision in Doucette v. Ives, 947 F.2d 21 (1st Cir. 1991).

FUTURE PAYMENTS

QUESTION 41: Must States hold future payments received in former or never-assistance cases until they are due?

ANSWER 41: No. Section 45 CFR 302.51(c) applies only to current assistance cases in which there is an assignment of support rights to the State as a condition of receiving assistance. States may retain until the due date, or immediately pay to the family, future payments in former or never-assistance cases, in accordance with State law.

QUESTION 42: If there are multiple obligors in a single assistance case, may amounts collected as support which represent payment on the required support obligation for future months be held and applied to such future months or must they be used to reimburse any existing unreimbursed assistance paid to the family?

ANSWER 42: Section 302.51(c) indicates that no amounts shall be applied to future months unless amounts have been collected which fully satisfy the support obligation assigned under section 408(a)(3) of the Act for the current and all past months. If there are no assigned arrearages for that obligor, the amount collected represents future support and may not be retained to satisfy any existing unreimbursed assistance.

FEDERAL REPORTING REQUIREMENTS

QUESTION 43: Will the OCSE reports be revised to reflect the new distribution requirements under section 457 of the Act?

ANSWER 43: Reporting forms will be revised by the Fall of 1997. A Definitions Workgroup was established to continue the work of the Performance Measures and Incentives Workgroups and used the progress in revising current reports of these Workgroups and under the Measuring Excellence Through Statistics (METS) Initiative as a starting point.

OTHER QUESTIONS

QUESTION 44: May a State retain assigned support up to the total amount of unreimbursed assistance, including assistance provided to a family under the TANF program in a form other than cash, e.g., child care or work subsidies or vouchers?

ANSWER 44: Under section 457 of the Act, the amount of assigned support that may be retained by a State is limited to the cumulative amount of unreimbursed assistance paid to the family. TANF-ACF-PA-97-1, dated January 31, 1997, defines "assistance" as every form of support provided to families under the Temporary Assistance to Needy Families program except for the following:

1601) services that have no direct monetary value to an individual family and that do not involve implicit or explicit income support, such as counseling, case management, peer support and employment services that do not involve subsidies or other forms of income support; and

1602) one-time, short-term assistance (e.g., automobile repair to retain employment and avoid welfare receipt and appliance repair to maintain living arrangements).

QUESTION 45: If a Judge orders a Non Custodial Parent to make a payment on arrearages, does this allow the IV-D agency to apply the entire amount paid to the Non Custodial Parent's arrearages, without applying any amount to current support paid to the family?

ANSWER 45: No. Any amount collected under the IV-D program must be distributed in accordance with section 457 of the Act, 45 CFR 302.51(a)(1) and this Action Transmittal. Under 45 CFR 302.51(a)(1), amounts collected, except for amounts collected through Federal income tax refund offset, must be treated first as payment on the required support obligation for the month in which the support was collected and if any amounts are collected which are in excess of such amount, these excess amounts shall be treated as amounts which represent payment on the required support obligation for previous months.

QUESTION 46: In a current assistance case, may a State wait until the end of the month to calculate the cumulative amount of unreimbursed assistance, and issue any payment of excess support to the household within 15 days of the end of the month? How will the 2-day distribution rule effective in October 1998 affect this?

ANSWER 46: Federal regulations at 45 CFR 302.32(f) governing timeframes for distribution of collections in IV-D cases remain in effect until the October 1, 1998 effective date of the requirement under section 454B(c) of the Act that the State disbursement unit disburse amounts payable under section 457(a) of the Act within 2 business days after receipt from the employer or other source of periodic income. Questions relating to the timing of disbursements by the State Disbursement Unit will be answered under a separate Action Transmittal on State Disbursement Unit requirements.

SYSTEMS-RELATED QUESTIONS

QUESTION 47: What are OCSE's plans for issuing revisions to the Federal test deck to incorporate PRWORA changes? Will there be more than one revision, taking into account the staggered distribution requirements?

ANSWER 47: OCSE plans to issue a revised distribution test deck which reflects PRWORA changes by the end of calendar year 1997. The revised test deck will address all of the distribution changes in PRWORA.

QUESTION 48: Section 454(24)(B) of the Act requires that States' automated systems meet PRWORA requirements by October 1, 2000. However, many of the individual requirements of PRWORA have earlier implementation deadlines, and they will often depend on systems support. Must States meet the implementation dates for individual program requirements or may a State delay full compliance with all requirements until October 1, 2000 when State systems must be operational?

ANSWER 48: The statute requires each State to have a Statewide computerized support enforcement system that meets all IV-D PRWORA requirements no later than October 1, 2000. The State is encouraged to automate PRWORA requirements sooner so that systems support is available when a PRWORA requirement becomes effective. The State can only rely upon the October 1, 2000 date as the implementation date for having a Statewide system as discussed above. The State must meet the specific effective date for each PRWORA requirement.

Attachment 263719Bibliography

Bibliography

Bane and Ellwood, The Dynamics of Dependence: The Routes to Self-Sufficiency, Report prepared for the U.S. Department of Health and Human Services. Cambridge, MA: Urban Systems Research and Engineering, Inc., 1983

Baron, Carol, Virginia Independence Program Evaluation Manager, Office of Planning and Policy, Department of Social Services, What Happened to Virginia8217s First 94 Two-Year Time Limit Cases?, August, 1998.

Edin, Kathryn and Laura Lein, Making Ends Meet, Russell Sage Foundation, New York, 1997.

Formosa, Carl, Washington State, Division of Child Support, The Effect of Child Support and Self-Sufficiency Programs on Reducing Direct Support Public Costs, December, 1998.

GAO, Welfare Reform: Child Support an Uncertain Income Supplement for Families Leaving Welfare, GAO/HEHS-98-198, Report to the Chairman, Subcommittee on Human Resources, Committee on Ways and Means, U.S. House of Representatives, August, 1998.

GAO, Welfare Reform: States Are Restructuring Programs to Reduce Welfare Dependence, GAO/HEHS-98-109, Report to the Chairmen, Committee on Finance, U.S. Senate, and Subcommittee on Human Resources, Committee on Ways and Means, U.S. House of Representatives, June 1998.

Green Book, Committee on Ways and Means, U.S. House of Representatives, May 1998.

Lewin Group: Study of Financing State Child Support Enforcement Programs, 1999

Meyer, Daniel R., Child Support and Welfare Dynamics: Evidence from Wisconsin, Institute for Research on Poverty, February, 1993.

Meyer, Daniel R., Cancian, Maria, Principal Investigators and Judi Bartfeld, Wisconsin Child Support Demonstration Evaluation, Quarterly Impact Report, Quarter 5, March 1998.

Pavetti, LaDonna, Against the Odds: Steady Employment Among Low-Skilled Women, A Report to the Annie E. Casey Foundation, July, 1997.

Pavetti, LaDonna, Moving Up, Moving Out or Going Nowhere? A Study of the Employment Patterns of Young Women, A Report to the Annie E. Casey Foundation, July 1997.

Schnexnayder, Deanna, and Jerome A. Olson, Daniel G. Schroeder, and Jody L. McCoy, The Role of Child Support in Texas Welfare Dynamics, Lyndon B. Johnson School of Public Affairs, University of Texas, September, 1998.

TANF Family Income Resource Models, Virginia Department of Social Services.

Turetsky, Vicki, Center for Law and Social Policy, Some States Continue $50 Child Support Pass-Through Despite Federal Disincentive, May, 1997.

Wisconsin Department of Workforce Development and University of Wisconsin8217s Survey Research Laboratory, Survey of Those Leaving AFDC or W-2, Preliminary Report, January to March 1998, January, 1999.