United States General Accounting Office GAO Report to the Committees on Finance, U.S. Senate, and Ways and Means, House of Representatives January 1999 SUPPLEMENTAL SECURITY INCOME Increase Receipt and Reporting of Child Support Could Reduce Payments GAO/HEHS-99-11 United States GAO General Accounting Office Washington, D.C. 20548 Health, Education, and Human Services Division B-279807 January 12, 1999 The Honorable William V. Roth, Jr. Chairman The Honorable Daniel Patrick Moynihan Ranking Minority Member Committee on Finance United States Senate The Honorable Bill Archer Chairman The Honorable Charles B. Rangel Ranking Minority Member Committee on Ways and Means House of Representatives In 1996, over a quarter of the nation’s children lived in a household where only one parent was present, usually the mother. About 42 percent of female-headed families with children had incomes at or below the poverty level, and a majority of these families received government aid to help meet their basic needs. For four of the largest federal programs that provide such aid—Temporary Assistance for Needy Families (TANF), Medicaid, the Food Stamp Program, and Supplemental Security Income (SSI)—child support received by a custodial parent is required to be considered as income in determining eligibility and benefit amounts. To reinforce the idea that parents have the primary responsibility for supporting their children and to minimize government costs of providing aid, single parents with children applying for or receiving TANF (since 1975) or Medicaid (since 1984) are required as a condition of eligibility to cooperate with state Child Support Enforcement (CSE) agencies in seeking child support income. Cooperation involves identifying and helping to locate the noncustodial parent of a child and, if needed, taking steps to help establish paternity. In addition, as of 1996, states have the option of requiring food stamp applicants and recipients to cooperate with CSE. In contrast, single parents receiving SSI on behalf of their children are not required to cooperate with CSE agencies. These parents are, however, required to report any child support income received to the Social Security Administration (SSA), which administers the SSI program. In June 1998, about 60 percent of the 904,554 children under age 18 receiving SSI lived in single-parent families. In about 9 percent of these cases, the parents Page 1 GAO/HEHS-99-11 SSI and Child Support B-279807 reported child support income to SSA. These statistics suggest that there is potential for increasing child support income in the other families. They also suggest that some families receiving child support may not report it to SSA. Because of congressional interest in enhancing parental responsibility and self-sufficiency among families receiving public assistance and concern about the need for improved management of the SSI program, we explored opportunities for increasing the number of SSI single-parent families receiving child support and for improving the reporting of such support to SSA. More specifically, we identified the potential for reducing SSI payments by increasing the extent to which (1) SSI children in single-parent families are served by the CSE program and have support collected for them and (2) support collected by CSE programs is reported by custodial parents to SSA. We focused our work on the CSE programs of Florida, New York, and Texas, three of the four states with the largest numbers of children receiving SSI.1 We did our work between May 1997 and December 1998 in accordance with generally accepted government auditing standards. (See app. I for information on our scope and methodology.) SSIbenefits could be reduced and single-parent families’ incomes Results in Brief increased if more children on SSI received CSE services. The Florida, New York, and Texas CSE programs provided services to almost 45,000 SSI children in single-parent families in those states and collected child support for more than one-quarter of those served. However, more than 52,000 other SSI children in single-parent families in those states did not receive CSE services. We estimate that if their parents had been required to cooperate with CSE services, annual SSI benefits to these children would have been reduced by about $4.2 million, while the net annual income (considering the child support and the resulting adjustments to the SSI benefits) of the SSI single-parent families would have been increased by $2.2 million. These potential benefit reductions would be offset by the costs for SSA to administer a child support cooperation requirement and by the costs, which could be considerable, for CSE programs to provide services. Even though savings to the government are not guaranteed, increasing the number of SSI children receiving CSE services would help promote parental responsibility and increase the incomes of single 1 The four states with the largest number of children receiving SSI, in descending order, are New York, California, Florida, and Texas. We did not include California in our review because it does not have a statewide automated CSE data system. The three states included in our review represented about 20 percent of SSI children nationwide in June 1998. Page 2 GAO/HEHS-99-11 SSI and Child Support B-279807 parents, helping them to achieve and maintain economic independence and reducing their dependence on public assistance. Among the SSI children in single-parent families for whom the three states’ CSE programs collected support, we found strong evidence that many parents had not reported the income to SSA as they were required to do. The Florida, New York, and Texas CSE programs collected support for 15,427 SSI children—12,841 living in those states and an additional 2,586 SSI children living in other states. For 68 percent of these children, SSI records did not contain a report of child support income, indicating that their parents did not report the income to SSA. We estimate that, in these three states, SSA overpaid $7.7 million in annual SSI benefits because this unreported support was not considered in calculating children’s’ SSI benefits. These overpayments could be minimized in the future if SSA and CSE established a routine method of exchanging information on SSI recipients and child support collections. The Congress established SSI in 1972 to ensure a minimum cash income to Background needy aged, blind, and disabled persons, including needy children under age 18 if they have severe disabilities. SSI, now the nation’s largest cash assistance program for the poor, in 1997 paid about $26 billion in federal funds to about 6.5 million low-income aged, blind, and disabled recipients. Although SSI primarily serves adults, the number of children receiving SSI has increased from 127,000 in December 1975 to 928,000 in December 1997. SSA administers the program, which is authorized by title XVI of the Social Security Act. In 1975, in response to growing numbers of families receiving Aid to Families With Dependent Children, the Congress created the CSE program to reduce welfare spending and help single-parent families achieve or maintain economic self-sufficiency.2 The program provides assistance in obtaining financial and medical support for children through locating noncustodial parents, establishing paternity and support obligations, and enforcing those obligations. The Congress expanded the CSE program over the years, and today state CSE agencies are required to provide services automatically to families receiving TANF, foster care payments or services, or Medicaid and to any other family that requests them.3 State CSE agencies have responsibility for administering the program through state and local offices. At the federal level, the Office of Child Support Enforcement 2 CSE refers to the program authorized by title IV, part D, of the Social Security Act. 3 Persons required to cooperate with state CSE agencies may not be charged fees for CSE services. Page 3 GAO/HEHS-99-11 SSI and Child Support B-279807 within the Department of Health and Human Services (HHS) has oversight responsibility for the program. The federal government pays 66 percent of the states’ administrative costs for the CSE program, with the states paying the remainder. The federal government also pays states incentive payments for performance in establishing paternity and support orders, collecting current support, collecting past due support, and operating cost-effectively.4 In 1997, the CSE program collected about $13.4 billion in child support for families receiving TANF and other families. About $10.8 billion of this was distributed to families and $2.2 billion retained by the government as recoupment of TANF payments. The Social Security Act requires that TANF recipients assign their rights to child support to the state government and that cash child support payments collected for TANF families be retained by the government as recoupment for benefits.5 In the same year, CSE collected child support for about 13 percent of the TANF child support cases. The requirement that persons cooperate with CSE as a condition of receiving TANF or Medicaid benefits on behalf of children emphasizes that parents, not the government, have primary responsibility for supporting their children.6 However, TANF, Medicaid, and Food Stamp Program applicants and recipients may not be required to cooperate with CSE if they claim to have good cause for refusing to do so and the state agencies administering these programs determine that the circumstances claimed make cooperation not in the best interests of the child. Good causes for refusing to cooperate may include, for example, fear of physical or emotional harm to the child and/or custodial parent. For TANF recipients who are required to cooperate but fail to do so, states are required to reduce these recipients’ TANF grants by at least 25 percent and may opt to eliminate TANF aid for the entire family.7 4 In the calculation of performance incentives, collections on TANF cases and former TANF cases receive twice the weight as collections on cases of persons who have never received TANF. 5 Medicaid recipients assign to the government only their rights to medical child support payments and payments for medical care from any third party. Thus, only cash child support designated for medical expenses is retained by the government. 6 Many Medicaid and food stamp recipients are also TANF recipients and are therefore required to cooperate with CSE. Non-TANF recipients of Medicaid are required to cooperate with CSE only in establishing paternity and seeking medical support. In those states that have opted to require it, non-TANF food stamp recipients are required to cooperate with all CSE services. As of December 1997, seven states had chosen to require food stamp recipient cooperation with state CSE agencies. 7 A survey conducted by the American Public Human Services Association in the summer of 1997 found that 16 states eliminate the entire family’s assistance for failure to cooperate with CSE. Page 4 GAO/HEHS-99-11 SSI and Child Support B-279807 The SSI program, on the other hand, does not require cooperation with CSE as a condition of eligibility for receipt of SSI benefits.8 However, under the Social Security Act, two-thirds of any child support received by, or on behalf of, an SSI recipient is to be counted as income in calculating SSI benefits. The amount of child support counted as income results in a dollar-for-dollar decrease in the amount of SSI benefits to which a child is entitled. These adjustments can reduce the benefit level, depending on the size of the benefit, to zero, making the child ineligible for benefits. When SSI parents receive child support for their children, they are required to report this income to SSA for benefit determinations. SSA relies on applicants and recipients to accurately report their income and assets, and SSA policy requires that its staff obtain documentation to verify the amount of income and resources that applicants report. To ensure recipients’ continuing financial eligibility and possibly detect some types of income and resources that clients may not have reported, SSA uses computer matches of SSI payment records against recipient financial information contained in the payment files of third parties, such as other federal and state government agencies. However, SSA does not match computer data with state CSE agencies’ child support collection systems. Our analysis in three states indicates that opportunities exist to serve Opportunities Exist to more SSI children through the CSE program, which should result in Increase the Number increased child support income for some families and reduced SSI of SSI Children payments. Providing CSE services to additional families would entail costs for SSI and CSE program administration, but would help promote parental Receiving Child responsibility, increase some families’ incomes, and reduce the need for Support public assistance. 8 Most SSI recipients are also Medicaid recipients and, if they are single parents, are required to cooperate with the CSE program for the establishment of paternity (if needed) and the collection of medical support for children in single-parent homes. Moreover, some families with a child receiving SSI may also have one or more other children receiving TANF. In those cases, the families would be required to cooperate with CSE to receive TANF benefits. Generally, a child does not receive both SSI and TANF. Page 5 GAO/HEHS-99-11 SSI and Child Support B-279807 More Than Half of the In Florida, New York, and Texas, 97,095 children under age 18 received SSI Children Receiving SSI in in single-parent families as of September 1997.9 This represented about Single-Parent Families in half of all SSI children in those states.10 The three states’ CSE programs provided services to a total of 44,756, or 46 percent, of the SSI children in Three States Did Not single-parent families in those states during the periods of our review. Receive CSE Services These services resulted in child support collections for a significant number of families. More than half (54 percent), however, of these states’ SSI children in single-parent families did not receive CSE services. As shown in table 1, we did not find a CSE record for 47, 53, and 65 percent, respectively, of children receiving SSI in single-parent families in Florida, New York, and Texas during the periods of our review.11 Table 1: Receipt of Child Support Services by Children Receiving SSI in Children receiving SSI in single-parent families Single-Parent Families in Three States Children’s state of Receiving CSE Not receiving CSE residence Total servicesa services Number Number Percent Number Percent Florida 30,459 16,235 53 14,224 47 New York 42,026 19,825 47 22,201 53 Texas 24,610 8,696 35 15,914 65 Total 97,095 44,756 46 52,339 54 Note: Data are for children in Florida receiving SSI as of July 1998, and in New York and Texas as of September 1997. a Services provided by the Florida CSE during the 12 months ending mid-July 1998, by the New York CSE for the 8 months ending August 31, 1997, and the Texas CSE for the 12 months ending July 31, 1997. Taken together, the three states’ programs collected support for 12,841, or 29 percent, of the 44,756 SSI children they served. Individually, the Florida, New York, and Texas CSE programs collected support for 32, 21, and 39 percent, respectively, of the SSI children living in those states whom 9 For the purposes of our analysis, the terms single parent and single-parent family do not refer to the marital status of parents or the number of parents in a household but rather to the situation in which children are living in a home where only one of their natural or adoptive parents is present. In some instances, someone other than a parent, such as an adult caretaker or representative payee, may receive SSI benefits on behalf of a child. 10 The SSI children in single-parent families in Florida, New York, and Texas represented 55, 54, and 45 percent, respectively, of all of the SSI children in each of these states. 11 We found a small number—less than 10 percent—of single parents of SSI children not receiving services who had reported child support income to SSA. They may receive child support through means other than the CSE program, such as through the courts or directly from noncustodial parents. Page 6 GAO/HEHS-99-11 SSI and Child Support B-279807 they served.12 The median annual amount of support collected for these SSI children was $487 in Florida, $350 in New York, and $501 in Texas. (See also tables II.1 and II.2 for more detailed data.) SSI Benefits Could Be Providing CSE services to SSI children in single-parent families who are not Reduced and Families’ currently receiving them can potentially increase child support income for Incomes Increased If More some of them and reduce reliance on SSI benefits. It is not possible to estimate precisely the potential for child support collections among the SSI Children on SSI Received children not currently receiving services because doing so requires CSE Services information about the potential for payment by noncustodial parents, which is not available. To estimate SSI benefit reductions that might result from more SSI children receiving CSE services in the three states included in our review, we applied each state’s collection rate for the SSI children it currently served to the children not currently served and assumed that the same median amounts of support could be collected. The actual amounts of CSE collections that may result from serving these families could be higher or lower than this estimate. With this methodology, we estimate that CSE could collect a total of $6.4 million in support for an additional 14,119 SSI children.13 With two-thirds of the support counted as income for SSI benefit calculations, we estimate SSI benefits for children in these states could be reduced by a total of $4.2 million annually—$1.3 million in Florida, $1 million in New York, and $1.9 million in Texas. Despite this reduction in benefits, the SSI families’ incomes would be increased by a net of $2.2 million because not all of the child support received would result in a reduction in SSI benefits. (See table I.1 for detailed calculations.) Requiring Parents of SSI Like any enforcement effort, requiring that all single parents applying for Children to Cooperate With or receiving SSI for their children cooperate with CSE services as a CSE Increases Costs for condition of receiving aid comes at a cost. SSI eligibility workers would need to perform additional tasks to implement the requirement, and CSE’s SSI and CSE but Ensures caseloads would increase, raising costs for these programs. However, such That Child Support Is a requirement would increase the likelihood that noncustodial parents Pursued for Those fulfill their financial responsibilities toward their children, thereby Receiving Aid reducing custodial parents’ dependence on the government for support. 12 State CSE agencies can also provide services to children residing in other states, which we discuss later in this report. 13 These data exclude SSI children whose SSA records indicated the receipt of child support. Page 7 GAO/HEHS-99-11 SSI and Child Support B-279807 A cooperation requirement would necessitate that SSI eligibility workers, like workers in other programs, provide information about the CSE program to single parents receiving or applying for SSI on behalf of their children. SSI eligibility workers would have to explain the requirement for cooperation with CSE services, obtain information from parents for the referral of children’s cases to CSE agencies, and make the case referrals. For the TANF program, for example, some CSE agencies develop an intake form to be used by the TANF agency in obtaining information on the family and the noncustodial parent to help the CSE agency establish an enforceable case.14 Also, in other programs, applicants must be notified of the right to claim good cause for not cooperating, and then determinations must be made about whether good cause exists.15 Additional time required for tasks like these would likely lengthen the initial SSI eligibility interview and increase associated paperwork. In November 1997, SSA began interviewing a nationwide sample of approximately 1,000 SSI recipients about child support issues to determine the extent to which SSI recipients try to obtain child support and to estimate the potential cost of adding a component on child support issues in the initial SSI eligibility interview. SSA officials told us that the agency has a draft legislative proposal under review by the Office of Management and Budget that would require persons applying for or receiving SSI on behalf of children to show they have tried to obtain child support. The proposal does not require these individuals to cooperate with or to seek support specifically through the CSE program—as opposed to other means—as is now required for TANF and Medicaid recipients. The nationwide study would also help to estimate the costs for SSA to implement its legislative proposal. Establishing a CSE cooperation requirement for SSI recipients would also increase the workload and federal and state costs of the CSE program.16 In New York and Texas, serving currently unserved SSI children would increase the total CSE caseload in those states (as of Aug. 1997) by about 2 percent; for Florida, the increase would be about 1.3 percent. There appears to be no generally accepted methodology for calculating the 14 In some jurisdictions, TANF applicants must cooperate with CSE (usually in the form of an interview with a state CSE agency) before financial assistance is provided. 15 In the Aid to Families With Dependent Children (now TANF) program for fiscal year 1995—the latest year for which data were available—8,387 claims of good cause were made and 5,462 were found valid. The monthly average number of families receiving AFDC was 4.9 million in fiscal year 1995. 16 State CSE agencies are required to provide services to anyone who requests them, and SSI recipients may all request services. Page 8 GAO/HEHS-99-11 SSI and Child Support B-279807 actual costs of serving additional families in the CSE program. Although we did not estimate CSE costs for providing services to additional families, we believe they could be significant, potentially large enough to offset the savings in SSI benefit reductions. While the CSE program would bear the costs of providing services to all of the SSI families with children, it would not collect child support for all of them, limiting the amount of savings through SSI benefit reductions. Moreover, 34 percent of these CSE costs would be borne by the states, which would generally not experience savings from the reduced SSI benefits. Nevertheless, government cost reduction is not the only goal of the CSE program. It also promotes parental responsibility and strives to increase the incomes of single parents, helping them to achieve and maintain economic independence and reducing their dependence on public assistance. In addition to exploring the potential for more SSI recipients to receive CSE SSI Overpayments services, we also determined the extent to which those already receiving Occur From CSE services were reporting to SSA the child support income collected for Nonreporting of Child them by state CSE agencies. Our analysis of data for three states indicated that many single parents of SSI children did not report to SSA the child Support Income support collected for them by the CSE program. On the basis of this analysis, we estimated that SSA overpaid about $7.7 million in SSI benefits in 1 year as a result of such nonreporting. SSA could identify or prevent many of these types of overpayments if it had information on CSE collections for SSI children from one of the existing or soon-to-be-operational sources of computerized CSE collections data. Many Parents Did Not The Florida, New York, and Texas CSE programs provided CSE services for Report Child Support SSI children living in those states as well as to some SSI children living in Income to SSA other states.17 The three states’ CSE programs served a total of 55,884 SSI children in single-parent families, including the 44,756 we identified living in Florida, New York, and Texas and an additional 11,128 SSI children residing in other states. These CSE programs collected support for 15,427, or 28 percent, of these SSI children.18 17 A state CSE agency gives services to all children in the state who have requested services or whose custodial parents are required to cooperate with CSE. In addition, a CSE agency in one state will provide services to a child in another state when asked to do so by the CSE agency in the child’s state of residence. 18 The Florida, New York, and Texas CSE programs’ child support collection rates for the SSI children they served, including children living within and outside those states, were 32, 21, and 37 percent, respectively. These individual state and aggregate rates are very similar to the collection rates cited earlier that are based only on the SSI children served who reside in those states. Page 9 GAO/HEHS-99-11 SSI and Child Support B-279807 Our comparison of SSA and CSE records indicates that many parents of SSI children with child support income did not report the income to SSA. As shown in table 2, we found no indication of child support being reported for 10,432, or 68 percent, of the SSI children for whom the three states’ CSE programs collected support. (See also tables II.1 and II.2 for more detailed information by state on the numbers and percentages served and for whom collections were made.)19 Table 2: Parental Reporting of Child Support Collections by Three States’ SSI children for whom CSE collected child supporta CSE Programs for SSI Children in Parent reported Parent did not report Single-Parent Families support to SSA support to SSA Total State CSEb number Number Percent Number Percent Florida 5,828 1,755 30 4,073 70 New York 5,720 1,704 30 4,016 70 Texas 3,879 1,536 40 2,343 60 Total 15,427 4,995 32 10,432 68 Note: Data are for children in Florida receiving SSI as of July 1998, and in New York and Texas as of September 1997. a Collections made by the Florida CSE during the 12 months ending mid-July 1998, by the New York CSE for the 8 months ending August 31, 1997, and by the Texas CSE for the 12 months ending July 31, 1997. The total amount of unreported child support collected by these three states for SSI children was almost $12 million.20 Because parents did not report this income to SSA, SSA was not able to adjust benefit amounts accordingly. We estimate that as a result, SSA overpaid $7.7 million in SSI benefits in 1 year in these three states (see table 3).21 19 Because we relied on a computerized match of extracts of CSE and SSI automated records and not a detailed review of case file information, we were not able to confirm that the child support collections recorded in the CSE records were in fact distributed to each child’s custodial parent and, therefore, were required to be reported to SSA. In some cases, for example, where other children in a family were receiving TANF, the support collected for the SSI child may have been inappropriately retained by the government and not distributed to the family. In other cases, the child support for an SSI child could have been past due for a previous period of time during which the child was receiving TANF; in this case, the child support collected may have been retained, appropriately, by the government and not distributed to the family. In both cases, the support would not be counted by SSA as income because the family did not receive it. 20 To arrive at this total, the collections by New York’s CSE over an 8-month period were projected to 12 months. 21 For our estimate, we calculated the estimated annual SSI overpayment for each child as two-thirds of the annual unreported child support (the amount considered as income under law) or the annual SSI benefit, whichever was less, and summed these amounts for all children for whom support was not reported. Page 10 GAO/HEHS-99-11 SSI and Child Support B-279807 Table 3: Estimated Annual SSI Overpayments Resulting From Estimated Nonreporting of Child Support annual Collected by Three States’ CSE State collecting support overpayment Programs Florida $2,858,205 New York 3,206,240 Texas 1,658,028 Total $7,722,473 Data on CSE Child Support We have in the past reported on the need for SSA to verify information Collections Are Available voluntarily reported by SSI recipients by using existing state and other for Verifying SSI databases.22 We specifically recommended that, to prevent overpayments or detect them sooner, SSA (1) identify additional income sources for Recipients’ Income which SSA does not currently have computer matches and (2) use on-line access to routinely check for unreported sources of income in states when it is cost-effective to do so.23 SSA has taken some actions in response to our recommendations, including expanding use of on-line data maintained by state agencies to better verify recipients’ financial information. In a recent report, we recommended that SSA accelerate efforts to identify more timely and complete sources for verifying financial eligibility information.24 In this report, we identify another opportunity for SSA to improve its administration of the SSI program. Although no single, nationwide source of computerized data on all states’ CSE collections exists, states are required by the Social Security Act to have three statewide computerized data files containing information on child support collections from which SSA may obtain data for SSI recipients. The first computerized data files from which SSA may obtain CSE collections information are states’ CSE automated data processing and information retrieval systems. State CSE programs were required by the Family Support Act of 1988 to have statewide automated systems that meet federal specifications by October 1, 1997. In addition, the 1996 welfare reform legislation required that states enhance these systems to be 22 Supplemental Security Income: Timely Data Could Prevent Millions in Overpayments to Nursing Home Residents (GAO/HEHS-97-62, June 3, 1997); Supplemental Security Income: SSA Efforts Fall Short in Correcting Erroneous Payments to Prisoners (GAO/HEHS-96-152, Aug. 30, 1996); Supplemental Security Income: Opportunities Exist for Improving Payment Accuracy (GAO/HEHS-98-75, Mar. 27, 1998). 23 Supplemental Security Income: Administrative and Program Savings Possible by Directly Accessing State Data (GAO/HEHS-96-163, Aug. 29, 1996). 24 Supplemental Security Income: Action Needed on Long-Standing Problems Affecting Program Integrity (GAO/HEHS-98-158, Sept. 14, 1998). Page 11 GAO/HEHS-99-11 SSI and Child Support B-279807 capable of electronically interfacing with other federal and state agencies. As of November 16, 1998, 37 states’ CSE programs had received federal certification for their data processing and information retrieval systems.25 Two other computerized statewide data systems were required by the 1996 welfare reform legislation to be in operation by October 1, 1998: (1) states’ centralized child support payment collection and disbursement units, required for the identification, receipt, and distribution of support payments on both CSE and non-CSE child support cases,26 and (2) states’ CSE case registries, required to comprise information on all CSE and non-CSE cases.27 For CSE cases, the registries must also include information on any amounts owed that have been collected. Information contained in the state case registries must be capable of being extracted from the state’s CSE system and shared and compared with information contained in other databases. SSA officials expressed concern that CSE data files do not contain sufficient detail for them to determine accurately the amounts of child support income received by SSI children that should be considered as income for the purposes of ascertaining eligibility and payment amounts for SSI benefits. In addition, they said that in the past they encountered problems in accessing various kinds of state data because of the differing capacities of state data systems and state policies on and willingness to share data considered confidential. SSA officials noted, for example, that after several years of effort, exchanges of data on quarterly wages have not been achieved with all states. They suggested that it would be more efficient if CSE was responsible for reporting to SSA collections for SSI children, rather than requiring SSA to access state data systems. We agree with SSA that some states’ CSE data files may not contain sufficient information upon which to base adjustments to SSI benefit calculations. But states’ CSE data could, at a minimum, indicate to SSA those cases for which child support may be a potential source of income that should be further investigated. In addition, we acknowledge that the ease and costs of obtaining CSE collections data will vary from state to state because different degrees of automation exist in the data files of state CSE 25 The 37 include Guam and Puerto Rico. 26 If in August 1996 local courts were being used to process payments, the state has an additional year before it must have a central payment disbursement unit. State disbursement units must process non-CSE cases issued on or after January 1, 1994, which are subject to income withholding. 27 State case registries must contain information on non-CSE cases established or modified after October 1, 1998. Page 12 GAO/HEHS-99-11 SSI and Child Support B-279807 agencies. However, as more state CSE systems become federally certified and the state case registries and centralized collection and distribution units become operational, we expect the ease of access to increase and costs to decrease. It is also possible that the most efficient and cost-effective option is for CSE to report collections to SSA. We did not estimate the potential costs of the various approaches for SSA and CSE agencies to exchange data on child support collections. The decision on whether state CSE programs should report data to SSA or SSA should access state data files should be based on an assessment of the relative costs and efficiencies of various possible approaches. The Congress established the CSE program and expanded it over the years Conclusions to minimize public expenditures for low-income families and enhance families’ self-sufficiency by helping to obtain child support income for them. Even though SSI is an important source of cash support for many children, single-parent families applying for or receiving SSI are not required to cooperate with CSE. Yet, we found that many SSI children who could benefit from CSE services are not receiving them. If more SSI children received CSE services and ultimately received child support, SSI benefits would be reduced, and families receiving support would be less dependent on public assistance and closer to achieving self-sufficiency. The Congress could ensure that families receiving SSI receive CSE program services by amending the Social Security Act to require cooperation with CSE as a condition of eligibility for SSI benefits. This would be in keeping with the principle that parents, not the federal government, have primary responsibility for supporting their children and should seek child support, if appropriate, when applying for or receiving federal public assistance. In addition, many single parents of SSI children who are already receiving child support through the CSE program are not reporting the child support to SSA, resulting in millions of dollars of SSI overpayments. SSA could prevent or detect these overpayments by identifying child support collected by CSE for SSI recipients. Several computerized sources of such data are available to facilitate such an effort. We recommend that the Congress amend the Social Security Act to require Recommendation to that all single parents applying for or receiving SSI benefits on behalf of the Congress children under age 18 be required to cooperate with CSE services, unless Page 13 GAO/HEHS-99-11 SSI and Child Support B-279807 they have good cause not to do so. The Congress also will need to consider how best to enforce such a requirement. We recommend that the Commissioner of Social Security and the Recommendation to Secretary of HHS take steps to implement a cost-effective method for the Commissioner of ensuring that data on CSE collections for children receiving SSI are made Social Security and available to SSA and used in making eligibility determinations. This may include seeking legislative changes to allow SSA access to states’ data on the Secretary of HHS child support collections. In commenting on a draft of this report, SSA agreed with our Agency Comments recommendation that parents applying for and receiving SSI for children and Our Response under age 18 be required to cooperate with CSE agencies. SSA said it is considering various options for effecting such changes through legislation. SSA also agreed with our recommendation that SSA and CSE explore ways to implement a cost-effective method for ensuring that data on CSE collections for children receiving SSI are made available to SSA and used in making eligibility determinations. SSA commented that it had already begun discussions with HHS’ Office of Child Support Enforcement about alternative ways to obtain child support payment data to improve the SSI eligibility determination process. It also noted that legislation may be needed to require all states to report CSE payment data to SSA. At the same time, SSA said it continues to negotiate on-line access to state data on an individual state basis, addressing privacy issues as needed. It also noted that to address privacy issues more generally and thus facilitate its access to a wide range of state data, including child support data, it has developed a legislative proposal that would deem that SSA’s privacy standard would meet all states’ privacy standards for sharing data. SSA’s comments appear in appendix III. In commenting on a draft of this report, HHS also agreed with our recommendation that parents applying for and receiving SSI for children generally should be required to cooperate with CSE agencies. HHS said it is prepared to work with SSA and the Congress to address the numerous issues related to such a change and to develop legislation that achieves our recommendation. Regarding the recommendation about working with SSA to ensure that CSE data are available to SSA, HHS said that discussions are already under way with SSA about ways to provide payment data and noted along with SSA that legislation may be needed to require all states to report CSE payment data to SSA. In the meantime, it is supporting SSA’s Page 14 GAO/HEHS-99-11 SSI and Child Support B-279807 efforts to negotiate on-line access to state data on an individual state basis. HHS also noted that the report appears to equate child support enforcement services with the IV-D caseload, stating that some SSI children may also receive services as part of the non-IV-D caseload. We have clarified this point in the report. HHS’ comments appear in appendix IV. We also provided copies of a draft of this report to the Florida, New York, and Texas CSE agencies. Florida and Texas officials expressed concerns about the costs to the states of providing CSE services to SSI clients. In addition, officials from all three states raised a range of implementation issues, including how strongly the cooperation requirement will be enforced for parents and their children; what the specific responsibilities would be for SSA and the child support agencies, such as for collecting relevant information from custodial parents and determining good cause exemptions; and the potential burden to their information systems of sharing information at a time of great demands on their systems as a result of welfare reform and federal requirements for updating their CSE automated systems. We noted in the report that states would bear some of the costs due to additional CSE caseloads and would not share in the potential savings in SSI benefits to children, which are federally funded. However, the federal government also pays incentives to states for their CSE program performance in specified ways, giving greater weight to TANF cases than non-TANF cases in the calculation of these performance incentives, and we have added this information to the report. In developing legislation to implement a cooperation requirement for parents of SSI children, the Congress could do the same for SSI cases, which could mitigate the cost impact on states to some extent. We also acknowledge that many implementation details will need to be addressed by the Congress, SSA, and states’ CSE agencies and that some changes to how states are currently operating may be needed. However, we believe that implementing a cooperation requirement and exchanging key program data are important goals in keeping with promoting personal responsibility among individuals receiving government aid and managing government programs efficiently and effectively. We are sending copies of this report to the Commissioner of Social Security, the Secretary of Health and Human Services, and other interested parties. Page 15 GAO/HEHS-99-11 SSI and Child Support B-279807 If you have any questions about this report, please contact Mark V. Nadel, Associate Director, on (202) 512-7215 or Gale C. Harris, Assistant Director, on (202) 512-7235. Other major contributors were Catherine V. Pardee, Evaluator-in-Charge, and Vanessa R. Taylor, Computer Systems Specialist. Richard L. Hembra Assistant Comptroller General Page 16 GAO/HEHS-99-11 SSI and Child Support Page 17 GAO/HEHS-99-11 SSI and Child Support Contents Letter 1 Appendix I 20 Identifying Children Receiving SSI in Single-Parent Families 20 Scope and Determining Extent of Receipt and Reporting of CSE Services 21 Methodology and Support Collections Estimating Potential SSI Benefit Reductions From Increased 22 Receipt of Child Support Services and Collections Estimating Potential Reductions in SSI Overpayments by 24 Improved Support Reporting Appendix II 25 Receipt and Reporting of Child Support Services and Income Through Three States’ CSE Programs for SSI Children in Single-Parent Families Appendix III 28 Comments From the Social Security Administration Appendix IV 29 Comments From the Department of Health and Human Services Related GAO Products 30 Tables Table 1: Receipt of Child Support Services by Children Receiving 6 SSI in Single-Parent Families in Three States Page 18 GAO/HEHS-99-11 SSI and Child Support Contents Table 2: Parental Reporting of Child Support Collections by 10 Three States’ CSE Programs for SSI Children in Single-Parent Families Table 3: Estimated Annual SSI Overpayments Resulting From 11 Nonreporting of Child Support Collected by Three States’ CSE Programs Table I.1: Calculation of Estimated Annual SSI Benefit 23 Reductions From Providing CSE Services to Additional SSI Children Table II.1: Receipt and Reporting of Child Support Services and 25 Income, SSI Children in Single-Parent Families in Florida, New York, and Texas Table II.2: Recipient and Reporting of Child Support Services and 27 Income Provided Through the Florida, New York, and Texas CSE Programs for Children Receiving SSI in Single-Parent Families in Other States Abbreviations CSE Child Support Enforcement HHS Department of Health and Human Services SSA Social Security Administration SSI Supplemental Security Income TANF Temporary Assistance for Needy Families Page 19 GAO/HEHS-99-11 SSI and Child Support Appendix I Scope and Methodology We examined the extent to which opportunities exist for reducing SSI payments by increased receipt of services from and reporting of child support income received through the CSE programs in three states—Florida, New York, and Texas. We used SSA computerized records to identify SSI children under age 18 in single-parent families nationwide. We matched this information with data from the computerized CSE records of Florida, New York, and Texas to determine which of these children received CSE services and child support income and reported support received to SSA. We then estimated potential reductions in SSI benefits that could occur if CSE services were provided and support collected for additional SSI children and if unreported child support was reported to SSA. The period of review varied among the states because of the availability of state CSE data. It covered the 12 months ending July 1998 in Florida, the 8 months ending August 1997 in New York, and the 12 months ending July 1997 in Texas. While we did not independently verify the accuracy of the data, the SSI and the state CSE information systems are subject to periodic quality assurance reviews, and the CSE records serve as the official state records for purposes of administering the CSE programs. We did our work between May 1997 and December 1998 in accordance with generally accepted government auditing standards. To identify children receiving SSI in single-parent families, we obtained Identifying Children from SSA extracts of the Supplemental Security Record—SSA’s main Receiving SSI in administrative files and payment record for the SSI program—of all SSI Single-Parent Families recipients with birth dates on or after July 1, 1979, who were receiving benefits in September 1997. These records also indicated living arrangements of the recipients and whether child support income was being reported to SSA. Using these records, we identified children receiving SSI who were living with only one natural or adoptive parent. More specifically, we selected cases SSA coded as a child living with only a mother or a father, or as living with both parents if the person to whom benefits were paid was coded as being a stepparent. Also, we selected these cases only if the payee of the benefits was also specified by SSA as a natural or adoptive mother or father, or a stepparent. About 1,600 SSI children lived in families with two parents, one of whom was a stepparent. We did not include children in foster care, who constitute less than 3 percent of SSI children under 18, or children in institutions or other living arrangements. Also, we excluded from our study children who had reached age 18 in September 1998. Page 20 GAO/HEHS-99-11 SSI and Child Support Appendix I Scope and Methodology To determine the extent to which children receiving SSI in single-parent Determining Extent of families received CSE services and had support collected for them and Receipt and Reporting reported to SSA, we obtained extracts of CSE records from the states of of CSE Services and Florida, New York, and Texas and matched these records with the SSA records using the children’s names, months and years of birth, and Social Support Collections Security numbers. Since we obtained CSE records only from Florida, New York, and Texas, we cannot generalize the results of our analysis to SSI children nationwide.28 The state CSE records we obtained included all CSE cases active at any time during January through August 31, 1997, in New York; August 1996 through July 31, 1997, in Texas; and July 1997 through mid-July 1998 in Florida.29 These records indicated all child support collections made on the cases during those time periods. In our calculations, we multiplied New York’s CSE child support collections during the 8-month period by 1.516 to estimate a full year’s child support collections. We matched the CSE records of New York and Texas against September 1997 SSA records of SSI recipients under age 18 receiving benefits at that time. We matched the Florida CSE records against July 1998 SSA records of children who had been receiving SSI in September 1997.30 We considered as support that should have been reported for SSI benefit determinations any support collections made by CSE for children receiving SSI whose SSA record did not show child support income as having been reported to SSA. We limited our work to nonreporting of child support—we did not attempt to identify underreporting by comparing amounts of support collected by CSE with amounts reported to SSA. When there was more than one child on a CSE case, we calculated the amount of support collected for the SSI child by dividing the total amount of support collected by the number of children involved in the CSE case. This was necessary because CSE collection amounts on the records we 28 SSI children living outside these three states—and receiving CSE services—would not have a record in the Florida, New York, or Texas CSE program unless the children’s state of residence requested services from these three states. 29 Some families may receive state child support services other than those provided by the CSE agencies, including assistance with the collection and disbursement of child support payments subject to income withholding. Such services were not within the scope of this study. 30 About 12 percent of the children under age 18 in single-parent families in Florida receiving SSI as of September 1997 were no longer receiving benefits as of July 1998 and were not included in the analysis. We did not identify children who began receiving SSI between September 1997 and July 1998. Page 21 GAO/HEHS-99-11 SSI and Child Support Appendix I Scope and Methodology obtained were not separately identified by child.31 In addition, it is possible that in a case where other children in a family received TANF, support collected for an SSI child may have been inappropriately retained by the government along with support collected for the children receiving TANF, or that, in a case where an SSI child formerly received TANF, support collected may have been retained by the government as recoupment for prior TANF payments. In these situations, the support we attributed to an SSI child would not have been reportable to SSA because it was not received by the family. From the computerized files we reviewed, we could not determine precisely how much collected support was distributed to families. We based our estimates of the potential annual SSI benefit reductions that Estimating Potential could occur if more SSI children received CSE services on the number of SSI SSI Benefit children we identified as not receiving services. We identified as not Reductions From receiving CSE services SSI children who did not have a CSE record and did not have child support income reported on their SSA record.32 Increased Receipt of Child Support To estimate potential annual SSI benefit reductions, we assumed that the states’ CSE programs would collect support for only the same proportion of Services and SSI recipients not receiving CSE services as the program was collecting for Collections those receiving services. We further assumed that (1) the CSE system could collect for each SSI recipient not receiving CSE services an amount of support equal to the median amount collected for SSI recipients receiving services and (2) each SSI recipient not receiving CSE services was receiving the median SSI benefit received by all SSI potential child support recipients who were not receiving CSE services. As shown in table I.1, we calculated the total amount of potential annual support collections by multiplying the total estimated number of cases to be collected on by the median annual support amount collected. We multiplied this by 66 percent to derive the total amount of annual support collections countable as income for SSI, which in this case represented the amount of potential annual SSI benefit reductions.33 31 The amount of child support ordered to be paid can vary for children in the same family by the ages of the children, when the orders were issued, and other individual family circumstances. 32 There may have been SSI children who first received CSE services in the month of September 1997 through the New York CSE program or in the months of August and/or September through the Texas program that we counted as not receiving CSE services because the CSE records we obtained did not include those months. 33 The Social Security Act specifically provides that one-third of child support income be excluded from determining income for eligibility of children for SSI benefits. Page 22 GAO/HEHS-99-11 SSI and Child Support Appendix I Scope and Methodology Table I.1: Calculation of Estimated Annual SSI Benefit Reductions From Providing CSE Services to Additional SSI Children Factor and computation Florida New York Texas Total Additional SSI children in the state to receive CSE services (CSE record not found and no child support income reported on SSA record) 12,726 20,817 14,296 47,839 Multiplied by collection rate for other SSI children in the state for whom CSE Not collects child support 32.4% 20.9% 39.4% applicable Yields additional SSI recipients in the state for whom CSE will collect child support 4,128 4,355 5,636 14,119 Multiplied by median annual collections per other SSI children in the state for Not whom CSE makes collections $487 $350 $501 applicable Yields estimated annual support collections $2,010,247 $1,524,931 $2,825,468 $6,360,646 Multiplied by percentage of child support counted as income for SSI Not 66% 66% 66% applicable Yields estimated SSI benefit reductionsa $1,326,763 $1,006,454 $1,864,809 $4,198,026 Note: Intermediate calculations appear inexact because intermediary numbers are shown in rounded numbers, whereas actual calculations use nonrounded numbers. a Had 66 percent of the median child support collection been greater than the median SSI benefit, estimated benefit reductions would be limited to the amount of SSI benefits. Median SSI benefits received by SSI children not receiving CSE services were $5,928 in Florida, $5,870 in New York, and $5,808 in Texas. We did not directly assess the potential for child support collections among the SSI children not currently receiving services, and thus our estimate of SSI reductions may be understated or overstated. Some research indicates that a significant portion of noncustodial parents may have limited ability to pay child support, and it is possible that many of those custodial parents that have not pursued child support through the CSE program are associated with the noncustodial parents least able to pay. If that were so, our estimate would overstate the potential for child support collections among those not currently receiving child support services. However, we did not have information on the socioeconomic characteristics of the SSI children and their noncustodial parents needed to assess their ability to pay. In addition, the potential SSI benefit reductions we estimated may be overstated if some of the SSI recipients for whom we did not find a CSE record were already receiving child support income but not reporting it to SSA or had earlier received CSE services but had their cases closed, for example, because the identity of the noncustodial parent was not known, the noncustodial parent could not be located,34 or the 34 Federal regulations allow case closure after repeated attempts to locate noncustodial parents using multiple sources over 3 years. Page 23 GAO/HEHS-99-11 SSI and Child Support Appendix I Scope and Methodology noncustodial parent died.35 Also, some SSI children may have received CSE services, but we did not find a CSE record because the Social Security number and/or name and month and year of birth of the SSI child was in error on either the SSA or CSE record. We estimated the annual reduction in SSI overpayments that would occur if Estimating Potential all unreported child support collected for SSI recipients by the Florida, Reductions in SSI New York, and Texas CSE programs were reported to SSA for SSI benefit Overpayments by determinations. The estimated overpayment reduction that could occur for SSI children if unreported CSE collected support were reported to SSA is Improved Support equivalent, for each child, to 66 percent of the annual amount of child Reporting support collected by CSE or the annual SSI benefit, whichever is less. We performed this comparison for each SSI child recipient whose CSE record indicated that support collections were made but their companion SSI record did not indicate child support income had been reported. The summation of these amounts for all cases represents the total estimated overpayment reductions. 35 However, our work does consider such cases to some extent. The CSE cases against which we matched did include some closed cases—those either closed or identified for closure during the periods of our review. Page 24 GAO/HEHS-99-11 SSI and Child Support Appendix II Receipt and Reporting of Child Support Services and Income Through Three States’ CSE Programs for SSI Children in Single-Parent Families Table II.1: Receipt and Reporting of Child Support Services and Income, SSI Children in Single-Parent Families in Florida, New York, and Texas Residing in Florida Residing in New York Residing in Texas Total, three states Number Percent Number Percent Number Percent Number Percent Not receiving CSE servicesra No child support income reported to SSAb 12,726 42 20,817 50 14,296 58 47,839 49 Child support income reported to SSA 1,498 5 1,384 3 1,618 7 4,500 5 Subtotal, not receiving services 14,224 47 22,201 53 15,914 65 52,339 54 c Receiving CSE services Child support being collectedd No child support income reported to SSA 3,707e 12 2,741f 7 2,050g 8 8,498 9 Child support income reported to SSA 1,559 5 1,406 3 1,378 6 4,343 4 Subtotal, support being collected 5,266 17 4,147 10 3,428 14 12,841 13 Child support not being collected No child support income reported to SSA 10,388 34 15,197 36 5,065 21 30,650 32 Child support income reported to SSA 581 2 481 1 203 1 1,265 1 Subtotal, child support not being collected 10,966 36 15,678 37 5,268 21 31,915 33 Subtotal, receiving services 16,235 53 19,825 47 8,696 35 44,756 46 h Total 30,459 100 42,026 100 24,610 100 97,095 100 (Table notes on next page) Page 25 GAO/HEHS-99-11 SSI and Child Support Appendix II Receipt and Reporting of Child Support Services and Income Through Three States’ CSE Programs for SSI Children in Single-Parent Families Note: Data are for children under age 18 as of July 1997 and who had not reached age 18 by September 1998, residing with only one natural or adoptive parent, and receiving SSI in Florida as of July 1998 and in New York or Texas as of September 1997. a For whom we did not find a state CSE record for the 12 months ending mid-July 1998 in Florida; the 8 months ending August 31, 1997, in New York; and the 12 months ending July 31, 1997, in Texas. b SSA records did not indicate child support had been reported. c Children receiving SSI in single-parent families for whom we found CSE computerized records. d CSE records indicated child support had been collected for the SSI child. e We also found 366 SSI recipients from other states for whom the Florida CSE program collected child support, but this income was not reported to SSA. f We also found 1,275 SSI recipients from other states for whom the New York CSE program collected child support, but this income was not reported to SSA. g We also found 293 SSI recipients from other states for whom the Texas CSE program collected child support, but this income was not reported to SSA. h Numbers do not add because of rounding. Page 26 GAO/HEHS-99-11 SSI and Child Support Appendix II Receipt and Reporting of Child Support Services and Income Through Three States’ CSE Programs for SSI Children in Single-Parent Families Table II.2: Recipient and Reporting of Child Support Services and Income Provided Through the Florida, New York, and Texas CSE Programs for Children Receiving SSI in Single-Parent Families in Other States Served by Florida CSE Served by New York Served by Texas CSE program CSE program program Total, three states Number Percent Number Percent Number Percent Number Percent a Child support being collected No child support income reported to SSAb 366 17 1,275 18 293 17 1,934 17 Child support income reported to SSA 196 9 298 4 158 9 652 6 Subtotal, support being collected 562 26 1,573 22 451 26 2,586 23 Child support not being collected No child support income reported to SSA 1,464 69 5,129 70 1,204 70 7,797 70 Child support income reported to SSA 106 5 583 8 56 3 745 7 Subtotal, support not being collected 1,570 74 5,712 78 1,260 74 8,542 77 Total 2,132 100 7,285 100 1,711 100 11,128 100 Note: Data are for children under age 18 as of July 1997 and who had not reached 18 by September 1998, residing with only one natural or adoptive parent, and receiving SSI in July 1998 for children served by the Florida CSE program and in September 1997 for children served by New York and Texas CSE programs. a CSE records indicated child support had been collected for the SSI recipient. b SSA records did not indicate child support had been reported. Page 27 GAO/HEHS-99-11 SSI and Child Support Appendix III Comments From the Social Security Administration Page 28 GAO/HEHS-99-11 SSI and Child Support Appendix IV Comments From the Department of Health and Human Services Page 29 GAO/HEHS-99-11 SSI and Child Support Related GAO Products Supplemental Security Income: Action Needed on Long-Standing Problems Affecting Program Integrity (GAO/HEHS-98-158, Sept. 14, 1998). Welfare Reform: Child Support an Uncertain Income Supplement for Families Leaving Welfare (GAO/HEHS-98-168, Aug. 3, 1998). Child Support Enforcement: Certification Process for State Information Systems (GAO/AIMD-98-134, June 15, 1998). Supplemental Security Income: Opportunities Exist for Improving Payment Accuracy (GAO/HEHS-98-75, Mar. 27, 1998). Child Support Enforcement: Strong Leadership Required to Maximize Benefits of Automated Systems (GAO/AIMD-97-72, June 30, 1997). Supplemental Security Income: Timely Data Could Prevent Millions in Overpayments to Nursing Home Residents (GAO/HEHS-97-62, June 3, 1997). Supplemental Security Income: Administrative and Program Savings Possible by Directly Accessing State Data (GAO/HEHS-96-163, Aug, 29, 1996). Supplemental Security Income: SSA Efforts Fall Short in Correcting Erroneous Payments to Prisoners (GAO/HEHS-96-152, Aug. 30, 1996). (116000) Page 30 GAO/HEHS-99-11 SSI and Child Support Ordering Information The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. Orders by mail: U.S. General Accounting Office P.O. Box 37050 Washington, DC 20013 or visit: Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (202) 512-6061, or TDD (202) 512-2537. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. For information on how to access GAO reports on the INTERNET, send an e-mail message with "info" in the body to: email@example.com or visit GAO’s World Wide Web Home Page at: http://www.gao.gov PRINTED ON RECYCLED PAPER United States Bulk Rate General Accounting Office Postage & Fees Paid Washington, D.C. 20548-0001 GAO Permit No. G100 Official Business Penalty for Private Use $300 Address Correction Requested
Supplemental Security Income: Increased Receipt and Reporting of Child Support Could Reduce Payments
Published by the Government Accountability Office on 1999-01-12.
Below is a raw (and likely hideous) rendition of the original report. (PDF)