Child Support Enforcement: Families Could Benefit From Stronger Enforcement Program (Chapter Report, 12/27/94, GAO/HEHS-95-24). Greater federal leadership coupled with equally intensive state efforts could help the national child support enforcement program to better serve the families that depend on it. The dramatically rising number of children needing support--the child support enforcement caseload soared 180 percent between 1980 and 1992--has focused attention on federal and state efforts to force parents to support their children. However, these efforts have been stymied by management weaknesses that keep the Office of Child Support Enforcement (OSCE) from (1) effectively leading the program and the states, (2) judging how well the program is working, and (3) setting effective policies. Although the federal role is substantial--most program funding is federal--child support enforcement is very much a state activity. Today, states face common problems, such as increasing workloads that outpace resources, inadequate computer systems, and fragmented authority and unstandardized procedures among others. In response, states have developed a number of strategies, including augmenting their staffs with volunteers and contracting with private collection agencies, improving automation, and using innovative enforcement techniques. Many welfare reform proposals would further expand child support enforcement. Unless OSCE strengthens its management of its current program, it may have difficulty implementing any new responsibilities. --------------------------- Indexing Terms ----------------------------- REPORTNUM: HEHS-95-24 TITLE: Child Support Enforcement: Families Could Benefit From Stronger Enforcement Program DATE: 12/27/94 SUBJECT: Welfare benefits Aid to families with dependent children Public assistance programs Child support payments State-administered programs Federal/state relations Law enforcement Collection procedures Agency missions Strategic planning IDENTIFIER: HHS Child Support Enforcement Program AFDC OCSE Measuring Excellence Through Statistics Program California Delaware Arizona Iowa Kentucky Massachusetts New York Oregon Texas Virginia ************************************************************************** * This file contains an ASCII representation of the text of a GAO * * report. Delineations within the text indicating chapter titles, * * headings, and bullets are preserved. Major divisions and subdivisions * * of the text, such as Chapters, Sections, and Appendixes, are * * identified by double and single lines. The numbers on the right end * * of these lines indicate the position of each of the subsections in the * * document outline. These numbers do NOT correspond with the page * * numbers of the printed product. * * * * No attempt has been made to display graphic images, although figure * * captions are reproduced. Tables are included, but may not resemble * * those in the printed version. * * * * A printed copy of this report may be obtained from the GAO Document * * Distribution Facility by calling (202) 512-6000, by faxing your * * request to (301) 258-4066, or by writing to P.O. Box 6015, * * Gaithersburg, MD 20884-6015. We are unable to accept electronic orders * * for printed documents at this time. * ************************************************************************** Cover ================================================================ COVER Report to the Chairman, Subcommittee on Federal Services, Post Office and Civil Service, Committee on Governmental Affairs, U.S. Senate December 1994 CHILD SUPPORT ENFORCEMENT - FAMILIES COULD BENEFIT FROM STRONGER ENFORCEMENT PROGRAM GAO/HEHS-95-24 Child Support Enforcement Abbreviations =============================================================== ABBREV ACF - Administration for Children and Families AFDC - Aid to Families With Dependent Children CSE - child support enforcement GPRA - Government Performance and Results Act of 1993 HHS - Department of Health and Human Services METS - Measuring Excellence Through Statistics OCSE - Office of Child Support Enforcement OBRA'93 - Omnibus Budget Reconciliation Act of 1993 Letter =============================================================== LETTER B-250916 December 27, 1994 The Honorable David Pryor Chairman, Subcommittee on Federal Services, Post Office and Civil Service Committee on Governmental Affairs United States Senate Dear Mr. Chairman: This report, prepared at your request, examines the national child support enforcement (CSE) program at both the federal and state levels. It discusses federal program management issues and what some state CSE programs are doing to overcome barriers hampering their child support enforcement efforts. The report makes recommendations to the Secretary of Health and Human Services to improve CSE (1) planning, (2) performance measurement, (3) federal accountability, (4) audits, and (5) program funding structure. As arranged with your office, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from its issue date. At that time, we will send copies of the report to the Secretary of Health and Human Services; the chairmen and ranking minority members of the House Subcommittee on Human Resources, Committee on Ways and Means and the Senate Committee on Finance; and other interested parties. We also will make copies available to others upon request. This report was prepared under the direction of Jane L. Ross, Director, Income Security Issues. Please contact David P. Bixler, Assistant Director, at (202) 512-7201 if you have any questions. Other GAO contacts and major contributors to this report are listed in appendix VI. Sincerely yours, Janet L. Shikles Assistant Comptroller General EXECUTIVE SUMMARY ============================================================ Chapter 0 PURPOSE ---------------------------------------------------------- Chapter 0:1 Nonpayment of child support contributes to childhood poverty as well as to increases in the number of families receiving welfare.\1 Today, more than one-fifth of America's children live in poverty, and it has been estimated that half will live in single-parent families at some point in their lives. To help obtain the financial support noncustodial parents owe their children and to help single-parent families achieve or maintain economic self-sufficiency, the Congress established the intergovernmental child support enforcement program in 1975. In 1992, the federal Office of Child Support Enforcement (OCSE) reported a nationwide caseload of about 15.2 million cases, about 60 percent of which received welfare. The Chairman of the Subcommittee on Federal Services, Post Office and Civil Service, Senate Committee on Governmental Affairs, asked GAO to review the child support enforcement program to determine how the federal government could improve services to the states and to the families that depend on the program. GAO's review focused on (1) whether the program has essential management tools in place at the federal level to fulfill its mission, (2) how well OCSE has fulfilled its role in fostering the development of state child support enforcement programs, and (3) what the state programs are doing to overcome barriers hampering their efforts. GAO also examined the implications of welfare reform proposals and the impact of the Government Performance and Results Act of 1993 (GPRA). -------------------- \1 In this report, "welfare" refers to cash assistance provided to families under the Aid to Families With Dependent Children (AFDC) program. BACKGROUND ---------------------------------------------------------- Chapter 0:2 The federal government and the states share child support enforcement responsibilities. OCSE, within the Department of Health and Human Services (HHS), is responsible for providing leadership, technical assistance, and standards to develop effective state programs, which actually deliver child support enforcement services to families. Services include establishing paternity and support orders; locating noncustodial parents; updating support orders to be current with a noncustodial parent's income; obtaining medical support, such as medical insurance, from noncustodial parents; and collecting support payments. To help ensure state program effectiveness, the federal government uses a "carrot and stick" approach involving incentive payments, audits, and penalties. The federal government matches about two-thirds of state program administrative costs and makes incentive payments to states based on collections. Federal and state governments first became involved in child support enforcement activities with the aim of recovering government welfare costs. Child support owed by noncustodial parents of families receiving welfare was to be collected by state and local child support programs and then returned to the government, with a small portion going to the families. As the number of families receiving welfare rose in the 1970s, the Congress incorporated existing state and local efforts into the national child support enforcement program. Over time, the program expanded beyond the original aim of recovering welfare because the Congress believed that early enforcement of child support obligations could help prevent families' need for government support. For example, families that are not receiving welfare but that request program services must be served equally. RESULTS IN BRIEF ---------------------------------------------------------- Chapter 0:3 Greater federal leadership coupled with equally intensive state efforts could better position the national child support enforcement program to serve the families that depend on it. Dramatically increasing numbers of children needing support--the child support enforcement caseload grew 180 percent between 1980 and 1992--are focusing attention on federal and state efforts to enforce parents' responsibilities to support their children. However, these efforts have been hampered by management weaknesses that keep OCSE from (1) effectively leading the program and the states, (2) judging how well the program is working, and (3) setting effective policies. Because of declining resources, OCSE has diminished the level of technical assistance provided to state programs. Also, various organization and staffing changes have created communication problems between federal and state program officials. OCSE audits and data collection efforts, while satisfying legal requirements for monitoring and tracking the states' programs, do not provide either OCSE or the states with adequate information on program results. Now, under the impetus of GPRA, OCSE is getting started on management improvements that could position it to better serve states and families. OCSE has also proposed changing the incentive payment structure to encourage improved state performance. While the federal role is substantial--most program funding is federal--child support enforcement is very much a state activity. Today, states face common barriers such as increasing workloads that outpace resources, inadequate computer systems, and fragmented authority and unstandardized procedures among others. In response, states have developed a number of strategies, including augmenting their staffs with volunteers and contracting with private collection agencies, improving automation to help staff be more productive, and using innovative enforcement techniques. Some of the techniques various states have adopted are (1) requiring employers to report newly hired employees so parents who owe child support can be located, (2) using central lien indexes and tax record matching so parents' assets can be located, and (3) revoking driver's and professional licenses to encourage parents to pay what they owe. Many welfare reform proposals would further expand child support enforcement. Unless OCSE takes steps to strengthen its leadership and management of its current program, it may have difficulty implementing any new responsibilities. PRINCIPAL FINDINGS ---------------------------------------------------------- Chapter 0:4 THE CHILD SUPPORT ENFORCEMENT PROGRAM HAS LACKED ESSENTIAL MANAGEMENT TOOLS -------------------------------------------------------- Chapter 0:4.1 The child support enforcement program has lacked certain essential management tools to assess and improve current program performance. A well-articulated mission, programwide planning and goal-setting, and accurate data on program performance have not been available to guide program management. As the program expanded beyond welfare recovery, its mission became increasingly less clear. Legislative amendments expanded program efforts to include families not receiving welfare and activities, such as medical support enforcement, that do not focus directly on collections. In practice, this expanded mission has given the program competing priorities--without guidance from the federal level on how to manage those priorities. In the face of the program's expanded mission, OCSE's planning efforts focused on the agency and its processes, not on outcomes for either the national program as a whole or its own operations. Nor did these efforts seek input from key stakeholders, such as the Congress and the states. Only one national goal exists for the program, and this is a congressionally mandated standard for state performance in paternity establishment. Now, in response to GPRA requirements, OCSE is beginning a planning discipline that identifies priorities and outcomes. The child support enforcement program has also lacked accurate and consistent data that could be used to set goals for, assess, and improve program performance. Despite 20 years of required performance reporting, OCSE has not developed universally understood data definitions, and states collect data in ways that make aggregation and comparison impractical. OCSE and state officials acknowledge that needed data are not available, and OCSE has taken some actions on this issue. However, as of December 1994, OCSE was not in a position to know how the national program was actually performing. In concept, federal incentive payments reward state programs according to performance, but this funding mechanism has yet to achieve its potential. In practice, all states--regardless of performance--have received some incentive payments. Moreover, the amount of incentive payments depends on a state's collections and does not reflect other important activities, such as paternity establishment and medical support enforcement. The impact of the incentives is limited because states are not required to use incentive funds to expand their child support enforcement activities. Some states have used incentive payments for activities other than child support enforcement; others have used federal payments to offset the state part of matching funds for child support enforcement, according to a 1991 report by HHS' Office of the Inspector General. OCSE HAS NOT EFFECTIVELY FOSTERED STATE PROGRAM DEVELOPMENT -------------------------------------------------------- Chapter 0:4.2 As the Congress originally envisioned it, OCSE's role included fostering state program development by providing technical assistance and training, developing standards, and exercising federal oversight through audits. As HHS experienced workforce reductions in the 1980s, however, OCSE resources diminished. Technical assistance and training, which had formed a large part of OCSE efforts, virtually disappeared. In several instances, regulations were finalized after the effective dates of a law. In addition, an HHS-wide reorganization left OCSE with no organizational control over those HHS regional staff serving as contact points for the states on some program matters. State program staff had to contact various offices within HHS for different child support enforcement matters, and miscommunications between OCSE, HHS regional staff, and state program staff contributed to strained working relationships. By the early 1990s, OCSE's monitoring role had come into greater prominence, with more than half of remaining OCSE staff devoted to compliance audits. These audits focused on state compliance with federal requirements for administrative procedures and service delivery rather than on outcomes of state actions, such as how many paternities or support orders were established. State program staff said that while some audits had helped them gain state legislative support, the audits concentrated too much on administrative details and were too untimely to be a useful management tool for them, with audit reports sometimes issued 2 years after the period audited. OCSE recognizes that while the audits have spurred state actions, the audit approach needs to be changed to provide better insight into state program performance. OCSE is seeking to shift the emphasis from compliance with administrative procedures to ensuring state data integrity. OCSE, however, believes that this shift will require a legislative change. STATE CHILD SUPPORT ENFORCEMENT PROGRAMS FACE COMMON BARRIERS, USE MULTIPLE STRATEGIES -------------------------------------------------------- Chapter 0:4.3 Although states have received substantial federal funding for program development, at least five common barriers hinder state child support enforcement efforts: Workload continues to grow and become more time-consuming. Estimates of worker caseload ranged from 300 to 2,500 cases per worker, and officials believe that many cases now take more time than before. For example, the growing number of out-of-wedlock births means that more cases need paternity establishment--more than 40 percent of the caseload in one state. Child support enforcement program functions are only partially computerized. Some states do not have statewide computer systems. Other states have computerized some functions but others remain manual. One state had the computer capacity to seize noncustodial parents' bank assets but did not have caseload intake and parent location functions computerized. Another state had extensive case tracking capability but lacked the technology to identify assets and levy administrative liens through automation. State program staff lack control over local units. Some states wanted more uniform state implementation and were frustrated by dispersed program control. One state's audits documented that some local districts failed to implement state regulations for case processing. In another state, some judges did not follow state procedures. State legislatures do not always support state programs' proposed initiatives. Some states have been more successful than others in getting legislative support for new enforcement techniques, such as in-hospital paternity establishment; revocation of driver's and professional licenses of noncustodial parents who are delinquent in support payments; and using administrative, rather than judicial, processes for establishing support orders. Referrals from welfare offices lacked information that the child support enforcement program needed to do its work. GAO found in earlier work that poor coordination between welfare and child support agencies resulted in inadequate information about noncustodial parents, including identity, location, and earnings data.\2 The experiences of the state programs that GAO reviewed suggest that very little has changed. States that GAO reviewed were at different points in developing or adopting strategies to deal with the barriers they faced. The strategies included adding staff; refining AFDC intake procedures to facilitate child support enforcement; contracting out some functions to private entities; and using volunteers. States also used various techniques to garner legislative support and were making increased use of automation. The actions some states have taken also show that many decisions about adding resources, improving automation, and expanding the program's administrative processes lie within the control of state leadership and reflect the investments states can make in the program without the impetus of a federal mandate. -------------------- \2 Child Support: Need to Improve Efforts to Identify Fathers and Obtain Support Orders (GAO/HRD-87-37, Apr. 30, 1987). WELFARE REFORM PRESENTS ADDITIONAL OPPORTUNITIES AND CHALLENGES -------------------------------------------------------- Chapter 0:4.4 All the welfare reform proposals that GAO reviewed would add new enforcement tools but could also require more of OCSE and the states. In addition, some of the proposals would change the funding structure and OCSE's audit approach. Some proposals require greater centralization of state operations and give some states tools that they have sought, such as employer reporting of new hires or the ability to suspend professional or driver's licenses. However, new and expanded responsibilities may also strain OCSE's ability to effectively implement changes and increase state workloads. New demands under welfare reform could include broadening OCSE's role in developing and coordinating expanded automated systems at the state or federal level. In addition, state programs may also need greater technical assistance from OCSE and HHS to ensure effective state implementation. RECOMMENDATIONS ---------------------------------------------------------- Chapter 0:5 GAO is making several recommendations to the Secretary of Health and Human Services to strengthen OCSE's management and leadership. Among other things, these recommendations are aimed at establishing performance goals for OCSE; promoting greater federal accountability; reengineering OCSE audits of state performance to be more outcome-oriented rather than process-oriented; and revising the program funding structure to better support program priorities. AGENCY COMMENTS ---------------------------------------------------------- Chapter 0:6 HHS stated that the report provides a balanced appraisal of OCSE and the national program's accomplishments, and that GAO's recommendations are well taken. HHS provided additional information about actions that it has planned or has in progress that address several of GAO's recommendations. (See app. V.) While most of HHS' actions are appropriate, GAO has some specific concerns about HHS' response to establishing performance goals for OCSE; promoting greater federal accountability; and changing the audits and funding structure. (See pp. 95-97.) HHS also provided technical comments, as did program officials from selected states and several child support enforcement experts; their comments have been incorporated as appropriate. THE CHILD SUPPORT ENFORCEMENT PROGRAM: SIGNIFICANCE, ORGANIZATION, PERFORMANCE, AND REFORMS ============================================================ Chapter 1 Today, child support enforcement is an issue for more American families than ever before. The number of out-of-wedlock births and divorces has escalated in recent years, and it is estimated about half of American children will spend at least some time in single-parent families.\3 Many of these single-parent families, particularly those headed by young mothers, are at risk of welfare dependency. Child support is critical to these families' self-sufficiency. In the last 5 years, more and more single-parent families have sought government assistance from the Aid to Families With Dependent Children (AFDC) program, which requires recipients to cooperate with the child support enforcement (CSE) program. In addition, an increasing number of single-parent families not receiving AFDC have voluntarily sought CSE's help in obtaining child support from noncustodial parents. But in 1992, according to data reported by the federal Office of Child Support Enforcement (OCSE), only about 19 percent of the cases in the CSE program received any child support payments. In 1975, in response to the growth in the numbers of families receiving AFDC, the Congress created the national CSE program. The federal OCSE was created to monitor and help develop existing state and local child support enforcement programs. Child support collected for families receiving AFDC was to be returned to the government; child support collected for non-AFDC families would help these families avoid a need for cash assistance from the government. Over the years, the Congress steadily expanded CSE program enforcement tools, program requirements, and federal funding to states. In 1992, the federal government paid $1.7 billion to states in matching and other funds to serve a caseload of about 15 million families--8.7 million of whom received AFDC.\4 Because of his concerns, the Chairman of the Subcommittee on Federal Services, Post Office and Civil Service of the Senate Committee on Governmental Affairs asked us to review the CSE program to determine how the federal government could improve services to the states and the families who depend on state CSE activities. Our review focused on (1)whether the CSE program has the essential management tools in place at the federal level that a program needs to fulfill its mission; (2) how well the federal office, OCSE, has fulfilled its role in fostering the development of state CSE programs; and (3) what state CSE programs are doing to overcome barriers hampering CSE efforts. In addition, we examined implications of welfare reform proposals and the Government Performance and Results Act of 1993 (GPRA) for the CSE program. -------------------- \3 Family Impact Seminar, "Reducing Family Poverty: Tax-Based and Child Support Strategies," background briefing report (Washington, D.C.: Dec. 4, 1992). \4 The caseload estimate is from the Department of Health and Human Services' (HHS) Child Support Enforcement Seventeenth Annual Report to Congress, which contains data for fiscal year 1992. However, refer to pp. 44 and 45 of this report for the limitations of these data. THE CONGRESS CREATED THE NATIONAL CHILD SUPPORT ENFORCEMENT PROGRAM TO MEET GROWING NEED ---------------------------------------------------------- Chapter 1:1 Child support is a parental responsibility. However, because of the growing number of parents who fail to assume this responsibility, the Congress created the national CSE program in 1975 and significantly amended it in 1984 and 1988. The program's purpose is to strengthen existing state and local efforts to find noncustodial parents, establish paternity, obtain support orders, and collect support payments. Today, the program serves two populations: families receiving AFDC and those who do not. Although the services provided are essentially the same, collections are handled differently for the two groups, with a portion of the AFDC collections being returned to the state and federal governments and non-AFDC collections going directly to the families. Increases in both groups have caused caseloads to climb, and the service needs of both groups have become increasingly complex. ESTABLISHING CHILD SUPPORT: MAKING PARENTS RESPONSIBLE -------------------------------------------------------- Chapter 1:1.1 When two people, whether married or not, have a baby, they incur an obligation to provide for their child. When parents live apart, the parent not living with and providing day-to-day care for the child (the noncustodial parent) is expected to help the other parent (the custodial parent) provide for the child. Although children are entitled to support from parents who live apart, a legally binding document--a separation agreement or a support order--is needed to establish the appropriate amount of financial support and make the payment enforceable on the noncustodial parent. If the parents were married at the time of the child's birth, a support order is usually established when the parents separate or divorce. If the parents were not married when their child was born, the legal presumption of paternity conferred by marriage is lacking, and a permanent support order cannot be established or enforced on either parent until there is a legal determination of paternity.\5 Paternity is established in either of two ways: (1) through voluntary acknowledgment by the father or (2) if contested, through determination based on evidence such as scientific (genetic testing) and personal testimony. In such cases, paternity establishment confers on children the legal rights and privileges that a child born within a marriage has. Among them are rights to inheritance, medical and life insurance benefits, social security, and possibly veteran's benefits. -------------------- \5 In some cases, a temorary support order may be established before paternity is established. GOVERNMENT INVOLVEMENT TRIGGERED BY GROWING WELFARE CASELOADS -------------------------------------------------------- Chapter 1:1.2 Before the national CSE program, federal and state governments became involved in child support enforcement primarily through the AFDC program, which provided welfare benefits to children who had inadequate parental financial support. In 1950, the first relevant federal legislation required welfare agencies to notify appropriate law enforcement officials when AFDC was furnished to a child who had been abandoned by a parent.\6 Since 1968, federal statutes pertaining to the AFDC program have specifically required states to have programs for establishing paternity and obtaining support for children who receive AFDC. By 1975, an alarming rise in welfare costs resulting from out-of-wedlock birth rates and parental desertion as well as a growing demand to relieve taxpayers of the financial burden of supporting these families prompted the Congress to create the national CSE program as title IV-D of the Social Security Act. The Congress believed government welfare expenditures could be reduced, and to some extent prevented, by recouping AFDC benefits from noncustodial parents' child support payments. In addition, the Congress believed that earlier enforcement of child support obligations for families not receiving AFDC could help prevent these families from needing government support. The national CSE program incorporated the already existing state programs. -------------------- \6 See the Social Security Act Amendments of 1950, P.L. 81-734, sec. 321(b). AFDC AND NON-AFDC COLLECTIONS HANDLED DIFFERENTLY -------------------------------------------------------- Chapter 1:1.3 Families who receive AFDC, and families who do not, enter the CSE program in different ways; and the child support payments CSE collects for them are distributed differently. A mother who requests AFDC for her child is required as a condition of eligibility to assign her rights to child support to the government and to cooperate with the CSE program in locating and identifying the father of the child.\7 If the child support that CSE collects for an AFDC family does not, together with family income, make the family ineligible for AFDC, then all but $50 of the monthly support payment goes to the state and federal governments in proportion to their AFDC assistance to the family. The family continues to receive its full monthly AFDC grant plus the first $50 of the support payment. If the family does not receive AFDC, all the child support collected goes directly to the family.\8 Parents who do not receive AFDC are under no obligation to establish paternity or to use CSE services to collect child support. They do so, if at all, voluntarily. -------------------- \7 There are procedures to exempt custodial parents from cooperation for good cause. \8 If a family has stopped receiving AFDC, not all the child support collected may go to the family. In almost half the states, support paid above the current obligation amount is used to reimburse any child support owed to the state under the AFDC assignment provisions prior to payments owed to the family. THE CSE CASELOAD, ESPECIALLY NON-AFDC FAMILIES, HAS GREATLY EXPANDED -------------------------------------------------------- Chapter 1:1.4 Between 1980 and 1992, the nationwide CSE caseload\9 grew 180 percent, from 5.4 million to 15.2 million cases.\10 Part of that growth stemmed from the increase in the number of single-parent families receiving AFDC. The average monthly number rose more than 1 million in the past 5 years, increasing from nearly 3.6 million families in 1989 to just over 4.6 million in 1993. The larger portion of the CSE caseload increase, however, was in non-AFDC cases, which grew from 16 to 43 percent of the caseload during this period. In fiscal year 1992, there were 8.7 million AFDC cases and 6.5 million non-AFDC cases. Figure 1.1 shows the increase in the CSE caseload from fiscal year 1980 through 1992. Figure 1.1: Child Support Enforcement Caseload, Fiscal Years 1980-1992 (See figure in printed edition.) Source: OCSE annual reports. -------------------- \9 Throughout this report, CSE caseload refers to all cases enforced under title IV-D of the Social Security Act--both non-AFDC recipients who specifically request services and AFDC recipients who are automatically referred to and required to participate in the CSE program. \10 State data about 1992 caseloads and expenditures were the most recent available about state programs. INCREASING NUMBER OF YOUNG, UNWED PARENTS MAKES CASELOAD INCREASINGLY DIFFICULT TO SERVE -------------------------------------------------------- Chapter 1:1.5 Today, the CSE caseload is more difficult to serve than in the earlier years because there are more young, unwed parents. Various studies indicate that these cases usually require more services and are more difficult to successfully serve than those involving married or formerly married parents. For example, cases involving unwed parents require paternity establishment, which may be contested rather than acknowledged voluntarily. Also, unwed parents are usually less able and less willing than married or formerly married parents to pay child support. Such parents are generally younger and have less education and earning capacity. They are likely to be unemployed or unemployable and have little or no income with which to pay support--some are still in high school. In addition, some unwed parents are not aware of or do not fully understand the parental responsibilities that CSE agencies try to enforce. The growing size and complexity of the CSE caseload reflect major changes in demographic characteristics of the American family. High rates of divorce and out-of-wedlock birth have resulted in more children living in single-parent households. For example, the number of births to unmarried mothers increased 82 percent between 1980 and 1991. During the same period, the proportion of out-of-wedlock births rose from 18.4 to 29.5 percent of all births.\11 U.S. Bureau of the Census data indicate that there were 9.8 million custodial mothers aged 18 years and older in 1990--39 percent more than in 1979, the first year for which this information was available.\12 The increasingly difficult-to-serve nature of the CSE caseload and the links between nonpayment of child support, poverty, and increases in families receiving AFDC benefits are underscored by socioeconomic statistics. In 1989, according to U.S. Census data, 62 percent of all custodial mothers did not receive child support payments, largely because they lacked support orders.\13 In that same year, 32 percent of all custodial mothers were living below the poverty threshold, and three-quarters of them had not received child support.\14 Receiving child support appears to be a particular problem for never-married custodial mothers and their children--they made up 52 percent of the AFDC caseload in 1991, up from 21 percent in 1976.\15 Census data indicate that of all never-married custodial mothers, 75 percent did not have child support orders and slightly more than half had household incomes below the poverty level in 1989. -------------------- \11 Child Support Enforcement Seventeenth Annual Report to Congress, pp. 3-4. \12 Current Population Reports, Series P-60, No. 173, Child Support and Alimony: 1989 (Washington, D.C.: 1991), tables A-D. \13 Of 6.2 million custodial mothers who did not receive support, 4.2 million did not have support orders. Of the others, 1.2 million had orders but did not receive any payments, and 800,000 had orders but payments were not due that year. \14 Of those not receiving support, 75 percent did not have an order. \15 Families on Welfare: Sharp Rise in Never-Married Women Reflects Societal Trend (GAO/HEHS-94-92, May 31, 1994) and Families on Welfare: Teenage Mothers Least Likely to Become Self-Sufficient (GAO/HEHS-94-115, May 31, 1994). PROGRAM RESPONSIBILITY IS SHARED BY FEDERAL, STATE, AND LOCAL GOVERNMENTS ---------------------------------------------------------- Chapter 1:2 CSE is an intergovernmental program involving the federal, state, and local governments. Federal responsibility for the CSE program lies within HHS in the Administration for Children and Families (ACF). Within ACF, OCSE manages the CSE program at the federal level. A separate office within ACF, the Office of Information Systems Management/CSE Information Systems, approves, monitors, and certifies state information systems projects for all ACF programs, (funded under the Social Security Act), including CSE, and reports on CSE matters to the director of OCSE. In addition, ACF staff in HHS regional offices are responsible for review and approval of states' plans for all ACF programs, including CSE, AFDC, Job Opportunities and Basic Skills Training, and others,\16 and for ensuring consistent and uniform adherence to federal requirements governing all ACF formula and entitlement programs. Figure 1.2 displays the organization of CSE responsibilities within HHS. OCSE is responsible for providing leadership, technical assistance, and standards. When the Congress established the CSE program, it envisioned an aggressive federal role in ensuring that states actually develop strong and effective CSE programs. Title IV-D of the Social Security Act mandates that OCSE provide technical assistance to states to help them plan, develop, design, and establish effective CSE programs. The act also directs OCSE to establish standards for state program effectiveness and monitor the operation of state programs through periodic audits. To help ensure program effectiveness, OCSE has the authority to assess financial penalties if an audit reveals a state has failed to meet certain program standards. In addition, OCSE is responsible for maintaining relationships with federal, state, and local government officials, private organizations, and for individuals interested in the CSE program, and for coordinating and planning CSE activities to maximize program effectiveness. State CSE agencies are responsible for all activities leading to securing financial support and medical insurance coverage from noncustodial parents for children involved in CSE cases. To meet federal requirements and receive federal funds, state CSE programs must have HHS-approved plans indicating compliance with federal law and regulations and must operate programs in accordance with those plans. HHS can withhold federal funding from states that do not have approved plans. There are significant differences in the ways state CSE programs are organized, where they are placed in the states' organizational hierarchies, what relationships exist between the CSE program and other state agencies, and the policies and procedures that are followed. These characteristics usually vary by traditional state-local service delivery structures, levels of court involvement required by state family law, population distribution, and other state variables. For example, some state child support agencies operate their CSE programs with state funds through a network of regional offices, while others share the funding with and supervise county and other local jurisdictions' operations. Under both circumstances, certain program responsibilities may be contracted out to other entities. In addition, the state CSE agencies are not all in human services agencies; in a few states, CSE is in the state department of revenue or the attorney general's office. -------------------- \16 These include foster care and adoption assistance, child welfare, homeless youth, child care, and developmental disabilities. FUNDING DESIGNED TO CREATE A FEDERAL-STATE PARTNERSHIP ---------------------------------------------------------- Chapter 1:3 The CSE funding structure was designed to share program costs between the federal and state governments. The federal government matches 66 percent of states' administrative costs and 90 percent of their costs of developing management information systems\17 and certain other services related to paternity establishment. The federal government also pays incentives to states for collection efficiency. Incentives are calculated separately for AFDC and non-AFDC collections by dividing each by total program administrative costs. Incentive payments for AFDC collection efficiency range in amounts equal to 6 to 10 percent of the AFDC collections. Incentive payments for non-AFDC collections also range in amounts equal to 6 to 10 percent of non-AFDC collections but cannot be greater than 115 percent of the incentive payments paid for AFDC collection efficiency. These incentive payments are funded from the federal portion of recovered AFDC collections. Though federal matching funds are restricted to CSE program costs, states may use the incentive payments to fund programs other than CSE. They must, however, share incentives with local governments that bear some of the administrative costs of the CSE program. -------------------- \17 The 90-percent federal match for systems development costs expires at the end of fiscal year 1995. SERVICE DELIVERY IS DIVERSE AND COMPLEX ACROSS THE NATION ---------------------------------------------------------- Chapter 1:4 CSE is an administratively complex program. While the basic services are the same in every location, different families need different services. In addition, how services are delivered varies geographically. The common services mandated by federal law include client intake (including case establishment); locating noncustodial parents; establishing paternity; establishing, reviewing, and modifying support orders; enforcing financial and medical support obligations; and collecting and distributing support payments. Appendix I briefly describes these services and some of the activities and organizations involved with each of them. Families entering the CSE program require different combinations of these services at different times, and thus their cases flow through the program on different pathways. In some cases, the child's paternity has not been established and the location of the alleged father is unknown. In these cases, the custodial parent needs the CSE program to help with every step: locating the alleged father, establishing paternity and a child support order, enforcing the order, and distributing collections received. In other cases, the custodial parent may have a child support order and CSE agencies must periodically review and possibly modify the order to ensure continued conformity to state guidelines.\18 Order amounts may be increased or decreased with changes in noncustodial parents' ability to pay, for example, because of wage increases or job losses. Figure 1.3 illustrates various combinations and sequences of services that cases require as they flow through the program. Figure 1.3: CSE Services and Case Flows (See figure in printed edition.) Delivery of CSE services frequently involves the executive, legislative, and judicial agencies of state and local governments, such as tax collection agencies, the courts and district attorneys, as well as private institutions, such as banks, employers, and credit bureaus. At the local level, both enforcement and paternity establishment may involve the courts to varying degrees. One study of paternity establishment practices in a sample of nearly 250 counties found that the local agency responsible for child support was the department responsible for human services in 75 percent of the counties.\19 Only 43 percent of the counties, however, had paternity establishment handled solely by the department of human services. In most other counties, paternity establishment was a joint responsibility of the human services agency and a prosecuting or private attorney. Paternity and support order establishment and enforcement may be handled through a judicial or administrative process, depending on the state in which the custodial parent lives. Under a judicial process, authority to make paternity judgments, establish support orders, and take enforcement actions rests with the courts; proceedings take place in a legal setting and typically may involve district attorneys, state's attorneys, county attorneys, judges, legal aid societies, or private attorneys. In contrast, under an administrative process, an agency of the executive branch of state government has authority to administer certain aspects of state child support law or regulation without court approval being necessary for legally binding actions. The degree to which administrative rather than judicial processes are used varies across all states and within states. The functions and enforcement actions to which administrative processes are applied differ, as well as the degree to which the processes are used; the spectrum of processes is described as ranging from purely administrative to quasi-judicial to purely judicial.\20 -------------------- \18 All AFDC cases must be reviewed every 3 years, and non-AFDC cases must be reviewed at the request of either parent. \19 Pamela Holcomb, Kristin Seerfeldt, and Freya Sonenstein, Paternity Establishment in 1990: Organizational Structure, Voluntary Consent, and Administrative Practices (Washington, D.C.: The Urban Institute, 1992). \20 In most states, however, even those using purely administrative processes, judges still enter support orders in divorce proceedings. PROGRAM RESPONSIBILITY AND TOOLS HAVE EXPANDED WHILE FEDERAL EXPENDITURES HAVE GROWN ---------------------------------------------------------- Chapter 1:5 To help meet the increasing challenge of child support enforcement, the Congress has significantly amended CSE legislation over the years. In addition, federal expenditures have grown considerably. Most states continue to report net savings while federal net costs for the CSE program have increased. LEGISLATIVE PROGRAM EXPANSION -------------------------------------------------------- Chapter 1:5.1 Over the years, the Congress has expanded the responsibilities of the CSE program. Under the Child Support Enforcement Amendments of 1984 (P.L. 98-378), states are to treat AFDC and non-AFDC cases the same. The amendments also established the current funding structure. The Family Support Act of 1988 (P.L. 100-485) and the Omnibus Budget Reconciliation Act of 1993 (OBRA'93, P.L. 103-66) strengthened requirements for paternity establishment, medical support, and support order fairness and currency. In addition to expanding program responsibilities, the Congress also expanded program tools. The 1984 amendments, for example, required universal wage withholding provisions in all support orders, interception of federal tax refunds, and reporting delinquent payers to credit bureaus. The Family Support Act encouraged states to establish voluntary, civil paternity acknowledgment procedures and provided for establishment of paternity standards that states must achieve to avoid being penalized with reduced funding, among other things. Appendix II lists the major requirements. More recently, OBRA'93 strengthened the program by (1) requiring increased opportunities for unwed parents to voluntarily acknowledge or otherwise establish paternity in a less complicated and more timely manner, including in hospitals; (2) removing impediments to the enforcement of health insurance provisions in child support orders; and (3) strengthening the paternity establishment performance standards initiated by the Congress in the Family Support Act. INCREASED FEDERAL FINANCIAL EXPENDITURES FOR STATE CSE PROGRAMS -------------------------------------------------------- Chapter 1:5.2 Between 1980 and 1992, annual federal expenditures for matching funds to state CSE programs increased more than four-fold, from about $350 million to more than $1.3 billion. Figure 1.4 illustrates this trend. Federal incentive payments to states, paid out of the federal portion of the recovered AFDC funds, also increased, from $72 million in fiscal year 1980 to $339 million for fiscal year 1992. (See fig. 1.5.) Figure 1.4: Federal Expenditures for Matching Funds to State CSE Programs, Fiscal Years 1980-1992 (See figure in printed edition.) Source: OCSE annual reports. Figure 1.5: Federal Incentive Payments to State CSE Programs From the Federal Share of AFDC Collections, Fiscal Years 1980-1992 (See figure in printed edition.) Source: OCSE annual reports. FEDERAL NET COSTS FOR THE CSE PROGRAM HAVE INCREASED WHILE MOST STATES REPORTED NET SAVINGS -------------------------------------------------------- Chapter 1:5.3 From 1980 to 1992, federal government spending on state CSE programs increasingly outpaced what the federal government saved through AFDC recovery. During that period, however, at least 75 percent of the states reported a net savings due to their CSE programs. In other words, federal payments to states for CSE--matching funds and incentives--were not offset by federally retained AFDC collections, but state-reported CSE costs were more than offset by the combination of their retained AFDC collections and federal matching funds and incentive payments. Through fiscal year 1988, there were net savings to the public; since that time, costs have exceeded savings.\21 (See table 1.1.) Table 1.1 Net Budget Savings/Costs for States, Federal Government, and Taxpayers From the CSE Program, Fiscal Years 1980-1992 (Dollars in thousands) Fisc State net Federal net al budget budget savings/ Net savings/ year savings costs costs to taxpayers ---- ------------ ------------------ ------------------ 1980 $230,152 $-102,698 $127,454 1981 260,969 -128,377 132,592 1982 307,309 -147,946 159,363 1983 312,296 -138,078 174,218 1984 365,522 -105,048 260,474 1985 317,334 -230,888 86,446 \a 1986 275,644 -263,177 12,467 1987 341,826 -337,278 4,548 1988 381,001 -355,424 25,577 1989 403,333 -480,056 -76,723 1990 338,469 -528,135 -189,666 1991 385,059 -586,454 -201,395 1992 434,317 -604,946 -170,629 ---------------------------------------------------------- Note: Federal net savings/costs refer to the federal share of AFDC collections less incentive payments and matching funds paid to the states. \a According to the Congressional Research Service, the sudden decrease in taxpayer savings in 1985 was caused primarily by the implementation of the $50 "pass through" to AFDC families, effective in fiscal year 1985. Source: OCSE annual reports. -------------------- \21 Both the federal and state governments may also experience savings in the AFDC programs by keeping or moving families off of AFDC through successful child support enforcement efforts. Such savings, however, are difficult to measure. PERCEPTION OF PROGRAM'S PERFORMANCE VARIES WIDELY ---------------------------------------------------------- Chapter 1:6 A widespread public perception exists that the CSE program is failing to adequately address the nationwide problem of child support nonpayment. Some state program directors, however, believe CSE performance has improved over the years, citing rising collections in recent years in the face of an increasingly large and complex caseload often involving teenage parents and out-of-wedlock cases. In part, disagreements about program performance stem from differing notions of what can realistically be expected from the child support enforcement system. Advocates for children, policy analysts, human service and CSE professionals, and members of the Congress have characterized CSE's effect on the national child support enforcement problem as ranging from "unacceptable" to "dismal." In support of this assessment, they cite Bureau of the Census and other national survey data on the size of the problem nationwide. These data indicate that 62 percent of all custodial mothers in the United States did not receive child support in 1989. One study has estimated the gap between total child support that should be paid nationwide and total child support paid as $34 billion in 1990.\22 Advocates and others also have made specific criticisms of CSE's program performance on its current caseload. Using data reported by states to OCSE, some have faulted CSE because only 19 percent of all custodial parents served by the CSE program in 1992 actually received support payments. These observers also point out that of the 15.2 million cases in the CSE program in fiscal year 1992, more than 6.7 million cases, or about 44 percent, were without orders to pay support. In addition, they cite the fact that states reported collecting only about one-fourth of the child support due for fiscal year 1992 and prior years. Others in the child support enforcement community, however, believe that CSE program performance has improved greatly in recent years in the face of growing caseloads, federal mandates, and increasingly complicated out-of-wedlock cases. Between 1980 and 1992, states reported that dollar collections steadily increased by over 400 percent, from about $1.5 billion to $8 billion. In addition, both AFDC recovery and the proportion of cases for which collections were achieved increased between 1980 and 1992 along with a rise in the caseload. The program's recovery of AFDC payments increased approximately 118 percent, from a national rate of 5.2 percent to 11.4 percent. Similarly, the proportion of cases for which some collections were received increased 36 percent, from a rate of 13.7 percent to 18.7 percent. In part, the disagreement between those who criticize the program and those who defend its accomplishments hinges on differing notions of what are realistic and achievable expectations for the program. Some expect the program to achieve support for more of the families now in the program and to reach out and help additional families. Others have more limited expectations for various reasons, including the ever-increasing number of out-of-wedlock births, the fact that some custodial parents choose not to pursue child support, and uncertainties about the ability of some noncustodial parents to pay support. The ever-increasing number of out-of-wedlock births requires CSE to establish paternity on more cases--a sometimes difficult and time-consuming task. In-hospital voluntary paternity acknowledgment is extending CSE's reach early and has been shown to be increasing the number of paternities established and the efficiency of establishment. However, this initiative depends on the cooperation of both parents and results to date suggest that much remains to be done. Research reported in 1991 from the state of Washington, a pioneer of this approach, showed that nearly 40 percent of unmarried supposed fathers were signing voluntary acknowledgments. Other state programs' report similar results. These results mean that more than half of unwed fathers will have to be pursued through other means if child support is to be collected. Complicating the question of what realistic expectations are for CSE is the fact that not all custodial parents needing child support are served by the CSE program, and some choose not to use its services. According to 1987 Bureau of the Census data, 162,000 teenaged mothers were not awarded or were awarded but did not receive regular child support payments. Nearly 45 percent said they did not receive payments because they did not pursue the matter or did not want child support. Similarly, results from state demonstration projects of review and modification of orders also illustrate the reluctance of some families to have CSE involved in their cases. Reasons that non-AFDC custodial parents gave for refusing to have their awards reviewed included (1) fear of jeopardizing current payments, (2) desire to avoid court proceedings, (3) belief the noncustodial parent is paying all he or she can afford, and (4) doubt that the review would result in payments. Finally, program efforts may not guarantee that cooperative noncustodial parents will be able to pay support. Our previous work has shown that 9 percent of noncustodial fathers aged 23 to 31 have no income to pay support.\23 Other research has revealed that 27 percent of the noncustodial fathers who did not pay child support in 1990 were unemployed for at least part of that year. In addition, fathers' ability to pay has been shown to influence their compliance with child support awards.\24 -------------------- \22 Elaine Sorensen, Noncustodial Fathers: Can They Afford to Pay More Child Support? (Washington, D.C.: The Urban Institute, Jan. 1994). \23 Child Support Assurance: Effect of Applying State Guidelines to Determine Fathers' Payments (GAO/HRD-93-26, Jan. 21, 1993). \24 Judi Bartfield and Daniel Meyer, Are There Really Deadbeat Dads? The Relationship Between Ability to Pay, Enforcement, and Compliance in Nonmarital Child Support Cases, DP#994-93 (Madison, Wis.: Institute for Research on Poverty, Mar. 1993). CSE AND FEDERAL REFORM INITIATIVES ---------------------------------------------------------- Chapter 1:7 The CSE program is at the forefront of congressional efforts to reform government operations in general and, more specifically, to reform welfare. In August 1993, GPRA became law and the nationwide CSE program was subsequently selected as a pilot for the implementation of GPRA. In addition, in late 1993 and 1994, several welfare reform proposals emerged from the Congress and the administration that included changes for the CSE program. CSE IS AMONG PILOTS FOR GOVERNMENT MANAGEMENT REFORM -------------------------------------------------------- Chapter 1:7.1 CSE is among the pilot projects for government management reform under GPRA. Unlike most other pilots, however, the CSE pilot encompasses the entire federal and state CSE program, not just federal operations. The Congress enacted GPRA to strengthen federal program management with goal-setting, performance measurement, and results-reporting requirements. GPRA is intended to accomplish several government reforms, including improved (1) federal program effectiveness and public accountability, (2) service delivery, and (3) congressional decision-making. To that end, GPRA requires all federal agencies to have strategic plans by September 30, 1997, and performance plans by fiscal year 1999. To initiate program performance reform and identify the best way to implement this new system across the federal government, GPRA requires a series of pilots. As a pilot, OCSE is required to have a fiscal year 1995 performance plan by the start of that fiscal year. An OCSE official said the plan was submitted to HHS' Assistant Secretary for Management and Budget on September 27, 1994. WELFARE REFORM PROPOSALS CALL FOR CHANGES IN CSE -------------------------------------------------------- Chapter 1:7.2 Major welfare reform proposals introduced in the 103rd Congress contained changes for the CSE program. Some proposals call for more aggressive and stringent paternity establishment procedures, central support award registries in states, and centralized state collection procedures, among other measures. In addition, some of the welfare reform proposals would add to OCSE's responsibilities by requiring it to establish and operate national databases for support orders and other information. SCOPE AND METHODOLOGY ---------------------------------------------------------- Chapter 1:8 In doing our work, we reviewed the history of the CSE program, focusing on the period from 1980 to 1992, the most recent year for which data were available at the time of our review. We interviewed current and former program staff in all parts of the CSE program--at OCSE, in HHS regional offices, and in state and local CSE programs--including caseworkers, local office managers, county attorneys, state CSE and social service program directors, HHS and state auditors, and others. We also talked with CSE professionals affiliated with private companies, universities, advocacy groups, and policy and social research organizations. We continually reviewed published literature on child support and attended meetings of professional associations concerned with CSE issues. We analyzed CSE program statistics as reported by states and published by OCSE and discussed these with officials. To identify barriers that hinder state program CSE efforts and how states are addressing the barriers, we performed case studies in eight states: Arizona, Iowa, Kentucky, Massachusetts, New York, Oregon, Texas, and Virginia. (See app. III.) We selected these states because they represented different CSE caseload sizes, both centralized and decentralized program structures, administrative and judicial case processing approaches, a variety of performance histories as reflected in data submitted for OCSE annual reports, and different regions of the country. In addition, we reviewed socioeconomic data, including rates of out-of-wedlock births, divorce, and unemployment for these states and others. The scope of our work did not include the issues surrounding interstate child support enforcement. To assess the potential implications of welfare reform on the CSE program, we reviewed several welfare reform proposals containing changes to CSE. (See app. IV.) We conducted our review from November 1992 to December 1994 in accordance with generally accepted government auditing standards. HHS provided written comments on a draft of this report. These comments are discussed in chapter 5 and included in appendix V. We also obtained comments on a draft of this report from selected officials in each of the states we studied and several child support enforcement experts. Their suggested revisions were incorporated, as appropriate, into this report. CSE PROGRAM LACKS ESSENTIAL MANAGEMENT TOOLS ============================================================ Chapter 2 The CSE program lacks essential management tools to help fulfill its mission. Missing are a well-articulated mission, programwide planning and goal-setting, and accurate, consistent data that can be used to measure progress toward goals. In addition, the program's incentive funding structure is weakly linked to program performance. As the need for CSE services has continued to grow because of rising divorce rates and numbers of out-of-wedlock births, an evolving history has given CSE a complex mission. Originally intended primarily to recover federal and state welfare costs from the child support payments owed to AFDC families, CSE today is required to serve AFDC and non-AFDC families equally. In addition, some services, such as enforcing medical support, do not result in child support collections. As a result, the program faces competing priorities. In the face of this expanded mission, CSE lacks a program plan and measurable program goals that could be used to set priorities. Only one national goal exists--paternity establishment--and that was mandated by the Congress. In addition, despite nearly 20 years of state reporting, program data are inadequate for decision-making because of unclear definitions from OCSE and inaccurate reporting by the states. Finally, CSE's funding structure has not been effective in achieving program improvements. The structure of incentive payments continues to reflect CSE's original, narrower mission of recovering welfare costs, and the link between incentive payments and state program performance is weak. CSE'S MISSION HAS BECOME COMPLEX ---------------------------------------------------------- Chapter 2:1 The CSE program's mission has become more complex over time as the federal government expanded program requirements but made few changes to critical program components, such as the funding structure. As a result, different stakeholders--federal, state, and local program officials; custodial and noncustodial parents; advocacy groups; and elected leaders--have differing expectations of the program. Initially, states and the federal government focused the program's mission on easing taxpayer burden through recovering AFDC payments. Over time, the program's mission has been broadened from primary concern with government economy to concern with the economic security of children in both AFDC and non-AFDC families. In addition, some program activities, such as establishing medical support enforcement, focus more on noncollection outcomes and may use program resources without generating increased child support collections in the short term or at all. Despite these changes to the mission, important program components, such as the performance indicators used in federal audits of state programs and the structure of federal incentive payments to states, continue to emphasize AFDC cost recovery and welfare savings. Lack of alignment among program expectations--recover AFDC but also serve all in need and provide services that do not result in child support collections--has left state programs in awkward positions with their state legislatures and executive leadership. As one state program director observed to us, "[M]ost states are caught in an identity crisis now . . .. [W]e sold our program when it [was] supposed to generate revenue to reduce state cost . . .. [T]he taxpayers were the clients who were reimbursed the welfare expense . . .. [W]e [are] moving now toward providing services to families, but the structure and the funding is [sic] not supportive. It's not clear . . .. I think a national agenda would be very desirable." Moreover, given the competing priorities in program mission and the differing pressures generated by program expectations and requirements, state program directors must decide how to spread their resources among required collection and noncollection activities. In addition, welfare reform may further expand CSE's mission. Some proposals broaden the program to include all child support cases in the country. CSE'S ORIGINAL MISSION OF RECOVERING WELFARE COSTS COMPLICATED BY EXPANDING REQUIREMENTS -------------------------------------------------------- Chapter 2:1.1 CSE's original emphasis on recovering welfare costs has diminished as expanding requirements created a much broader and more complicated program mission. In addition to welfare recovery, there is now more emphasis on activities that require a long-term view of the program's overall social as well as financial benefits for both AFDC and non-AFDC families, such as medical support and paternity establishment. When the national CSE program was established in 1975, it focused on recovering AFDC payments. The priority placed on recovering AFDC payments was supported by state laws and the funding structure. Many states had laws allowing courts to set current support orders equal to a family's public assistance grant to recover all of the AFDC payments. In addition, the incentive payment structure established in 1975 rewarded states only for recoveries of AFDC collections. Thus, the primary beneficiaries of the early program were taxpayers, not families and children. As noted previously, in 1984 and 1988, the Congress extended CSE's authority and services, thus expanding CSE's mission. While continuing a focus on welfare cost recovery, these amendments raised the priority of several other aspects of CSE. The 1984 legislation increased emphasis on welfare prevention by providing that states serve AFDC and non-AFDC families in the same way. Providing equivalent service to non-AFDC families, however, has significantly increased most states' CSE caseloads as described in chapter 1 and illustrated in figure 1.1. In some states, the non-AFDC caseload is greater than the AFDC caseload. In addition, 1984 amendments and subsequent legislation\25 expanded state responsibilities by mandating the inclusion of medical support in child support orders and the provision of services to families who only receive Medicaid. To avoid a potential reliance on Medicaid, these mandates require state programs to ensure that parents provide medical insurance for their children whenever they have a reasonable opportunity to do so. However, medical support enforcement requires an investment of staff and automated resources that does not produce child support collections. Amendments in 1984 further reduced the emphasis on welfare recovery by setting a $50 limit on the amount of child support that AFDC families could retain.\26 This change allowed AFDC families to keep some of the child support collected, while the government retained less. However, some state program officials in the states we visited commented that this provision is an administrative "nightmare" and it added greater complexity to the distribution of collections. The 1988 amendments continued to broaden and complicate the program's mission by requiring states to (1) meet performance standards for paternity establishment and (2) undertake periodic review of child support orders and states' guidelines governing them to keep orders up-to-date. Paternity establishment is an essential first step toward obtaining a child support award and collection. However, it can be time-consuming and may not result in collections for some time, if at all, depending on case complexity and collectibility.\27 Another potentially time-consuming and labor-intensive activity, periodic review and adjustment of child support orders, has also been shown to substantially increase collections, but for a small percentage of cases. In addition, states now must respond to requests for review and modification from both noncustodial and custodial parents. -------------------- \25 The Omnibus Budget Reconciliation Act of 1987 (P.L. 100-203, sec. 9141) amended program requirements to provide CSE services to families who receive Medicaid. \26 The 1975 legislation had set a $20 limit, but that provision expired in 1976. No limit was in effect thereafter until the 1984 amendments. \27 Case complexity and collectibility depend on such factors as whether the case is contested; whether location, blood tests, subpoenas, and other program efforts are required; and the noncustodial parent's ability to pay. AUDIT PERFORMANCE INDICATORS AND INCENTIVE PAYMENTS ARE AIMED AT RECOVERING WELFARE COSTS -------------------------------------------------------- Chapter 2:1.2 Despite the movement to broaden CSE's mission, important program components--audit performance indicators and incentive payments-- remain focused on AFDC cost recovery, not the broader goal of contributing to families' economic security. OCSE is required to audit every state program at least once every 3 years, and the audit performance indicators for state programs emphasize AFDC collections. In addition to determining regulatory compliance, the audits assess the performance of state CSE programs by three indicators, two of which focus on AFDC collections: (1) AFDC collections divided by AFDC payments (less payments to unemployed parents),\28 (2) non-AFDC collections divided by administrative costs,\29 and (3) AFDC collections divided by administrative costs. The incentive payment structure continues to give priority to AFDC collections despite the changes in 1984. While expanding program emphasis beyond recovery of welfare costs, the 1984 amendments changed the incentive structure to reward state efforts to make non-AFDC collections as well as AFDC collections. However, in this scheme, AFDC collection efficiency still has priority because non-AFDC collection incentive payments are capped at 115 percent of AFDC collections. That is, no matter how well a state program does in collecting non-AFDC support, its efforts may not be fully rewarded. In addition, the lack of incentive payments for medical support efforts have given these efforts a low priority. AFDC collection efforts, therefore, drive state performance. Furthermore, HHS' Office of the Inspector General found that some state and local officials are concerned that using collection efficiency for determining incentives tends to penalize states for incurring additional significant costs that are not expected to yield increased total collections, both AFDC and non-AFDC, during the same year, such as costs for in-hospital voluntary paternity establishment programs aimed at new unwed parents. -------------------- \28 States make AFDC payments to children in single-parent families and two parent-families, who are needy because of the unemployment of one of their parents. In the calculation for the first indicator, AFDC payments to two-parent families are subtracted from the total amount of AFDC benefits paid. \29 The administrative costs used in these calculations may not include laboratory costs incurred in determining paternity if a state chooses to exclude them. SOME STATE LEADERS STILL EMPHASIZE CSE'S RECOVERY OF WELFARE COSTS AND GENERATION OF ADDITIONAL STATE REVENUE -------------------------------------------------------- Chapter 2:1.3 Despite the program's shifting emphasis away from recovery of welfare costs, some state leadership--legislative and executive decisionmakers--still expect CSE to generate revenue, according to state CSE officials. In the eight states we visited, many program officials said that state leadership tended to focus more on the program's revenue potential than on the social goals of promoting families' economic security. Some state program directors told us that their states' interest in CSE as a revenue generator makes it difficult for them to achieve a balanced program. For example, a senior Oregon program official said that it is difficult to go to the legislature with a message balancing both social and fiscal goals. He explained that the only way to get more money from the legislature is to emphasize the program's money-making potential. If the program's management went to the legislature and talked about economic self-sufficiency for custodial families, he believed they would lose the budget battle. Overall, he concluded, dollar goals are this program's clearest mission and the easiest to defend. New York Office of Child Support Enforcement officials expressed similar sentiments about their state legislature's goals. These officials believed the legislature's primary focus is on collections. As a result, state CSE officials made a deliberate decision to link child support enforcement to revenue generation, rather than presenting it as a programmatic expenditure along with other social programs to obtain more resources for the program. To meet the fiscal expectations of their legislatures and executive leaders, some state programs set priorities and risk audit findings of deficiencies that could result in financial penalties. For example, OCSE's 1988 audit of Iowa's CSE program revealed medical support enforcement deficiencies, and program officials were not surprised. The CSE director acknowledged that, in response to legislative expectations that the program generate revenue, they had focused their limited staff and automated resources on revenue-generating program activities. As a result, they did not give medical support enforcement priority attention before the audit because of the staff time and automated resources it would have required. WELFARE REFORM MAY FURTHER EXPAND CSE'S MISSION -------------------------------------------------------- Chapter 2:1.4 Among other changes, some welfare reform proposals would expand the responsibilities of CSE to create a central registry containing all child support orders, even those not now enforced under the program. The states, through a central public agency, would be responsible for collection and distribution of the payments. All but one of the proposals for including all orders in the program have criteria that allow parents not to participate in the order registry and all would allow parents not to participate with the central state collection agency. These proposals, however, might further move the focus of the program's mission away from welfare recovery because a larger portion of previously unserved non-AFDC families would be included. PLANNING AND GOAL-SETTING EFFORTS FALL SHORT ---------------------------------------------------------- Chapter 2:2 In the face of multiple and sometimes competing expectations among stakeholders, OCSE's past planning and goal-setting efforts have failed to set program priorities or outcomes; they have focused on OCSE's activities rather than on the CSE program as a whole. OCSE has not had a planning process that focused on outcomes for (1) the program as a whole or (2) its role in leading the program. Furthermore, except for paternity establishment, national and annual program goals that states are expected to achieve have not been defined for the program. The paternity establishment goals were defined by the Congress when it legislated paternity establishment standards in 1988. While results-oriented planning has not been legislatively required of OCSE in the past, it is recognized as an important tool for effective program management, and GPRA now requires it.\30 In response to the GPRA mandate for all federal agencies to develop strategic plans by September 1997, OCSE has taken the initiative to develop a strategic plan early. In addition, OCSE is a GPRA implementation pilot agency and, as such, is also preparing an annual performance plan before one is required. While this new planning effort appears to be avoiding some past weaknesses by including stakeholders and establishing long-term goals for the national CSE program, several issues remain to be addressed as OCSE continues to build its planning discipline. -------------------- \30 Our previous work (and that of others) has identified the importance of strategic planning and outcome measurement to effective program management. See "Management Vision Needed" in Department of Education: Long-Standing Management Problems Hamper Reforms (GAO/HRD-93-47, May 28, 1993) and explanation of strategic management in Management of VA: Implementing Strategic Management Process Would Improve Service to Veterans (GAO/HRD-90-109, Aug. 31, 1990); Program Performance Measures: Federal Agency Collection and Use of Performance Data (GAO/GGD-92-65, May 4, 1992); U.S. Department of Agriculture: Revitalizing Structure, Systems, and Strategies (GAO/RCED-91-168, Sept. 3, 1991); and Social Security: Sustained Effort Needed to Improve Management and Prepare for the Future (GAO/HRD-94-22, Oct. 27, 1993). PLANNING EFFORTS HAVE NOT FOCUSED ON OUTCOMES -------------------------------------------------------- Chapter 2:2.1 Although expanding responsibilities established multiple priorities for the CSE program over the years, OCSE had not engaged in a planning approach that solicited stakeholders' ideas, identified long-term goals for implementing the program's mission, or focused on program outcomes. Instead, previous planning efforts have principally focused on OCSE, not the national program, with an emphasis on process tasks and activities instead of on the desired outcomes of these activities. Past OCSE planning efforts were abandoned or were focused primarily on implementing 1984 and 1988 amendments. An early effort in 1980, entitled "Strategic Plan for the 1980's," was developed by OCSE staff with input only from HHS regional staff, not other stakeholders, but this plan was dropped a year after it was produced. OCSE's more recent planning efforts, before its GPRA pilot status, focused on its own activities rather than on what it planned to accomplish with these activities. For example, OCSE's plan listed activities such as "award and monitor new technology transfer contract" and "conduct judicial training conference using judicial curriculum guide." However, the plan did not identify what outcomes were expected or desired from these activities. In addition, OCSE did not develop or identify the indicators for measuring the effectiveness of these activities in the plan, the tracking document, or the periodic progress reports. EXCEPT FOR PATERNITY ESTABLISHMENT, NATIONAL AND ANNUAL PROGRAM GOALS ARE MISSING -------------------------------------------------------- Chapter 2:2.2 Although OCSE has advised state programs to develop goals, it has not done so for the national program. In the face of OCSE failure to set national or annual goals, the Congress mandated performance standards for state paternity efforts\31 in 1988 and amended them in 1993.\32 These paternity establishment standards effectively set a national goal for all states to eventually attain and differing annual goals for states, depending on their past paternity establishment performance. The standards hold all states accountable for improving performance until they reach a maximum, rather than a minimum, level of performance. As shown in table 2.1, the paternity establishment standard requires state programs to continuously improve on their individual performance until at least 75 percent of the cases that need paternity established have it established. In addition, greater proportionate improvement is required from low performers. Table 2.1 Required State Paternity Establishment Percentages Based on Past Year's Performance (For CSE cases requiring paternity establishment) Percentage established Required annual increase ----------------------------- ----------------------------- Less than 40 6 percentage points 40 or greater but less than 5 percentage points 45 45 or greater but less than 4 percentage points 50 50 or greater but less than 3 percentage points 75 75 None ------------------------------------------------------------ Source: 42 U.S.C. 652(g). In the absence of federally established goals in other areas, some states have developed their own goals. The Texas legislature, for example, holds the CSE program accountable for meeting a series of annual output and outcome goals. Output goals are established for such things as the number of (1) paternity actions filed, (2) notices of delinquencies filed, and (3) income tax refund offsets submitted to the Internal Revenue Service.\33 In addition, the Texas CSE program has an efficiency goal for its collections-to-cost ratio. Outcome goals include the number of (1) paternities established, (2) orders established or modified, and (3) collections obtained. -------------------- \31 The Senate Finance Committee stated in its report on the 1984 CSE amendments that OCSE "has not fully implemented the requirements for the establishment of standards of effectiveness." Child Support Amendments, Report No. 98-387 (Apr. 9, 1984). \32 Because of technical problems with the modification, regulations have not yet been issued implementing the revised standards. \33 Since our visit to Texas in 1993, the CSE program has changed its output goals. It no longer measures paternity actions filed. OCSE IS BEGINNING TO BUILD RESULTS-ORIENTED PLANNING DISCIPLINE -------------------------------------------------------- Chapter 2:2.3 In response to GPRA's mandate, OCSE is beginning to build a planning discipline that includes stakeholders and defines expected results and the means of measuring achievement, not just processes and activities or the quantity of output. OCSE is developing a strategic plan intended to articulate the fundamental mission of the program and establish long-range performance goals to implement its mission. Furthermore, unlike past planning efforts that did not include program stakeholders, OCSE has assembled a core planning team of OCSE and HHS regional office officials and state program representatives. In addition, OCSE has engaged state CSE program directors in the planning process and held focus groups at a national CSE conference to obtain comments and suggestions on its draft strategic plan from advocates and CSE professionals. OCSE's August 10, 1994, draft strategic plan goes beyond what GPRA requires of a strategic plan. In addition to a program mission and two general goals focusing on paternity and order establishment, OCSE has defined numerous objectives and many quantitative performance indicators for each goal. In contrast to past planning efforts, the objectives and performance indicators address such key areas as paternity and order establishment and medical support enforcement as well as collections. Also, in terms of collections, an objective is "to increase the collection rate" and various performance indicators are listed that can be used to measure the accomplishment of this objective. Such an objective will help the program identify what progress can be made in getting support payments for a larger proportion of families rather than simply increasing the absolute dollars collected. As a GPRA pilot project, OCSE has also developed an annual performance plan for fiscal year 1995 before such a plan is required from all federal agencies. The annual performance plan is intended to provide a direct link between the program's longer-term goals and what state programs and OCSE's managers and staff will need to accomplish each year. According to GPRA and the Office of Management and Budget guidance, such plans should include several elements such as (1) one or more performance goals, (2) performance indicators that will be used in measuring the goals, and (3) a description of the means to be used to verify and validate measured values. According to an OCSE official, the performance plan was submitted to the Office of Management and Budget by September 30, 1994, but it was not available for our review. However, another OCSE official said that the performance plan is not so detailed as to contain outcome goals for OCSE itself. The same OCSE official said these matters may be addressed in an in-house OCSE operating plan. SOME ISSUES REMAIN TO BE ADDRESSED IN OCSE'S PLANNING EFFORTS -------------------------------------------------------- Chapter 2:2.4 As of September 1994, OCSE's new planning efforts under GPRA had avoided repeating some past planning weaknesses and omissions, but OCSE had yet to address several important issues including outcomes and effectiveness measures for its own activities. However, OCSE officials said they recognize that the planning process is ongoing and that the strategic plan is a "living document" to which they expect to make continuous revisions. OCSE's August 1994 draft strategic plan described some important activities OCSE will undertake to achieve some of the plan's objectives, but the plan does not define how OCSE will measure the results of these activities or the outcomes expected from services they currently provide to the states. Failing to associate outcomes with activities is a past weakness that OCSE has the opportunity to correct through its current planning efforts, although it is not required of the GPRA strategic plan. For example, the plan states OCSE will "conduct an in-depth assessment of OCSE regulations, services, standards, and programs to identify contradictory and confusing regulations, unnecessary or ineffective services, and standards that do not contribute to customer service." However, OCSE does not indicate how it will measure the accomplishments from this activity. In addition, OCSE has not defined outcomes for itself regarding the services it provides to the states such as operating the Federal Parent Locator Service and processing state requests for federal tax refund interception by the Internal Revenue Service.\34 OCSE has also not provided outcomes for future timeliness of regulations, responsiveness to states about audits and policy questions, and regional HHS office commitment and feedback. These are the types of issues that OCSE will need to address if it is to build a meaningful planning discipline. An OCSE official told us that OCSE's services and results will be addressed in subsequent OCSE operating plans. Other issues that have not yet been addressed by the planning process are the actions needed by OCSE or the Congress to align the existing funding structure and audit approach with the CSE program's broad mission. However, an OCSE official told us that changes to the funding structure and audit were contained in the Clinton administration's welfare reform proposal submitted to the 103rd Congress. -------------------- \34 Federal tax refunds may be intercepted by the Internal Revenue Service and sent to state CSE programs to satisfy unpaid child support obligations. BETTER DATA ARE NEEDED FOR ACCURATE PERFORMANCE MEASUREMENT ---------------------------------------------------------- Chapter 2:3 Despite nearly 20 years of annual performance reporting, program data remain seriously flawed because of inadequate reporting standards and limited state reporting capabilities. As we have reported in the past, the resulting lack of accurate and consistent data hinders meaningful planning, analysis, performance measurement, and management improvement.\35 OCSE is attempting to improve program data reporting, but its main effort has been on hold because of the transition between administrations in 1993 and pending welfare reforms that could alter reporting requirements. Given these circumstances, current data collection will not adequately support the performance reporting GPRA requires. -------------------- \35 See Child Support: Need to Improve Efforts to Identify Fathers and Obtain Support Orders (GAO/HRD-87-37, Apr. 30, 1987). STATES ARE REQUIRED TO REPORT PROGRAM DATA TO OCSE ANNUALLY -------------------------------------------------------- Chapter 2:3.1 OCSE is mandated to prepare an annual report to the Congress no later than 3 months after the end of each fiscal year using data reported by the states. This report is supposed to include, among other information, program costs and collections, OCSE costs and staffing, caseload size, and service data from the states. Since 1977, there have been 17 such annual reports with increasing amounts of information about the CSE program. OCSE HAS NOT DEVELOPED UNIVERSALLY UNDERSTOOD DATA REPORTING STANDARDS -------------------------------------------------------- Chapter 2:3.2 OCSE has not developed universally understood or applied definitions and procedures by which the individual state CSE programs can collect and report data. An unduplicated caseload count, essential for accurate national program performance measurement, for example, is difficult to obtain because of the lack of universally applied definitions and procedures. Within OCSE's definition of a "case," for example, differences occur in reporting because of different state interpretations. OCSE defines a CSE case as "every absent [noncustodial] parent who is now or may eventually be obligated under law for the support of one or more dependent children." Some jurisdictions establish separate child support cases for each child with a custodial mother for whom the father is not identified because their fathers may be different; however, not all jurisdictions do this. Duplicate case counting can also occur when an AFDC case becomes a non-AFDC case as the result of successful collection efforts. The family may be counted both as an AFDC case and as an AFDC "arrears-only" case because delinquent child support is still owed for the time the family was receiving AFDC. In addition, OCSE's description of when a case may be closed is open to interpretation. As a result, collection rate statistics as well as total caseload figures can be misleading since all states are not equally aggressive in closing cases. OCSE has defined the criteria for case closure, but it has not set a time frame within which a case must be closed. Consequently, states that keep cases open longer than others may report higher caseloads and lower collection rates than those that close cases as soon as they meet OCSE's case closure criteria. STATE DATA ARE NOT AVAILABLE, RELIABLE, OR CONSISTENT -------------------------------------------------------- Chapter 2:3.3 Data basic to meaningful assessment of CSE program performance at the state and local levels are not available from all states, reliable, or consistent. Information that is available in one of the states we visited but not reported to OCSE, for example, includes the proportion of cases needing different services, such as location of the noncustodial parent, enforcement, wage withholding, and collection monitoring. Without this information, OCSE and state programs cannot develop an accurate picture of caseload needs for resource planning or measure progress in serving families. For example, simply knowing the absolute number of paternities established in a given year does not give program managers or policy analysts a way to measure the need for such a service. Both OCSE and state program officials told us that statistical analyses using the state-reported data to determine effective practices among states would be highly flawed because of poor data quality and inconsistent reporting. The percentage of collections from wage withholding, for example, is significantly underreported by one state program we visited because of limitations in its payment processing system, according to program officials. While this information is reported to the Congress, the state program does not find it a key piece of information for managing its operations. OCSE's annual reports contain numerous footnotes submitted by states to explain year-to-year variances in their reported data; these footnotes reveal the nature and range of problems casting doubt on the reliability of program data. Explanations for data variances provided in state notes include such problems as the following: contamination of information, lax data entry by the local office, local failure to report data, previous counting methodology resulting in inflated statistics, and adjustments in reporting methods. Cost data are also questionable because of varied state practices. States use different methods to collect and compute expenditure data, including caseload distribution, cost allocation, time studies, and educated guesses. Accounts receivable reported by states differ according to their procedures for writing off debts, including statutes of limitations, judgment-setting practices, and public assistance debts laws. Furthermore, states cannot accurately break down administrative costs by service, such as establishing paternity, locating noncustodial parents, and distributing support collections. OCSE HAS TRIED TO IMPROVE DATA AND ESTABLISH PERFORMANCE INDICATORS -------------------------------------------------------- Chapter 2:3.4 Recognizing that data quality is crucial to better program management, OCSE has made efforts to improve data quality and establish outcome-based performance indicators; but its main effort was suspended in late 1993 until changes to CSE as part of welfare reform are known. OCSE's main effort in this area, Measuring Excellence Through Statistics (METS), was implemented specifically to improve the quality of state statistical and financial data submissions. It involved meetings with state CSE directors and ACF regional staff and soliciting comments from all state CSE directors. Active from February 1992 through late 1993, METS encompassed efforts to revise the forms and instructions states are required to use to report program data, develop ideas for focusing the audit process on outcomes, and develop results-oriented performance measures. As of April 1993, most reporting definitions and procedures had been completed, except for a definition for accounts receivable and a procedure for counting interstate cases that would protect against double counting. More recently, in September 1994, an OCSE official told us that the audit staff is trying to build on the METS effort by looking into how states interpret OCSE's data definitions as well as evaluating state data collection systems and the quality of state data. CURRENT DATA COLLECTION WILL NOT ADEQUATELY SUPPORT GPRA PERFORMANCE REPORTING -------------------------------------------------------- Chapter 2:3.5 OCSE's current data collection from the states will not be adequate to support GPRA's performance reporting requirements. In addition to strategic and annual performance planning, GPRA requires annual performance reporting starting in March 2000 on what was actually accomplished and what goals were not met. This is a critical element for measuring the success of the planning efforts because reporting provides essential feedback to managers, policymakers, and the public. Therefore, it is important that data collected regarding CSE results be accurate. Annual planning will be more difficult if the inaccuracies and inconsistencies of past reporting are allowed to continue. FUNDING STRUCTURE NEEDS REEXAMINATION ---------------------------------------------------------- Chapter 2:4 Although performance-based incentive payments and funding flexibility are consistent with GPRA's focus on encouraging results-oriented government,\36 several problems with CSE's funding structure, including the incentive system, warrant reexamination. In addition to not having good data, a major problem is the weak link between performance and incentives. Since states may use incentive payments to fund other programs, not all states reinvest incentive payments in the program; instead they deposit the payments in the states' general revenue funds to be used to fund other programs. Also, states that do reinvest the funds in CSE frequently use the payments to cover a variety of CSE operational costs, thus reducing state investment in CSE, rather than supplementing state investment. Welfare reform may change this practice because some proposals include new funding structures. -------------------- \36 The National Performance Review led by Vice President Gore also emphasized results-oriented government. INCENTIVE STRUCTURE WEAKLY LINKED TO PERFORMANCE -------------------------------------------------------- Chapter 2:4.1 Despite congressional expectations, the present incentive structure is weakly linked to performance. Since 1984, the incentive payments have been expected to "encourage states to develop and improve efficient, cost-effective child support programs which balance services for AFDC and non-AFDC cases, both interstate and intrastate."\37 But all states, regardless of their performance as defined by collection-to-cost ratio, receive some incentive payments. In addition, continued minimal performance will still earn a state incentive payments. In contrast to the paternity establishment standards, goals are not set for improvement in ratios either year to year or over the long term. In effect, the potential of this structure as a management tool to change state performance is limited. OCSE has recognized the shortcomings in the existing structure, but proposed changes have not yet been enacted. Under the present incentives structure, all states receive incentives of at least 6 percent of their AFDC and non-AFDC collections, regardless of their collections-to-cost ratio. States achieving ratios of between 1.4 and 2.8 can earn up to 10 percent of their collections depending on their achieved ratio. As the Congressional Research Service has observed, "under the current financing arrangement, states can run inefficient programs and still make a profit from the CSE program."\38 In practice, few states earn the maximum incentive payments. An analysis of the collection-to-cost ratios for AFDC collections for fiscal years 1986 through 1992 showed most states earned 6 to 7 percent incentive payments on AFDC collections. In addition, fewer than five states earned the maximum of 10 percent each year. See figure 2.1 for this distribution. Figure 2.1: Distribution of Child Support Enforcement Incentive Payments Based on AFDC Collection-to-Cost Ratio, Fiscal Years 1986-1992 (See figure in printed edition.) Source: OCSE annual reports. During the same period, fiscal years 1986 through 1992, several states earned non-AFDC incentives from 7.5 to 8.5 percent, and 6 to 18 states earned the maximum non-AFDC incentive. However, few of these states were able to take advantage of the full incentive earned because the dollar amount of the incentive payments is capped at an amount equal to 115 percent of the AFDC incentive payments. See figure 2.2 for this distribution. Figure 2.2: Distribution of Child Support Enforcement Incentive Payments Based on Non-AFDC Collection-to-Cost Ratio, Fiscal Years 1986-1992 (See figure in printed edition.) Source: OCSE annual reports. Note: States may not receive the full dollar amount reflected in the percentages because the dollar amounts are capped at 115 percent of the AFDC incentive payments. OCSE has recognized the limitations of the incentive payment structure and has made two separate legislative proposals to make the structure more performance-based. In 1991, OCSE proposed rewards for performance in such activities as paternity establishment, order establishment, and AFDC terminations resulting from child support collections. It also suggested reducing incentives for cost-effectiveness. In addition, the proposal would have required states to reinvest the incentives in CSE. Officials at OCSE, however, said that the proposal was not pushed because of the change in administrations. Currently, OCSE is proposing to eliminate incentive payments and replace the entire funding structure with a graduated matching formula that would recognize different levels of state program performance. Under this proposal, certain levels of performance in at least five key areas would be rewarded with a larger federal match: (1) paternity establishment, (2) support order establishment, (3) the portion of child support cases in which payments are received, (4) the amount of collections compared to support owed, and (5) cost-effectiveness. Proposed legislative language for this type of funding structure was incorporated in the administration's 1994 welfare reform bill.\39 -------------------- \37 Child Support Enforcement Amendments of 1983, Committee on Ways and Means, U.S. House of Representatives, H.R. Rep. No. 98-527, (Nov. 10, 1983), p. 41. \38 The Child Support Enforcement Program: Policy and Practice, U.S. Congress, House Committee on Ways and Means, Subcommittee on Human Resources (report prepared by the Congressional Research Service), 101st Cong., 1st sess., 1989, Committee Print WMCP 101-19. \39 The administration's proposal is contained in H.R. 4605 and S. 2224 introduced in the 103rd Congress. INCENTIVE FUNDS ARE SOMETIMES USED TO SUPPLANT RATHER THAN SUPPLEMENT OTHER STATE CSE FUNDING -------------------------------------------------------- Chapter 2:4.2 Although many states do use at least some portion of the incentive payments to fund their CSE programs, the funds are sometimes used as part of the state match for federal funds rather than supplementing the state CSE budget and further expanding the program. In 1991, HHS' Office of Inspector General reported that nine states had laws or regulations that specifically mandated how incentives must be used.\40 In those nine states, incentive payments were passed along to local CSE operations or credited directly to the CSE program. Of the remaining states, 36 said they directed all or some portion of incentive payments to CSE. Those states indicated that they used incentive payments for CSE activities, including ongoing CSE operations, improving CSE automated systems, piloting CSE demonstration projects, and funding additional CSE agency employees. Ten states had used incentive payments to fund special CSE projects such as (1) a support order review and modification demonstration, (2) contracting with a private collection agency to pursue very difficult cases, and (3) automation development and preparation of cases for data entry or conversion to new systems. Further work by the Office of Inspector General and a survey of states conducted by Delaware found that incentive monies were used to make up the states' share of the CSE programs' operating or administrative costs. In a follow-up audit of nine selected states in 1991, the Office of Inspector General found, "Federal CSE incentives generally were used for the CSE program, reducing or eliminating the need for state/local general funds to fulfill matching requirements."\41 In Arizona, for example, the audit revealed that 70 percent of the incentive payments was used to reduce state or local program costs. The Inspector General concluded, "[T]he primary impact of incentive payments is that state CSE programs are funded, sometimes exclusively, by the federal government through the federal cost share and incentives." In fiscal year 1992, approximately 84 percent of states' total CSE administrative costs were reimbursed by the federal government through matching funds and incentive payments. -------------------- \40 See Child Support Incentive Payments, HHS, Office of Inspector General, Report #OEI-05-91-00750 (June 1991). "States" refers to the 50 states, the District of Columbia, the Commonwealth of Puerto Rico, and the territories of Guam and the Virgin Islands. \41 See The Use and Equity of Child Support Enforcement Incentive Payments at Selected States, HHS, Office of Inspector General, Report #A-09-91-00034 (Apr. 22, 1992). The nine states were Alabama, Arizona, California, Kentucky, Michigan, New Mexico, New York, Pennsylvania, and Washington. WELFARE REFORM PROPOSALS INCLUDE DIFFERENT FUNDING STRUCTURES -------------------------------------------------------- Chapter 2:4.3 Three of the eight welfare reform proposals we reviewed would eliminate the existing incentive payment system and allow increasing federal match rates for different levels of performance. Under one, for example, states could earn a 5-percent increase in the match rate for improved performance establishing paternities and a 10-percent increase for improved overall performance. All three proposals would reward paternity establishment performance. In addition, one of the three proposals would reward state compliance with program and staffing standards. The other two propose aligning the funding structure with overall performance in a number of key areas, not just cost-effectiveness. Overall performance would include performance in (1) the areas of support order establishment, (2) child support cases in which payments are collected, (3) amount of collections compared with support owed, and (4) cost-effectiveness, as defined by the Secretary of HHS. By changing the funding structure to one that uses a matching formula exclusively, these proposals may eliminate some of the complexity in the current arrangement. The only expenditures that would be matched with federal funds would be CSE program expenditures. States would no longer receive incentive funds that (1) may be reinvested in the CSE program or spent elsewhere, (2) must be shared with local jurisdictions, and (3) are complicated to forecast because of the effect of the cap on non-AFDC incentives. Under all three proposals, the maximum federal matching rate a state could achieve would be equivalent to 90 percent of the state's costs--if that state achieved the highest level of performance. However, it is difficult to predict if any of the three proposals would actually increase total federal costs. We do not know how many states, if any, will achieve maximum performance under two of the proposals because the performance goals are not yet defined. Under the third proposal, a state would have to establish paternity in at least 80 percent of the cases needing paternity establishment, meet the program standards in at least 80 percent of the cases, and comply with staffing requirements set by the Secretary of HHS. To date, very few states have been able to establish paternity in 80 percent of the cases needing it, according to data reported by the states. OCSE HAS NOT EFFECTIVELY FULFILLED ITS ROLE OF FOSTERING STATE CSE PROGRAM DEVELOPMENT ============================================================ Chapter 3 Although the Congress envisioned that OCSE would take an aggressive role in ensuring that states develop strong and effective CSE programs, OCSE has not done so. Specifically, the Congress tasked OCSE with providing leadership, technical assistance, and standards for the CSE program as well as monitoring state programs. But as the program evolved, OCSE's ability to direct, assist, and communicate with state CSE programs did not keep pace with growing requirements. An HHS-wide reorganization left OCSE with little or no control over HHS regional staff who deal with state CSE program staff. In addition, its financial resources were reduced, nearly eliminating funding for training and technical assistance contracts to the states. Subsequently, communications between OCSE, HHS regional staff, and state program officials deteriorated, and working relationships became strained. On several occasions, OCSE has finalized regulations after the statutory effective dates of the legislated provisions, creating uncertainties for state programs. Also, while OCSE audits--its principal tool for monitoring state programs--have spurred state action in some cases, these audits have focused more on compliance with administrative procedures than on program outcomes. The audits have provided limited insight into how state programs could achieve better program results, and have come too late to be useful to some states. OCSE recognizes it needs to revise its audit approach and is taking steps to do so. Program personnel in many of the states we visited wanted more training and technical assistance from OCSE; some of these officials also expressed a desire to be more involved as OCSE drafts regulations. OCSE STAFF, FINANCIAL, AND TECHNICAL ASSISTANCE RESOURCES REDUCED ---------------------------------------------------------- Chapter 3:1 Since the mid-1980s, OCSE's ability to fulfill its mission has been affected by an HHS-wide reorganization and workforce reductions. The number of staff directly accountable to OCSE and OCSE's financial resources decreased. OCSE's capacity to provide on-site technical assistance to states, through both contracts and OCSE and regional staff, has nearly been eliminated. All areas in OCSE, such as policy development, program operations, and audits, experienced staff reductions; but today, more OCSE staff remain devoted to audits than technical assistance. HHS REORGANIZATION AND RESOURCE REDUCTIONS -------------------------------------------------------- Chapter 3:1.1 As the result of a fiscal year 1986 HHS-wide reorganization, regional staff devoted to CSE activities are no longer directly accountable to OCSE but instead report to HHS regional administrators. In 1986, OCSE was combined with five other major HHS programs, including AFDC, into the Family Support Administration to emphasize the family. (In 1991, the Family Support Administration became the Administration for Children and Families [ACF].) In addition to regional staff, OCSE's budgetary and administrative functions and automated systems unit were transferred to other Family Support Administration units. This reorganization changed the composition of OCSE staff and the amount of resources OCSE had at its disposal. To illustrate, in fiscal year 1986, before the reorganization, OCSE controlled 342 full-time positions--95 in regional offices and 247 in OCSE's Washington, D.C., office and area audit offices located around the country. In fiscal year 1987, however, after losing control of CSE regional office staff, OCSE's authorized positions dropped to 151. In addition to the loss of direct accountability from regional, financial, and information systems staff, OCSE's staff devoted to technical assistance, training, policy, planning, evaluation, audit, and research and regulations\42 was also reduced. From fiscal year 1980 to 1992, OCSE staff assigned to these activities decreased from 181 to 151. See figure 3.1. From fiscal year 1987 to 1992, staffing of the OCSE division responsible for training and technical assistance declined 37 percent, from 43 to 27. Figure 3.1: Staff Directly Accountable to OCSE, Fiscal Years 1980-1992 (See figure in printed edition.) Note: Fiscal year 1982 numbers not available. Source: OCSE annual reports. OCSE financial resources were also reduced beginning in the mid-1980s. Along with the staff resources that were removed from OCSE's control after the 1986 reorganization, OCSE's financial resources were reduced by over 50 percent. Part of this reduction included about $2 million in contracts for technical assistance, publications development, data processing, and other services. From fiscal year 1986 to 1987, these contracts declined from $5.3 million to $3.2 million. Total financial resources under OCSE's direct control in fiscal year 1992 were $10.4 million compared with $21.3 million in fiscal year 1986 when the regions were under OCSE's control. See figure 3.2. Figure 3.2: OCSE Financial Resources, Fiscal Years 1980-1992 (See figure in printed edition.) Note: The drop in OCSE's resources from $21.3 million in FY 1986 to $9.8 million in FY 1987 reflects an HHS reorganization during FY 1986 in which OCSE lost direct control of regional CSE staff and headquarters financial and information systems staff. Source: OCSE annual reports. -------------------- \42 That is, excluding financial and information systems management, and regional office staff transferred to other parts of ACF during 1986. TECHNICAL ASSISTANCE AND TRAINING TO STATES GREATLY REDUCED -------------------------------------------------------- Chapter 3:1.2 Despite its mission to provide technical assistance to state CSE programs, OCSE's resources for both direct and contracted work of this kind have been reduced. Throughout the 1980s, OCSE funded several contracts for training and technical assistance, but today has none. Also during the 1980s, particularly the early 1980s, both OCSE headquarters and regional staff provided extensive, direct, on-site technical assistance to the states and conducted formal management studies of local and state programs. Joint management studies are no longer done and on-site technical assistance is minimal in most regions. In fiscal year 1992, fewer than a quarter of OCSE's staff were devoted to providing training and technical assistance. Of OCSE's fiscal year 1992 staff of 151, 86 people, or 57 percent, were devoted to conducting compliance audits. OCSE funds for contracted training and technical assistance have been eliminated. OCSE officials said that in the early 1980s the office had contracting authority for $3 million in training and technical assistance. By fiscal year 1990, however, this had been reduced to $1 million and by fiscal year 1993 to $300,000. OCSE officials said they had no funds for contracted training and technical assistance in fiscal year 1994. OCSE's total contracts for publications development, printing, data processing, and other services, as well as training and technical assistance, were reduced from $7.1 million in fiscal year 1985 to $2.1 million in fiscal year 1992. At least six technical assistance efforts funded through contracts in the 1980s are no longer funded: (1) transferring best practices; (2) training state, local, and federal personnel; (3) assisting states in drafting legislation; (4) providing orientation and training to state judges and judicial and quasi-judicial staff; (5) providing links with the legal community of court clerks, the private bar, and prosecutors; and (6) publishing educational articles. In addition, in the first half of the 1980s, when HHS regional offices were still directly in the chain of command to OCSE, they provided extensive on-site technical assistance to state programs; they no longer do. They conducted formal management studies of local and state programs, sponsored conferences for judges, and met with and testified before state legislative committees. Regional staff also made presentations before public interest groups; state and local officials, including legislators, court personnel, judges, attorneys, and child support administrators; and others; and participated in educational panels. They also helped write television and radio public service announcements, assisted in drafting state legislation, and appeared on television and radio talk shows. At the time of our review, most HHS regions were providing minimal on-site technical assistance to states. OCSE and regional office staff jointly conducted comprehensive management reviews of state programs and provided on-site technical assistance during the early 1980s. Today, OCSE staff's technical assistance efforts are limited primarily to participating in national conferences, distributing pamphlets and self-help guides, writing letters, making telephone calls, and posting information on electronic bulletin boards. In fiscal year 1981, OCSE's program operations division performed 14 management reviews; whereas in fiscal year 1993, OCSE officials said they no longer do management reviews. The OCSE division director responsible for technical assistance told us the small staff and travel budget devoted to technical assistance severely limits their ability to compile and disseminate best practices and provide direct on-site training and assistance to state programs. In February 1993, 25 OCSE staff were devoted to training and technical assistance. At the same time, the travel budget for these staff was limited to $10,000. The OCSE official said that staff responsible for technical assistance are able to meet with regional offices only once a year; they have monthly telephone conference calls in lieu of more frequent personal contact. HHS CURRENT ORGANIZATION FOR CSE HAS NOT BEEN THE MOST EFFECTIVE ---------------------------------------------------------- Chapter 3:2 The current organization of CSE responsibilities within HHS has not been the most effective for providing program leadership. OCSE's ability to communicate, direct, and assist state programs is limited. For example, OCSE has no direct control over HHS regional resources for CSE, has no process for feedback from the regions on CSE work, and receives no definite up-front commitment of regional office resources to CSE. Furthermore, state CSE programs must deal with several points of contact at the federal level. Under this organization, misunderstandings and miscommunications of program policy and requirements have occurred between states, HHS regional offices, and OCSE. REGIONAL OFFICE CSE RESOURCES NOT ACCOUNTABLE TO OCSE -------------------------------------------------------- Chapter 3:2.1 OCSE has no direct control over CSE regional resources and no formal process for feedback from the regions. Before the 1986 reorganization that created the Family Support Administration (which later became ACF), regional CSE staff were directly accountable to OCSE.\43 Under the current organization structure, all regional staff report to regional administrators for ACF. These administrators must balance resources among all ACF programs, including CSE, to meet mandated ACF program requirements. However, there is no process for getting firm, up-front regional office commitment of resources to the CSE program--OCSE can only set priorities for areas of state programs to be examined by regions. In addition, there is no formal process for regional office feedback to OCSE on the results of their expenditure of resources on CSE. ACF's director of regional operations staff officials told us that it was impossible to determine the dollars and staff time spent by the HHS regional offices for CSE state technical assistance, training, or program reviews because of the intermingling of funds that occurred after the 1986 reorganization. Also, OCSE officials did not know if certain regional efforts cited in ACF's strategic plan for performance by regions were accomplished; they said the regions had not reported on them. For example, the plan stated that regions were going to develop joint plans with states to increase AFDC recovery and the number of paying cases, but OCSE did not know what had been done. -------------------- \43 The April 1991 reorganization combined the Family Support Adminstration, HHS' Office of Human Development Services and the Maternal and Child Health block grant programs administered by the Public Health Service into ACF, where OCSE remains today. SEVERAL POINTS OF CONTACT FOR STATES -------------------------------------------------------- Chapter 3:2.2 HHS' organization of CSE responsibilities affects communication with state programs. Currently, states must deal with at least five different HHS units for their CSE programs as illustrated in figure 3.3. An HHS regional official said it is confusing for the states to be dealing sometimes directly with OCSE for some things and other times with the regions. The nature of these contacts is summarized in table 3.1. Figure 3.3: HHS Points of Contact for State CSE Programs (See figure in printed edition.) Note: Highlighted boxes show states' points of contact within HHS. Table 3.1 HHS Points of Contact for State CSE Programs HHS organizational unit Nature of contact -------------------- ------------------------------------ OCSE, Washington, Federal tax refund intercepts; D.C. checks of Internal Revenue Service information; federal parent locator service; requests for waivers of program requirements; and verification of social security numbers with the Social Security Administration HHS regional offices State plan submission and approval; exemption requests; required data- reporting reviews; and financial systems reviews ACF Office of Matters concerning certification of Information Systems child support computer management and Management, information systems Washington, D.C. OCSE area audit Compliance audit scheduling and offices and issues Washington, D.C. HHS Office of Independent investigations of the Inspector General CSE program ---------------------------------------------------------- POOR COMMUNICATION BETWEEN OCSE, STATES, AND REGIONS -------------------------------------------------------- Chapter 3:2.3 The HHS reorganization placing CSE regional office staff in a different chain of command rather than having them report to OCSE impaired CSE communications between OCSE, the regions, and the states. An HHS regional CSE official said the added layers of bureaucracy impeded communication with states and their communication with OCSE. Similarly, OCSE officials confirmed there have been communication breakdowns with the regions. They said that regional staff have misconstrued information and communicated inaccurate information to the states. States have complained of vague and inconsistent policy interpretations from OCSE and HHS regions. One state CSE director described communication channels as "convoluted layers," similar to a "black hole" where things just "disappear." In addition, state representatives testified in 1991 before the congressionally established Commission on Interstate Child Support that CSE policy was often interpreted differently by HHS regional offices and OCSE auditors.\44 During our visits to eight states in summer 1993, program officials in one state cited OCSE guidance about the requirement to review and adjust support orders every 3 years as an example of unclear and untimely communication. Though the required implementation deadline was October 1993, these state officials were not clear what date should be used as the base date for the 3-year review cycle in situations in which families go off AFDC and later go back on. They could not discern from the regulations which of at least three possible dates could start the review cycle: (1) the date the family first went on AFDC, (2) the date the family went off AFDC, or (3) the date the family went back on AFDC. -------------------- \44 Supporting Our Children: A Blueprint for Reform, U.S. Commission on Interstate Child Support (Washington, D.C.: Government Printing Office, 1992), p. 249. The Commission was charged by statute to make recommendations on improvements to the interstate establishment and enforcement of child support awards. WORKING RELATIONSHIPS STRAINED ---------------------------------------------------------- Chapter 3:3 Working relationships between state programs and OCSE and HHS regions have been more adversarial than cooperative. State officials cite the need for more technical assistance and timely regulations from OCSE rather than policing of processes. According to state officials, the untimeliness of regulations has made it more difficult for them to comply with the law. In some cases, the late issuance of regulations caused them to go back to their legislature to request additional legislative changes. State officials also want more input into regulations development. STATES WANT MORE TECHNICAL ASSISTANCE -------------------------------------------------------- Chapter 3:3.1 Many state officials we talked with want more technical and training assistance from OCSE and HHS regions, like that done in the past. State officials want more nationwide training from OCSE and help in developing their own training capabilities. They believe the HHS regional offices need to be more proactive in sharing information with the states and helping states provide additional support to one another. SOME STATES AND HHS REGIONAL STAFF TRYING NEW MODEL FOR COOPERATIVE TECHNICAL ASSISTANCE -------------------------------------------------------- Chapter 3:3.2 In the face of diminished OCSE resources for technical assistance, states in two federal regions are developing model approaches to providing technical assistance and training.\45 States in one region have established a steering committee composed of the chief executive officers of states' human service agencies and the ACF regional administrator. The ultimate goal is better coordination of human services. Under the steering committee, four work groups address (1) child support enforcement, (2) welfare reform, (3) child welfare services, and (4) administration and organizational issues. The child support enforcement work group is focusing on common training needs and federal-state communication. In another region, state child support training coordinators are forming a regional training network. The idea originated during a national training conference in 1991. The group plans to share materials, attend each others' training sessions, develop training packages for line workers, and serve as consultants to each other and to their CSE directors. -------------------- \45 The regions are Region VII and X. Region VII is Iowa, Kansas, Missouri, and Nebraska. Region X is Washington, Oregon, Idaho, and Alaska. UNTIMELY REGULATIONS MAKE STATES' IMPLEMENTATION MORE DIFFICULT -------------------------------------------------------- Chapter 3:3.3 Some officials from the states we visited complained that OCSE's late issuance of federal regulations made it difficult for them to pass required state legislation and comply with the law in time and damaged the program's credibility with state legislatures. As shown in table 3.2, for example, final regulations implementing requirements in the 1988 amendments for wage withholding on delinquent non-AFDC cases and all new or modified CSE cases were issued in July 1992, almost 1-1/2 years after the November 1990 statutory effective date. Table 3.2 Statutory Effective Dates and Dates of Final Regulations for Selected CSE Program Requirements Statutory Dates of effective final Program requirement dates regulations ---------------------------- -------------- ------------ Wage withholding on November 1990 July 1992 delinquent non-AFDC cases and all new or modified IV- D cases\a Presumptive state October 1989 May 1991 guidelines\b Medical support in all August 1984 September orders\c 1988 ---------------------------------------------------------- \a Wage withholding is a procedure by which automatic deductions are made from wages or income to pay a debt such as child support. It may be voluntary or involuntary. \b Presumptive guidelines are a standard method for setting child support orders based on the income of the parent(s) and other factors. Child support orders set according to the guidelines are presumed to be correct. Judges and others must use the state guidelines or demonstrate that the application of the guidelines would be unjust or inappropriate in a particular case. \c Medical support includes (1) legal provision for payment of medical and dental bills or (2) access to the noncustodial parent's medical insurance or (3) both. In the absence of federal regulations, states are left to interpret laws on their own. When regulations are issued later, they may have provisions in them that make state law incorrect or otherwise require state CSE program officials to go back to their legislatures for changes. In 1989, at the time of proposed regulations governing use of presumptive state guidelines, OCSE proposed additional requirements only 1 month before the statutory effective date--long after such requirements should have been known. State officials complained that some requirements would cause undue burden on those who had taken steps in good faith to achieve compliance but now would be out of compliance. Even when OCSE has responded to legislation by issuing regulations promptly, state officials often find themselves having to address later revisions. For example, regulations establishing time frames for distribution of support in interstate cases were required by statute to be issued by August 1989 and were, but OCSE made significant modifications--in essence finalized the regulations--almost 2 years later in May 1991. SOME STATES WANT A MORE COLLABORATIVE APPROACH TO DEVELOPING REGULATIONS -------------------------------------------------------- Chapter 3:3.4 Some program officials in the states we visited suggested a new approach to developing regulations in which OCSE would solicit more involvement from states earlier in the process. Some believed that OCSE does not get enough state input before proposing regulations or does not adequately respond to state comments on the impact of implementing new policies and proposed regulations. These officials believed the approach currently used for developing federal regulations provides for state comments too late and does not encourage a federal-state dialogue. They would like to see states involved before proposed regulations are drafted, when OCSE is wrestling with the issues. One program official suggested an alternative approach in which OCSE would identify issue areas and use HHS regional CSE specialists to discuss the issues with the states and collect ideas from the states for OCSE. Program officials in other states believed this approach would allow states to participate in the process well before the regulations are drafted and public comments are solicited. They also believed this approach would save time by reducing state comments later, after the draft regulations are published.\46 -------------------- \46 What the states suggested is essentially similar to "regulatory negotiation" recommended by the Administrative Conference of the U.S. in 1982 and 1985. The conference noted that "bringing interested parties together in a cooperative setting at the front end of the rulemaking process . . . has the capacity to reduce the likelihood of litigation, to produce faster and less costly rulemaking--and to create objectively better rules." David M. Pritzker and Deborah S. Dalton, Negotiated Rulemaking Sourcebook, Office of the Chairman, Administrative Conference of the United States (Washington, D.C.: Jan. 1990). AUDITS PROVIDE LIMITED INSIGHT INTO PROGRAM OUTCOMES, BUT STILL SPUR STATE ACTIONS ---------------------------------------------------------- Chapter 3:4 OCSE compliance audits as presently structured provide only a partial picture of state performance. The audits provide little insight into ultimate case outcomes and the extent to which the actions taken have been productive. The audits focus on compliance with federal requirements for written procedures and service delivery and thus highlight what state programs have not done, rather than what state programs have accomplished and what methods have been particularly successful. In addition, audits are untimely with final formal audit reports sometimes issued a full 2 years after the period of performance audited. Although state officials complained that audits are too late to be a useful management tool for them, they acknowledged that the audits and the threat of penalties have helped their programs get state legislative attention and resources. To ensure compliance with OCSE regulations for program standards in anticipation of audits and to improve program management, some states are developing internal auditing and monitoring capabilities. OCSE is attempting to make the audits more outcome-focused and timely. AUDITS ASSESS COMPLIANCE WITH ADMINISTRATIVE PROCEDURES MORE THAN OUTCOMES -------------------------------------------------------- Chapter 3:4.1 Despite congressional pressure to measure states' program effectiveness, OCSE has continued to rely on an audit approach that has over 50 compliance criteria and three performance indicators to judge program effectiveness. The performance indicators are limited to measuring three outcomes: AFDC and non-AFDC collections-to-cost ratios and AFDC recovery rates; they do not address other CSE functions such as paternity and order establishment. The compliance criteria include 29 criteria with numerous subcriteria for auditing state compliance with federal requirements for certain procedures, such as written procedures for publicizing the availability of support enforcement services. In addition, there are 23 criteria with numerous subcriteria for ensuring that states are providing child support services in accordance with their approved state plans. In the past, OCSE has considered an effective state program one that is in substantial compliance with both aspects of the audit: performance indicators and compliance criteria. To determine substantial compliance with the performance indicators, the level of state performance reached in each of the three performance indicators is assigned a numerical score. A state's combined score must equal or exceed 70 for substantial compliance. To determine substantial compliance with the criteria, OCSE auditors examine state written procedures to determine whether they (1) comply with federal laws and regulations and (2) are used in 75 percent of the cases sampled. In addition, each sampled case may be audited for more than one criterion. For example, a case may be reviewed for its compliance with both location and paternity establishment criteria. Furthermore, a case may fail one criterion but pass another, and thus be counted as deficient in one instance and acceptable in another--regardless of the case's ultimate outcome. OCSE believes that this audit approach is performance-based. In addition, OCSE believes the audits provide clear indicators of state program performance because they highlight what services were not provided by states during the audited period. However, determining compliance--the extent to which procedures are in place and actions are taken on cases--does not include an assessment of the effectiveness of the states' procedures in achieving the desired outcomes such as paternity and support order establishment. In addition, while many states have failed the compliance criteria, only 3 states have also failed to achieve a score of at least 70 on the performance indicators since fiscal year 1986. Therefore, audit results present a limited view of state performance. AUDITS TOO LATE TO BE USEFUL -------------------------------------------------------- Chapter 3:4.2 The usefulness of compliance audits for judging state performance is also limited because they are not done in a timely fashion. In some instances, final formal audit reports were issued 2 years after the period of performance audited. In addition, OCSE has yet to audit state implementation of the Family Support Act of 1988 because the audit regulations are not yet final. State officials, HHS regional staff, and advocates shared the view that the audits' usefulness is limited. State officials expressed confusion over the timing and purpose of the audits and believed that penalties were often applicable to past conditions that no longer exist. In addition, some HHS regional CSE specialists told us the audit reports are too late to be useful to the states. STATES SAY AUDITS PROVIDE SOME IMPETUS FOR CHANGE -------------------------------------------------------- Chapter 3:4.3 Although state officials complained about the utility and timing of compliance audits, some acknowledged positive effects from audits. Program officials acknowledged that audits and penalties or the threat of penalties\47 have helped their programs get state legislative attention and resources. With the attention of their legislatures, states were able to make a number of legislative, procedural, and system changes to correct deficiencies and hire additional staff. In addition, some states said the audits were useful in getting offices to use the administrative process. -------------------- \47 Approximately $21 million in audit penalties have been assessed Ohio, New Mexico, Arizona, Mississippi, the District of Columbia, Guam, the Virgin Islands, and Puerto Rico as a result of OCSE compliance audits since 1975. SOME STATES HAVE DEVELOPED THEIR OWN PROGRAM MONITORING STRATEGIES -------------------------------------------------------- Chapter 3:4.4 Some states have developed their own internal audit procedures and monitoring systems to ensure compliance with federal standards and provide more timely performance evaluation for program management. Although OCSE is not yet auditing state programs for requirements of the 1988 amendments, some states are auditing themselves against these requirements. OCSE considers a number of state efforts exemplary and highlighted seven states' internal monitoring programs in an information memorandum to all states in early 1993. In addition, our review of eight states identified some additional efforts. In California, for example, county performance is reviewed annually, feedback from state evaluators is described as immediate, and follow-up on corrective action plans for noncompliance is quarterly. California formed a separate unit that does the reviews in 21 small counties. That unit also validates reviews done in 37 larger counties by local staff who are trained by the state in the review methodology and funded by the state. Several counties have incorporated software in their case management systems for tracking time frames, which has helped the state prepare for statewide automation by increasing the uniformity of case management throughout the state. An analysis of the review efforts showed marked improvement in county compliance with federal program operation standards. In another state, Delaware, the quality control system that has been developed evaluates not only the performance of the CSE agency but also that of the Delaware Family Court. An automated sampling process ensures that cases selected had activity in the specified program area being examined during the specified time period. OCSE IS TRYING TO CHANGE ITS AUDIT PROCESS -------------------------------------------------------- Chapter 3:4.5 Over the last few years, OCSE has been trying to change the audit process, and its officials are on record that the audits need to be more outcome-oriented. Working with state CSE program directors and the National Governors' Association, OCSE has been exploring ways to make the audits more outcome-oriented and has taken both a regulatory and legislative approach to effect a change. In September 1993, OCSE published proposed revisions to existing audit regulations that officials did not believe required any statutory changes. In the proposed approach, achievement of desired case outcomes in sampled cases would take priority over compliance with process and time frames in determining substantial compliance. If a desired outcome, such as paternity establishment, was achieved in a case during the time period audited, the case would be considered compliant; deviations from the process or time frames leading to that outcome would not be counted as noncompliant. If a desired outcome was not achieved, then the case would have to meet process and time frame criteria. The proposed regulations, however, would not add new performance indicators or revise the current indicators that are limited to cost-effectiveness and AFDC recovery. OCSE has chosen to delay revisions to the performance indicators until more refined indicators can be devised and states have had sufficient time to implement the new standards. While the proposed audit regulations would reduce some requirements, they would add others, making their net effects on audit timeliness difficult to predict. Although the proposed audit regulations combine and consolidate current criteria, they would add many significant criteria to allow the auditing of compliance with provisions of the 1988 amendments. The proposed regulations narrow the focus of audits to 14 of 52 existing criteria. These 14 criteria reflect service areas that at least 10 percent of the states have failed in the past. However, all of at least 17 additional criteria (with numerous subcriteria) would be audited for compliance with the 1988 amendments. Moreover, the substantial compliance standards for maintenance of records and case closure would be raised from 75 percent to 90 percent of cases sampled. While the revised audit regulations have not been finalized, OCSE officials have decided that further changes need to be made to the audit process to focus more on outcomes and less on detailed process requirements. To do this, however, OCSE officials believe that the statutory mandate for comprehensive audits must be changed. Proposed legislative language effecting this change was incorporated in the administration's welfare reform proposal. Pending further legislative action, OCSE has suspended its scheduled compliance audits for 1994. Instead, audit staff are evaluating state data collection and reporting systems as noted earlier in chapter 2. PROPOSED WELFARE REFORMS WOULD PLACE GREATER DEMANDS ON OCSE ---------------------------------------------------------- Chapter 3:5 While the federal CSE role varied among the eight welfare reform proposals we reviewed, all would give some new responsibilities to OCSE,\48 but few would add resources for technical assistance or change the audit process. Under some of the proposals, OCSE would be required to develop and coordinate new or expanded national automated information systems. Others would require OCSE to work with the states in such areas as training development and establishing staffing standards. New responsibilities placed directly on the states may also require greater technical assistance efforts and communication by OCSE and within HHS to ensure effective state implementation. -------------------- \48 The proposals generally assigned the responsibilites to the Secretary of HHS, but presumably implementation responsibility would be delegated to OCSE. SOME WELFARE REFORMS PROPOSE LARGER COORDINATING ROLE AND MORE RESPONSIBILITIES FOR OCSE -------------------------------------------------------- Chapter 3:5.1 A number of the proposals we reviewed would require OCSE to develop and coordinate new or expanded automated information systems, develop a comprehensive training program for federal and state program staff, and conduct staffing studies of each state's program. Four of the proposals, for example, would direct OCSE to develop a national database of all child support orders issued or enforced by state CSE programs. Six proposals would require reporting of newly hired employees to state or federal agencies, and one of these would also require OCSE to maintain a database of the newly hired employees through a modified W-4 tax withholding reporting procedure. Also, additional responsibilities for OCSE for training and staffing standards for state programs were included in some of the welfare reform proposals. Two proposals specifically called for OCSE to provide training as well as technical assistance to states. One of these proposals would require OCSE to develop (1) a core curriculum and training standards for state programs to follow and (2) national training for state CSE program directors. In addition, these two proposals would direct OCSE to conduct staffing studies in each state, and a third would set a deadline for HHS to set staffing requirements as already required under section 452(a)(2) of the Social Security Act. NEW STATE RESPONSIBILITIES COULD REQUIRE MORE TECHNICAL ASSISTANCE AND EFFECTIVE COMMUNICATION FROM OCSE -------------------------------------------------------- Chapter 3:5.2 Under many of the welfare reform proposals we reviewed, states would also have new mandates and requirements that may require more technical assistance and effective communication from OCSE for implementation. For example, seven proposals would impose new requirements affecting states' paternity establishment programs. One called for OCSE to develop regulations for voluntary paternity acknowledgment in nonhospital settings. Another would give state CSE programs authority to determine if an AFDC applicant is cooperating in the paternity establishment effort. However, OCSE is responsible for developing the standards state programs must use if they believe the custodial parent should be exempt from cooperation. In addition, there are proposed mandates for states to establish such measures as employee reporting, professional license suspension if child support is delinquent, less judicial case processing, central collections and distribution, cost-of-living adjustments in new or modified child support orders, and inclusion of delinquent child support payers in their job search and training programs for welfare recipients. While some states are implementing some of these now, others may need assistance from OCSE to enact and implement these requirements. In addition, under many of the proposals, states would be required to enhance their current systems and establish new databases. These state efforts would require technical assistance and monitoring activities by OCSE program staff and ACF management information systems personnel to ensure efficient development and capacity for linking state systems to a national system where necessary. Seven of the proposals we reviewed, for example, would require states to establish central or automatically linked local registries that would have a record of all child support orders being enforced by the CSE program. Some of these proposals would require all child support orders issued in the state to be included. SPECIFIC FUNDING SCHEME PROPOSED FOR TECHNICAL ASSISTANCE -------------------------------------------------------- Chapter 3:5.3 One proposal included a provision for a specific funding scheme for federal activities assisting state programs. Under this proposal, funds from the federal portion of recovered AFDC payments would be used to fund a wide range of activities. Among these activities would be information dissemination, technical assistance, training of state and federal staff, and operation of the federal databases containing child support orders and employee information. AUDITS TARGETED FOR CHANGE UNDER SOME WELFARE REFORM PROPOSALS -------------------------------------------------------- Chapter 3:5.4 OCSE would be required under three of the welfare reform proposals to change the audit process to different degrees. One proposal specified the process be streamlined to (1) be less burdensome on states in substantial compliance, (2) produce more timely analysis of states not in substantial compliance, and (3) possibly shift resources to providing technical assistance. Another proposal required OCSE to create, after a year of study, a new process--with new criteria, standards, and reporting--that emphasizes outcomes. Finally, a third proposal would make the most specific and immediate changes. This proposal would require (1) all states to have an annual monitoring and reporting capability that meets OCSE standards and (2) OCSE to change the focus of its audits to verifying state data, data systems, and financial management. Since this proposal would also change the incentive structure to reward states for achieving certain levels of performance in key areas, such as paternity and order establishment, data verification performed by the audits would be critical to the accuracy of performance assessment and incentive adjustments. This proposal incorporates the statutory change desired by OCSE. STATE CSE PROGRAMS USE MULTIPLE STRATEGIES TO ADDRESS COMMON BARRIERS TO STATE CSE EFFORTS ============================================================ Chapter 4 Despite substantial federal funding, mandates, and oversight, CSE is still very much a state program in which state and local leadership, operating in unique political and fiscal environments, is primarily responsible for program performance. Our work in eight states\49 and our past studies--and the research of others-- suggest that state programs face common barriers hindering their CSE efforts and are at different points in developing the specific management and enforcement tools they need to overcome or reduce the barriers. State programs must deal with increased caseloads; limits in their computer systems' capacity to serve those caseloads; limited control over local program activities; a need to garner legislative support for new state CSE initiatives; and poor communication between CSE and AFDC staff that impedes CSE's ability to provide services to AFDC clients. The states we visited had developed multiple strategies to deal with these barriers. Over the years, all had added some staff, increased their use of administrative rather than judicial case processing, and developed innovative enforcement tools such as requiring employers to report newly hired employees to a central state registry and using state tax data to locate noncustodial parents. In addition, these states all recognized automation as a critical tool for dealing with a variety of barriers and were expanding their computer capacity. State programs used a variety of techniques, such as having a legislative liaison office, to convince state legislatures to enact laws needed to authorize new enforcement tools. States also have made changes to improve CSE-AFDC coordination, such as locating both programs in the same state agency. Table 4.1 provides an overview of the various barriers identified in our review and the strategies state CSE programs initiated to overcome them. As shown in the table, multiple strategies are often used to address different barriers. Appendix III presents selected information about the state CSE programs we visited. Table 4.1 Barriers States Face and Some Strategies to Overcome Them Barriers Strategies ------------------ -------------------------------------- Increasing More staff workloads and resource Administrative rather than judicial constraints processes for establishment and enforcement of paternity and support orders Innovative practices such as new-hire reporting to reach difficult populations Increased use of automation Contracts with private collection firms Volunteers Limited computer Enhancing existing and acquiring new system capacity computer software and technology Dispersed program Automation used to control and lack of procedural --centralize program data and uniformity functions --standardize case processing procedures --monitor compliance with processing standards Lack of Legislative activism and consensus- legislative building through support for initiatives and --assigned staff to promote program resources mission to legislature --public awareness campaigns and lobbying efforts emphasizing goals and achievements --communication with advocacy groups and others --improved program management and services Inadequate Communication initiatives and program communication liaisons between AFDC and CSE programs Mutual education and training programs for staff Staff co-location Revisions to intake and referral process ---------------------------------------------------------- -------------------- \49 Arizona, Iowa, Kentucky, Massachusetts, New York, Oregon, Texas, and Virginia. EXPERIENCES OF EIGHT STATES ILLUSTRATE COMMON BARRIERS IMPEDING STATE CSE EFFORTS ---------------------------------------------------------- Chapter 4:1 To varying degrees, the eight state CSE programs we reviewed have experienced and continue to experience common barriers. These barriers include (1) increasingly complex and growing workloads coupled with resource constraints, (2) limited computer system capacity, (3) dispersed program control and lack of uniform procedures, (4) lack of legislative support, and (5) inadequate communication between the AFDC and CSE programs. STATE AND LOCAL PROGRAMS MUST DEAL WITH MORE DEMANDING WORKLOADS AND RESOURCE CONSTRAINTS -------------------------------------------------------- Chapter 4:1.1 State and local CSE programs must deal with more demanding workloads despite constrained resources. State programs are confronting (1) a growing number of cases that more frequently need paternity establishment, (2) increasing time-consuming interaction with custodial parents, (3) limited ability to reach some noncustodial parents, and (4) federal requirements for medical support enforcement and case review and modification. At the same time, state budgets have been seriously cut back and some states are seeking ways to "downsize" their governments. As a result of rising out-of-wedlock birth rates, paternity establishment is becoming an increasingly large aspect of CSE. In Texas, for example, as of June 1993, more than 40 percent of Texas' total caseload needed paternity established, according to state figures (see figure 4.1). As described in chapter 1, paternity establishment is the first step in obtaining child support. It can be time-consuming and involve genetic testing and judicial hearings as well as CSE efforts to interview custodial parents and locate the alleged fathers. Figure 4.1: Texas CSE Caseload Profile, June 1993 (See figure in printed edition.) In addition, AFDC custodial parents' reluctance to cooperate with CSE staff has been cited by program staff across the country as contributing to the workload difficulty. Staff in the eight states we visited also cited contacts from increasing numbers of non-AFDC custodial parents as part of their growing workload. Program officials in all eight states also expressed frustration with the difficulty of pursuing noncustodial parents who are self-employed or change jobs frequently. CSE officials explained that the inability to reach such parents with existing resources and enforcement tools impeded collections and overall program performance. The expanding CSE program mission also increased CSE case complexity. In addition to the traditional establishment and enforcement services, program resources are being devoted to medical support enforcement and review and modification of support orders. The additional case complexity and the resources required are reflected in the longer time it takes to train new caseworkers and bring them up to a full-performance level. In Kentucky, officials said it takes 15 months to bring a new caseworker up to the full-performance level; whereas, it used to take 9 months. At the time of our visits, states were beginning to wrestle with review and modification requirements that became effective October 13, 1993. Based on the results of state demonstrations of the review and modification process, some state program officials anticipated further resource shortfalls. The demonstrations revealed that the process is very labor intensive, with two states estimating the need for 11 to 13 percent more staff to meet the requirements. One researcher noted that review and modification "can require considerable persistence to obtain accurate income information from parents, perform one or more guideline calculations, try to obtain an order agreeable to both parties, and prepare the case for court if necessary. In addition, part of the workload burden derives from the specifications of the Family Support Act, which imposes a complex scheme of notices and timeframes."\50 Although staffing levels have increased over time for all the states we reviewed, program personnel told us that the programs were still understaffed relative to caseload size and the amount of work required for a typical child support case. Program officials' estimates of worker caseload ranged from 300 to 2,500 cases per worker.\51 In commenting on workload difficulties, a New York local CSE official estimated that his office was 40 percent understaffed. In addition, a 1992 Arizona Auditor General report noted that staffing levels were inadequate to address rapidly growing caseloads. Telephone inquiries alone can consume staff time. State officials in Iowa and Texas, for example, reported that centrally located operators handle an average of 7,000 and 31,000 calls a month, respectively. These calls are in addition to the calls made to those states' automated inquiry systems and do not include the number of calls handled directly by field staff. Although CSE programs have received substantial federal funding and state funding increases over the years, they have not always escaped budget cuts. The Iowa program, for example, experienced a reduced legislative appropriation in 1991 along with all other state agencies. A number of CSE officials in the states we visited expressed concern that state funding for the program remains insufficient and resources are thinly spread. -------------------- \50 Robert G. Williams, Ph.D., President, Policy Studies, Inc., in testimony before the Subcommittee on Human Resources, Committee on Ways and Means, U.S. House of Representatives (Mar. 15, 1994). \51 See Interstate Child Support: Wage Withholding Not Fulfilling Expectations (GAO/HRD-92-65BR, Feb. 25, 1992). In a survey of 136 local CSE offices, the median estimate was 1,000 cases per worker. LIMITED COMPUTER SYSTEMS CAPACITY A BARRIER TO STATES -------------------------------------------------------- Chapter 4:1.2 Despite nearly 15 years of 90-percent federal funding and a requirement that CSE have statewide computer systems by October 1995, some state CSE programs lack the systems capacity to serve growing caseloads and address additional federal case processing requirements. In extreme cases, such as Virginia and Kentucky in 1993, states still do not have single statewide computer systems. Instead, they rely on multiple stand-alone systems or manual processing to manage and track caseload actions. In Kentucky, for example, the largest portion of the state's caseload was being tracked by both a largely manual system in the state area office and a local computer system in a corresponding county attorney's office. Kentucky officials noted this arrangement caused duplication of casework between state and county offices. In both Virginia and Kentucky, multiple systems and the absence of integrated databases make case processing and tracking more difficult and require more manual intervention. Oregon's CSE program was among the first to implement automated CSE systems but now suffers from aging technology and strained capability to carry out federally required CSE functions such as case initiation and enforcement.\52 Oregon, although it implemented a system between 1975 and 1978, is now in virtually the same situation as some other states developing systems for the first time. Over the years, Oregon made incremental enhancements to its system to handle increased caseload complexity and federal case processing requirements. Unfortunately, these incremental enhancements have strained the system's functional and data storage capacities. A 1991 OCSE review noted major deficiencies and recommended Oregon replace its current system. Automated systems in Massachusetts and Iowa also continue to have limitations. Massachusetts, a leader in using automation to seize noncustodial parents' bank assets, still has a limited system for case management. The current system provides little support for case initiation, management, and automated tracking of efforts to locate parents, paternity establishment, and court proceedings. Caseload intake and noncustodial parent location functions are primarily handled manually by CSE personnel. In contrast, Iowa's automated CSE system has extensive case tracking and reporting capabilities, but its ability to seize bank assets is hindered by its current automation capacity. As of June 1993, Iowa lacked the technology to identify assets and levy administrative liens automatically. One program official said that the ability to do a tape-to-tape match of CSE records with self-employment tax and asset records would be an "invaluable" enforcement tool. -------------------- \52 Other functional requirements under federal law include case management, financial management, security, and reporting. PROGRAM CONTROL IS DISPERSED AND UNIFORMITY IS LACKING -------------------------------------------------------- Chapter 4:1.3 Personnel in most of the state programs we visited were frustrated with the dispersed and variable nature of their program structure, but not all to the same degree. Program personnel in New York, Kentucky, and Oregon complained that multiple and autonomous particpants (for example, judges, county attorneys, local governments) with competing priorities prohibited centralized control and posed a barrier to state CSE program performance. These programs lacked uniform statewide procedures and as a result had difficulty implementing statewide CSE initiatives. Our previous work and the research of others have identified dispersed program control as a barrier to improving CSE.\53 According to one study, involvement by the courts, prosecutors, advocacy groups, local human services agencies, and the CSE agency itself creates a complex interplay of competing traditions and values likely to have a limiting effect on the local implementation of any reform effort.\54 In a state where county governments exercise significant control of program operations, the New York CSE program has experienced difficulties providing program direction and implementing statewide practices. Clients obtain CSE services through 58 Department of Social Services district offices. In the districts, local commissioners control program operations and have considerable autonomy in determining local management philosophies, organizational structure, and case processing procedures and priorities that often conflict with the state CSE program's initiatives. State reviews of district offices from 1982 to 1992 also documented that local districts have failed to implement state procedures and staffing recommendations and fully utilize the automated information system, among other issues. Kentucky has experienced similar difficulties managing the program as a result of fragmented authority and court and county attorney involvement. In certain places, judges use their own individualized procedures rather than following uniform state procedures. As a result, the state procedures manual is replete with different procedures required in different counties. Some judges, for example, will not honor support orders from other counties within the state when the custodial parent lives in their jurisdiction; they establish new orders and will not collect arrears under orders from other counties within the state. Even states with greater centralized program control, such as Massachusetts, Virginia, Iowa, and Texas, have had some difficulty implementing uniform procedures despite having direct control over caseworkers and legal staff in local offices. While lines of authority are more clearly defined and initiatives are more easily coordinated from the program's headquarters, control and uniformity issues do arise primarily in relation to the courts. -------------------- \53 GAO/HRD-87-37, Apr. 30, 1987. \54 Charles F. Adams, Jr., David Landsbergen, and Larry Cobler, "Welfare Reform and Paternity Establishment: A Social Experiment," Journal of Policy Analysis and Management, Vol. 11, No. 4, (1992), p. 667. LACK OF LEGISLATIVE SUPPORT AFFECTS STATE PROGRAMS' ABILITY TO IMPLEMENT INITIATIVES -------------------------------------------------------- Chapter 4:1.4 Despite successful efforts in some states, many CSE programs have experienced legislative opposition to proposed initiatives over the last decade. Some state programs we visited tried but were unable to obtain legislative support to implement new processes and tools. In at least two cases, however, legislative support was withheld because of historically poor program management. Examples from the eight state programs we visited illustrate the importance of legislative support to adopting practices considered innovative by program officials. In 1993, the Massachusetts legislature failed to support the CSE program's proposal for certain administrative enforcement techniques opposed by the courts and interest groups. The proposal called for implementing in-hospital paternity establishment, administrative order establishment, and administrative review and modification of orders. While in-hospital paternity establishment passed, other elements did not. The CSE director explained that the legislature was not convinced that administrative processes would be more efficient than judicial processes. In addition, the legislature was concerned that due process rights would not be protected if the courts were not involved in all child support matters. CSE officials in other states we visited told us legislatures had been reluctant to approve additional enforcement tools. In New York, past proposals for new-hire reporting and driver and professional license revocation authority failed because of the legislature's concern that the measures would hurt business. Iowa's CSE program has also been unable to obtain legislative support for a central lien index that would allow the program to quickly identify property held by noncustodial parents and place liens on it. Real estate professionals strongly opposed the measure. INADEQUATE COMMUNICATION BETWEEN AFDC AND CSE RESULTS IN BACKLOGS AND DELAYS -------------------------------------------------------- Chapter 4:1.5 Inadequate communication has resulted in incomplete AFDC referrals to CSE programs and contributed to caseload backlogs and processing delays. Communication between AFDC and CSE is important because CSE relies on AFDC staff to conduct initial intake interviews and obtain noncustodial parent information. Over the years, ineffective communication between the two programs has been cited by OCSE and other studies as a barrier to performance. In addition, in our 1987 report we found that poor coordination between welfare and CSE agencies resulted in inadequate information about noncustodial parents, including identity, location, and earnings data.\55 The experiences of the state programs we reviewed suggest that very little has changed. CSE personnel in the states we visited cited problems with the interaction between AFDC and CSE. Referrals from the AFDC program frequently lacked information CSE staff needed to do their work, either because the AFDC staff did not collect the information or the parent did not provide it. AFDC personnel appeared to be reluctant to penalize AFDC recipients who did not cooperate in providing needed information. As a result of incomplete referrals, CSE workers spent time and resources tracking down additional information, reinterviewing clients, and providing services to clients unwilling to identify the absent parent. -------------------- \55 GAO/HRD-87-37, Apr. 30, 1987. SOME STATES HAVE DEVELOPED SUCCESSFUL STRATEGIES TO ADDRESS PROGRAM BARRIERS ---------------------------------------------------------- Chapter 4:2 The eight state CSE programs we visited have implemented or are in the process of implementing many strategies to reduce or eliminate program barriers. The design, implementation, and success of some of the common strategies reflect the political and fiscal environment and CSE program history in each of the states. Multiple strategies are directed at growing workload size and complexity. These strategies include hiring more staff and increasing the use of (1) administrative rather than judicial processes, (2) innovative enforcement techniques, (3) automation, and (4) contracts with the private sector. In addition, one state is making extensive use of volunteers to extend staff resources. Some of these strategies, such as increased automation and private contracting, are also helping to improve program control and uniformity within state programs. Because automation is recognized as a critical tool for dealing with a variety of problems, all the states we visited are enhancing their automated capabilities. In addition, some state programs are actively working with their state legislatures and others, such as judges and county officials, to build consensus for initiatives and legislative support. Some state programs have also found ways to improve communication with the AFDC program. Although opportunities exist for all states to implement these approaches, we found that some state programs demonstrated more initiative than others in such areas as legislative activism. However, we also saw instances where legislative activism was not completely successful. In addition, most of the barriers have not been completely eliminated and will continue to require extensive state effort. MULTIPLE STRATEGIES HELP STATES ADDRESS WORKLOADS AND SUPPLEMENT LIMITED RESOURCES -------------------------------------------------------- Chapter 4:2.1 No single strategy is used to meet the demands of a growing workload. Despite limited resources, all CSE programs we reviewed have added some staff over the years. They have also increased their reliance on administrative processes, innovative enforcement tools, and automation to address growing and more complex caseloads. In addition, some of the programs have engaged the private sector to extend staff resources through contracts with collection agencies and volunteer recruitment. SOME STATE PROGRAMS HAVE HIRED MORE STAFF ------------------------------------------------------ Chapter 4:2.1.1 Most state programs we visited have been moderately successful in adding staff over the years, although some have experienced staff cuts and hiring freezes as well. In Texas, for example, as its CSE caseload tripled between 1984 and 1992, program personnel more than tripled. Massachusetts' CSE program recently raised its staffing level approximately 25 percent. The Kentucky program also experienced two large increases of 30 percent and more in 1979 and 1989, but more recently, staffing was cut in 1992 and 1993 by 6 and 7 percent, respectively. ADMINISTRATIVE PROCESSES ALLOW GREATER EFFICIENCY ------------------------------------------------------ Chapter 4:2.1.2 In most of the states we visited that have implemented administrative alternatives to court processing for such actions as paternity establishment, order establishment, and review and modification, CSE personnel believe the administrative process has added or will add to their programs' efficiency. However, the type and extent of administrative processes used in these states vary. State program personnel in some of the states we visited believed that administrative processes have reduced court involvement, created greater procedural uniformity, and allowed CSE programs to serve increasing workloads more efficiently. Iowa, for example, has had extensive administrative procedures for years and continues to add to them. The Iowa CSE program expanded its administrative means for paternity establishment because program officials believed it would make the process faster and cheaper by reducing the amount of attorney and court time. In addition, Iowa is implementing an administrative process for the review and modification of support awards to more efficiently meet federal time frames for review and modification of support orders. In contrast, the Texas program has traditionally relied on the judicial process for order establishment and enforcement. In the past few years, however, as court backlogs have grown, it has added administrative alternatives. Texas is now implementing an administrative process for order establishment that relies on a mediation conference with the parents instead of a court hearing to set the award amount. In promoting this approach to the Texas state legislature, program personnel estimated that more than 50 percent of the caseload could be resolved using the administrative process. INNOVATIVE ENFORCEMENT TOOLS AND APPROACHES EXTEND PROGRAM'S REACH ------------------------------------------------------ Chapter 4:2.1.3 The eight states we visited are using innovative enforcement tools to extend their CSE programs' reach in locating noncustodial parents or their assets or to create greater incentives to pay support. These tools include requiring employer reporting of newly hired employees, using state tax data to locate noncustodial parents, and requiring noncustodial parents who do not pay child support to perform community service or revoking their professionl, trade, or driver's licenses. Like several other states around the country, Massachusetts, Iowa, Oregon, and Virginia have implemented mandatory employer reporting of newly hired employees for a number of industries and professions. The Texas legislature also recently approved new-hire reporting. Texas employers, however, are not required to participate but are encouraged to volunteer. To reach self-employed noncustodial parents and nonwage income, Iowa's legislature authorized its CSE program and Department of Revenue and Finance in 1993 to implement a tax record matching procedure to identify self-employed individuals with child support obligations. Self-employed noncustodial parents with delinquent child support obligations are sent a notice on Department of Revenue and Finance letterhead requiring the payment of child support and the quarterly submission of a state estimated tax form. For individuals who do not comply, CSE tries to identify interest- bearing assets from the tax return and attempt garnishment. Iowa's CSE program has also obtained legislative authority to order noncustodial parents who claim they are unemployed to seek employment. Failure to seek employment can serve as evidence to the court in determining a willful failure to pay. Iowa has also given the courts the power to impose community service as an alternative sentence for contempt of court or failure to pay child support. Other practices that program officials believe will enhance collections include revocation or suspension of professional licenses. Arizona and Oregon are among at least 18 states that have obtained authority to revoke or block renewal of professional licenses. AUTOMATION INCREASES EFFICIENCY ------------------------------------------------------ Chapter 4:2.1.4 CSE officials in the states we visited commonly noted that a greater reliance on automated enforcement and parent-locating tools has allowed their programs to do more work with fewer staff and resources. In general, all eight states have expanded their use of automation for case processing and asset attachment and are making a considerable number of collections through automated wage withholding and interception of other income such as unemployment benefits and tax refunds. In New York, for example, automated enforcement techniques produce an estimated 75 percent of total collections. Since 1992, Massachusetts has begun to automate a number of innovative enforcement tools. Accordingly, the CSE program implemented a "Mass Enforcement" initiative that focuses on cases with the greatest potential for collections and uses computerized file matches to identify employment and asset information. Increased automation has helped to free up line staff from administrative duties and allow them to do more casework. For example, staff from some states said that innovations such as computerized voice-activated response systems using "800" numbers have allowed them to spend more time working cases and less time responding to simple telephone inquiries about payments. Iowa program officials said they were able to reduce their central payment inquiry staff from 14 to 8 as the result of a voice-activated response system that informs callers when a payment has been received by the state. CONTRACTING WITH THE PRIVATE SECTOR EXTENDS LIMITED STAFF RESOURCES ------------------------------------------------------ Chapter 4:2.1.5 Contracting with private sector companies extends limited state program staff in a variety of ways, although not all the states we reviewed are using contractors. New York, for example, is contracting with a vendor for centralized collections and payment processing. Previously, payments were made and processed at 58 district offices throughout the state. The state estimates that a privately operated centralized collections process will save the program between $5 million and $6 million a year. Increased workloads and resource considerations have also caused Massachusetts, Texas, and Virginia to contract for collection services. Massachusetts contracts with two private companies to make collections on a small number of difficult cases. Texas started using a private contractor in 1993 to supplement staff resources. The contractor is expected to pursue collections on cases that are at least 6 months delinquent and have arrears of $150 or more. Virginia also relies on a contractor to make collections, but only in AFDC cases in arrears, which represent about 5 percent of the program's total caseload. Virginia recently created two additional district offices, expanding from 19 to 21, by using a private firm to run all operations of the new offices. This contract has allowed the Virginia program to expand services geographically while staying within its state-mandated staffing levels. VOLUNTEERS AND WELFARE TRAINEES ENHANCE STAFF RESOURCES ------------------------------------------------------ Chapter 4:2.1.6 One state--Texas--supplements staff resources with volunteers, including student interns and trainees referred by the AFDC program. These unpaid staff answer telephones, file, and perform other office duties. In Texas' fiscal year 1993, volunteer staff put in 130,000 hours in field operations and on special projects. The volunteers' field office hours represented approximately one full-time equivalent in each of the 58 field offices. The volunteer program director estimated that in addition to the office hours, field office volunteers made 225 community presentations about CSE, including the responsibilities of parenthood and paternity, that reached approximately 175,000 people. HHS' Regional Office VI is very supportive of the volunteer program and has approved federal matching funds for computer equipment used by volunteers. STATES ARE IMPROVING AUTOMATION CAPACITY -------------------------------------------------------- Chapter 4:2.2 All state CSE programs are planning and developing statewide systems to meet federal certification requirements by October 1995. For some states, like Oregon and Texas, this has meant abandoning old, patched-together systems in favor of entirely new systems. As a result of a 1991 OCSE review that noted deficiencies in Oregon's current system, state officials decided to identify, transfer, adapt, and implement an automated system that was already implemented in another state. In addition, some states, like Arizona, are implementing statewide computer systems for the first time. In December 1992, Arizona implemented a new computer system that has increased capacity. According to Arizona officials, more than 300,000 cases are now in a centralized database. The new system has allowed the program to implement automated tax intercepts, credit bureau reporting, and lottery intercepts. The system also has an improved location capacity because of additional connections with other computer data systems. Arizona CSE personnel said the computer system helped them monitor the stages of each case better and had improved automated interfaces. AUTOMATION AIDS CENTRALIZING PROGRAM CONTROL AND UNIFORMITY -------------------------------------------------------- Chapter 4:2.3 Automation has helped some states centralize program control, ensure uniformity in procedures, and centralize a number of CSE functions. In general, the strategy has been to automate more processes and transfer control from local offices to the central CSE administration in the state. At the time of our review, New York was beginning to use automation and a private contractor to transfer collection processing from the 58 district offices to the state office. With centralized collections, payments from noncustodial parents are received at one central location, posted by the system, and distributed automatically to the custodial parents. Before this initiative, the collection process was primarily manual and controlled by local Department of Social Services offices throughout the state. New York has also used automation to centralize many other functions, including lottery intercepts, tax intercepts, administrative penalties for nonpayment, and document generation. Iowa provides another example of standardization through automation. The Iowa CSE director said automation has given the program a single process for handling cases and that staff no longer have a choice of how to proceed with actions such as order establishment. In addition, the program's case tracking system allows the CSE program to centrally monitor and document case actions in terms of federal case processing time frames. Reports showing case actions that are not in compliance with the time frames are shared with staff at all levels. LEGISLATIVE ACTIVISM USED TO GARNER SUPPORT FOR PROGRAMS -------------------------------------------------------- Chapter 4:2.4 The CSE programs we visited have used a variety of techniques to increase legislative support. These techniques include legislative liaisons, public awareness campaigns, lobbying efforts, consensus building among stakeholders, and other initiatives to improve internal management and case processes. In many instances, such efforts have resulted in additional program resources, authority to implement innovative practices, and greater program control. These efforts, however, have not always been successful. Some state programs continue to have difficulties obtaining necessary legislative support. In Texas during the 1993 legislative session, senior program officials, in addition to the program's legislative liaison, lobbied the state legislature and made compromises in wording and procedures to institute an administrative process. This lobbying effort included an orientation session for all legislators and staff at the start of the session. During the orientation, CSE staff explained the CSE program and what kind of assistance the program can give legislators responding to constituent inquiries. CSE convinced the legislature that the judicial process took longer than necessary and kept the program from achieving its potential. The Texas CSE program also uses other measures to reach legislators, judges, and county officials to build program credibility and a consensus for action. The program is required by law to report back to the legislature on program outputs and outcomes. This requirement has helped keep the legislature informed about the effectiveness of CSE initiatives. In addition, units of the CSE litigation division work with judges on standardizing court procedures and processes to improve CSE and establishing case law with carefully selected cases. The CSE director and staff also work with county law enforcement and judicial officials to identify program improvements. The Massachusetts CSE program has taken several actions to obtain legislative support for its initiatives, and program officials believe this support has allowed the program to become one of the most aggressive in the country. The program has a legislative liaison who meets regularly with each member of the legislature and a cadre of key supporters to discuss upcoming legislation. In addition, the liaison generates statistical packages highlighting program impacts for the legislature. To improve program credibility, the Massachusetts CSE program also sponsors public service announcements and holds press conferences to publicize collections figures. The program also publishes a "10 Most Wanted List" of nonpaying noncustodial parents and other pamphlets. As a result of its efforts to gain the legislature's support, the Massachusetts CSE program has had more successes than failures. Successful legislative initiatives have included (1) a statute that requires all employers to report new hires within 14 days of starting employment; (2) an administrative lien provision that creates a child support lien when a child support order is due and unpaid; (3) voluntary paternity acknowledgment; (4) authority to enforce court-ordered health insurance provisions by directly contacting the employer or insurance providers; and (5) access to employment, utility, and licensing records to locate noncustodial parents. However, the CSE program has been unable to get its entire package of legislative initiatives, which include administrative processes for new and modified orders, passed in the legislature. ACTIONS TAKEN TO IMPROVE COMMUNICATION BETWEEN CSE AND AFDC -------------------------------------------------------- Chapter 4:2.5 Although problems remain, the state CSE programs we visited have implemented initiatives to improve communication with AFDC. Some approaches are aimed at improving the referral process, while others have focused on improving cooperation through procedural revisions, education programs for welfare and child support personnel, and other more innovative techniques. Oregon appears to have been more successful than the other states in confronting and reducing this particular barrier to performance. Oregon CSE officials voiced a positive view about the program's relationship with the AFDC program and staff and attributed this view to several factors. The CSE director believed that good communication was due, in part, to both CSE and AFDC having the same oversight agency and a single computer system. In addition, past problems with referral data had been corrected through education and refinement of AFDC intake documentation. The CSE program developed a training program for AFDC staff on how to enter information necessary and useful to CSE efforts directly onto the computer system. The CSE and AFDC programs also refined the AFDC intake document to obtain child support information. These efforts reduced the amount of information AFDC staff were required to get from clients for CSE purposes and improved cooperation between the programs. Some of the other states we visited are looking at ways to improve the quality of data from the AFDC intake process and to better communication at senior management levels in each program. For example, Arizona recently conducted a study of the intake interview and referral process and was planning to establish a pilot office in which CSE personnel would be co-located with AFDC staff. CSE personnel will conduct intake interviews to ensure that the necessary information is obtained and will educate their AFDC counterparts about appropriate information gathering approaches. The Texas CSE program is examining the intake process in coordination with AFDC staff and is focusing on mutual education, training programs, and discussions with AFDC management. AFDC and CSE management staff meet quarterly to discuss common issues and to ensure cooperation. Program officials believe cooperation has improved as a result of these regular meetings. In addition, liaisons have been established for both agencies to assist personnel throughout the state with case processing problems. WELFARE REFORM COULD INCREASE STATES' WORKLOAD BUT COULD HELP SOME MAKE CHANGES ---------------------------------------------------------- Chapter 4:3 All eight welfare reform proposals we reviewed contained new mandates for state CSE programs. In some instances, these mandates would add to the workload states are already struggling to address as well as add more automation requirements. However, the mandates would give some state programs the enforcement tools, such as new-hire reporting and driver's license suspension, that they have been unable to obtain from state legislatures. In addition, these mandates would help some state programs achieve greater centralization and control of their operations. WELFARE REFORM COULD INTENSIFY SOME BARRIERS STATES ARE FACING -------------------------------------------------------- Chapter 4:3.1 All the welfare reform proposals we reviewed would expand state program responsibilities to different degrees and potentially intensify some of the barriers states are now facing. For example, under at least one proposal, the CSE caseload would be expanded to include all child support orders, even those not now enforced under the Social Security Act. While a portion of these new cases may never need enforcement services, they would still be subject to the requirement that all cases be automatically reviewed and considered for modification every 3 years. In addition, three of the proposals require state programs to make greater paternity establishment efforts through outreach and voluntary paternity acknowledgment. Two proposals would require state parent locator services to be used to handle requests from noncustodial parents to locate custodial parents for enforcement of visitation issues. Welfare reform proposals may further strain state automated information systems. As indicated in chapter 3, most of the proposals would require state programs to have automated central child support order registries or the capability to integrate local registries. In addition, one of the proposals would require state programs to match employer reports of new employees with the state order registry. Another would require state programs to have automated on-line access to information in any database maintained by the state. WELFARE REFORM PROPOSALS PROVIDE NEW ENFORCEMENT TOOLS FOR SOME STATE PROGRAMS AND REQUIRE GREATER CENTRALIZATION -------------------------------------------------------- Chapter 4:3.2 The welfare reform proposals we reviewed would provide new enforcement tools to some states and require greater centralization of operations for some as well. For example, seven proposals would require employers to report new hires either to a state or federal unit that then makes the information available to all states. In addition, two proposals require states to set up central collection and disbursement agencies to handle all child support payments throughout the state. Furthermore, one proposal offers additional federal funding as an incentive to centralize all of CSE under one state-controlled organization. Also, two proposals provide for driver's license suspensions. CONCLUSIONS AND RECOMMENDATIONS ============================================================ Chapter 5 The history of the CSE program is one of progressively expanded responsibilities. Despite annual federal and state funding of nearly $2 billion, the CSE program is struggling to keep up with the needs of an ever-expanding constituency of welfare, working poor, and middle income families. Given nationwide trends toward increasing numbers of out-of-wedlock births, federal and state governments are likely to continue to face many challenges in managing their programs. In addition, welfare reform proposals we reviewed suggest that OCSE and the states may be tasked with greater responsibilities in the near future. To date, OCSE has been less than effective in the leadership role the Congress envisioned when it established the national CSE program. As program mission and caseload have expanded over time, OCSE has continued to lack some important management tools, including a clearly articulated program mission, programwide planning and goal-setting, and a means to measure progress toward goals. At present, OCSE does not have accurate and consistent data on state program performance. OCSE needs such data if it is to assess program performance and take action to improve it. OCSE will also need these data if it is to successfully implement the management reforms required under GPRA. OCSE's ability to lead effectively would also be helped by a funding structure that is better aligned with the current realities of CSE's mission and that could help promote service improvement at the state and local levels. In addition, OCSE's working relationships with state programs in recent years have been characterized as more adversarial than cooperative. Miscommunications between OCSE, HHS regional staff, and state program staff over CSE policy interpretation and other matters have fueled states' dissatisfaction. HHS' current organizational structure, which divides CSE responsibilities between OCSE staff and HHS regional staff not accountable to OCSE management, bears some responsibility for these ongoing difficulties in the federal-state partnership that is at the core of the CSE program. OCSE's audits of state programs are perhaps its most important management tool, and some state program staff said that OCSE audits had spurred state action. But the audits have focused on monitoring state compliance with federally prescribed administrative procedures. As a result, the audits give OCSE and state program directors limited insight into the ultimate outcomes of program activities, such as whether paternities were established, noncustodial parents located, and collections achieved. OCSE is aware of the audits' limitations and has taken some actions to change the audits. Our work confirms and reinforces a need to reengineer OCSE audits. Such a reengineering could not only help federal-state program relations but also serve as a means for OCSE and state program staff to obtain accurate and consistent performance data--key management data that could help OCSE and states as they seek to better serve the families that depend on CSE efforts. In this chapter, we make some specific recommendations. We believe implementing GPRA's management reforms may help OCSE improve its own program management as well as serve states better. But whether through GPRA implementation or through other means, some rethinking of federal management tools, processes, and relations with state CSE programs is needed if the issues we have identified are to be addressed. We recognize that the CSE program faces many difficult tasks and that even given optimum CSE program management, some noncustodial parents may never be located and others may never be able to pay support. But continued and strengthened efforts by OCSE and states would seem virtually certain to help some families. Such efforts are all the more important if OCSE is to be given new authorities and responsibilities under welfare reform. GPRA CAN HELP OCSE DEVELOP NEEDED MANAGEMENT TOOLS ---------------------------------------------------------- Chapter 5:1 As a result of the GPRA mandate and OCSE's pilot project status, OCSE is beginning to build a planning and measurement discipline that focuses on outcomes. Such a process should enable OCSE to better (1) communicate program expectations, (2) identify the tasks needed to accomplish the outcomes, (3) identify the data needed to assess accomplishments, and (4) use the information from performance assessment for policy decisions and plan revisions. OCSE's actions to date show promise, but we believe some of the issues we have identified remain to be addressed. As of September 1994, OCSE had developed a draft strategic plan that exceeds what GPRA requires and had avoided repeating some past planning weaknesses. To develop the plan, OCSE has sought comments and involvement from key stakeholders such as the regional HHS CSE personnel and state program representatives. In addition, OCSE had defined a program mission, vision, and two general goals with objectives and many quantitative performance indicators for each goal. The draft mission statement affirms the broad scope and expanded mission the program has assumed over the years. The goals focus on paternity and order establishment in support of this broad mission, and the objectives and performance indicators specifically address outcomes not only for collections but for noncollection activities such as medical support enforcement as well. Articulating the broadness of the mission and establishing outcomes should help to clarify the program's mission among the stakeholders. In addition, through a welfare reform proposal, OCSE is attempting to align the existing funding structure and audit approach with the broad mission OCSE has defined. Such an alignment is important for stakeholders to clearly understand program priorities. In addition to the strategic plan, OCSE has also developed an annual performance plan as required by GPRA. According to GPRA, the annual plan is intended to provide a direct link between the program's longer-term goals and what state programs and OCSE's managers and staff will need to accomplish each year. Setting demanding but realistic long-term and annual measurable outcomes for the national program and for state programs will be challenging. The variation among states in adopting innovative practices, such as voluntary paternity acknowledgment and early employer reporting of newly hired employees, suggests a need for OCSE to work with each state program individually to set goals that require continuous improvement from year to year. One way to do this would be for OCSE and states to set goals that reflect and build on past state performance, similar to the approach used for the paternity establishment standards. We believe that an individualized, cooperative, goal-setting effort would encourage greater state responsibility and accountability for achieving the goals and the data used to measure performance. A cooperative process would seem more likely to promote such accountability than having OCSE or the Congress imposing annual goals on state programs. OCSE's planning efforts, however, will need to address some of the issues we have raised in this report, including data quality and OCSE's own outcome goals. Improving data quality is essential if OCSE and the states are to achieve effective annual performance reporting and OCSE has taken some preliminary steps toward this end. However, OCSE needs to define data definitions and set standards to promote consistent data reporting. Building on initial efforts to clarify definitions and obtain consistent state reporting, OCSE officials said OCSE audit staff are currently investigating (1) data definitions used by states, (2) state data collection systems, and (3) state data quality. However, these officials did not have a plan or timetable for the completion of this effort. Furthermore, once measurable outcome goals have been established for the program, OCSE and the state programs will need to identify the key data needed for annual performance reporting and develop the necessary definitions. Data quality, however, relies heavily on state program efforts. States will thus need to assume greater accountability for the quality of data they submit. Another important issue for OCSE to consider is its own performance as both a partner with and manager of state program efforts. While OCSE has expressed an intention to assess its processes for serving the states, it has yet to set goals for itself regarding the future timeliness of regulations, responsiveness to states about audits and policy questions, and regional HHS office commitment and feedback. Given the failure of OCSE's past planning efforts to focus on outcomes, measurement, and continuous improvement, we believe OCSE will need to address these issues in future planning. FUNDING STRUCTURE SHOULD BE REDESIGNED ---------------------------------------------------------- Chapter 5:2 Our work identified several problems with the current program funding structure that, taken together, suggest that the funding structure should be redesigned. The current incentive payment structure has not been aligned with the expanding CSE mission. Furthermore, it does not meaningfully encourage continuous improvement, nor does it provide for the cost sharing between state and federal governments originally envisioned for the program. Instead, the incentive payments make it possible for states to shift most, if not all, program costs to the federal government. If incentive payments are to promote better state performance, we believe they need to be aligned with performance expectations as defined in the mission priorities and goals established in the strategic plan. For example, if such a plan recognizes a broader CSE mission and includes measurable outcome goals for collection and noncollection results, such as paternity and order establishment and medical support enforcement, then the incentive payment structure should be used to reward state improvements and maintenance of the highest level of performance in these areas. We endorse the concept of performance-based funding embodied in the incentive payments and believe this concept should be retained in the redesigned structure. THE FEDERAL-STATE PARTNERSHIP NEEDS TO BE REINVENTED ---------------------------------------------------------- Chapter 5:3 OCSE should continue to work with states to find new ways to strengthen the federal-state partnership that is at the core of the CSE program. New ways to deliver technical assistance and training, as well as new ways to develop regulations, would be important steps to this end. In addition, long-standing communication problems between state officials, HHS regional staff, and OCSE staff, problems due in part to the current organizational structure, will need to be addressed. We encourage OCSE to explore more cooperative approaches to developing regulations and delivering technical assistance and training. In chapter 3, we described some initiatives states would like to see or are trying for developing regulations and delivering technical assistance. These may be potential models for OCSE to consider or develop further in working with the states. As part of reinvigorating the state-federal partnership, we believe HHS and OCSE officials should consider alternatives to the current federal organizational structure for the CSE program, since this structure has produced ineffective communication and accountability. To this end, OCSE's strategic planning can help clarify OCSE's role by establishing performance goals for the agency. As with the CSE program, these goals should be focused, as much as possible, on results such as more timely issuance of regulations and responsiveness to states. The needed organizational changes, therefore, should support not only CSE program mission priorities, but OCSE's agency performance goals as well. In developing a better functioning organization, we believe HHS should consider (1) what the role of regional HHS personnel is, (2) how regional resources should be spent on CSE, (3) what mechanisms are necessary to ensure accountability to OCSE by regional and ACF staff outside of OCSE who are working on CSE priorities, and (4) where federal resources should be devoted to foster state goal achievement. THE AUDIT IS A CANDIDATE FOR REENGINEERING ---------------------------------------------------------- Chapter 5:4 Whatever changes may be made to the CSE program, we believe it is essential that the federal government maintain a program monitoring function to ensure that children who need financial support are being effectively served. However, in view of the problems we and others have identified with the program's audits, we believe, and OCSE agrees, that the scope and approach of this monitoring function need to be reconsidered. While the audits have spurred state action in some cases, we believe it is time for the program audit role to be reexamined and reengineered to provide a monitoring capability that is not only more outcome-oriented, but supports accurate state performance reporting as well. Two existing conditions make this an opportune time to change the audit approach. First, GPRA implementation demands accurate, consistent data and assessment of progress toward stated goals--elements the current program lacks. GPRA annual performance reporting, therefore, should provide insights about program outcomes that the current audit approach has been unable to provide. To ensure these insights are based on sound data, however, the data, not the administrative procedures, should be subject to audit. The OCSE audit role, when placed within the GPRA framework, can be shifted from ensuring compliance to ensuring data integrity and investigating performance problems identified in annual performance reporting, thus achieving a more outcome-oriented approach. A second condition that enables serious reconsideration of the existing audit approach is the growth of state-initiated self-monitoring. As we noted in chapter 3, a number of states are conducting their own program audits or evaluations using OCSE-promulgated program standards or their own evaluation criteria. These state program monitoring efforts can have a more timely effect on state operations than OCSE audits and reduce the need for OCSE to conduct the type of detailed audits of state administrative procedures it has in the past. OCSE is currently considering ways to use state monitoring efforts, including replacing the existing federal audit approach with required annual state monitoring and OCSE review of the results. OCSE agrees that the audit needs to be reengineered but believes its ability to make significant changes to the audit process is constrained under current law. Existing legislation requires OCSE to conduct periodic comprehensive audits of state programs to ensure substantial compliance with all federal requirements. WELFARE REFORM PRESENTS OPPORTUNITIES AND ADDITIONAL CHALLENGES FOR CSE ---------------------------------------------------------- Chapter 5:5 Welfare reform proposals could have various effects on the CSE program. Adding federal mandates, as reflected in the welfare reform proposals we reviewed, may result in a more standard national CSE program. In addition, welfare reform holds some promise for a redesigned funding structure and audit approach. However, under welfare reform, OCSE would likely need to assume more responsibilities despite its limited capacity and organizational accountability. Any reform is likely to require greater federal leadership than has been exhibited in the past to achieve reform goals. While increasing requirements would challenge OCSE's leadership capacity, some of the welfare reform proposals provide opportunities for better alignment among stated priorities and incentive funding. In addition, some target the audit process for change. By focusing financial rewards on performance in critical areas in addition to cost-effectiveness, the proposed funding structures emphasize the importance of activities such as paternity establishment that may not bring immediate collections. Furthermore, by focusing the audits more on program outcomes and the quality of the data used to measure the outcomes rather than on compliance with certain requirements, welfare reform provides the opportunity to improve information for future decision-making. Many of the current proposals for welfare reform would add to OCSE's responsibilities for providing technical assistance and training and for monitoring state programs. As demonstrated by our findings in this report and our earlier testimony,\56 OCSE is already struggling with existing requirements in these critical support areas. Absent the reengineering of some OCSE processes, such as the audit, and an organizational structure with clear lines of authority and OCSE accountability, welfare reform seems likely to present OCSE with serious difficulties in achieving desired results. In addition, new requirements in the various welfare reform proposals would force legislative action in some states where the CSE programs have been unable to obtain the authority to pursue certain measures. However, the new requirements would also expand the authority and responsibilities of state CSE programs, requiring more stringent paternity establishment procedures, central registries, centralized collections, and work requirements for noncustodial parents, among others. Some of the new requirements could also add to the confusion about CSE's mission, goals, and priorities described in chapter 2. Given all these likely impacts of welfare reform on OCSE, it is all the more important that changes be made to address the current lack of alignment among stated priorities, expectations, performance measurement, and incentive funding. -------------------- \56 Child Support Enforcement: Federal Efforts Have Not Kept Pace With Expanding Program (GAO/T-HEHS-94-209, July 20, 1994). RECOMMENDATIONS TO HHS ---------------------------------------------------------- Chapter 5:6 We recommend that the Secretary of HHS take three actions. First, direct OCSE to address the following issues in its future planning efforts, in conjunction with major program stakeholders such as the Congress, state program managers, regional CSE specialists, and appropriate advocacy groups: individualized state performance goals based on such factors as past state performance, demographics, and degree of automation; data quality improvement; funding-structure changes that will retain the current performance-based approach but make it supportive of mission priorities; and methods for improving federal-state communication and working relationships. Second, the Secretary should direct OCSE to establish performance goals that are linked to the fulfillment of OCSE's mission. In this process, we believe OCSE should consider setting outcome goals, in addition to output goals, in the areas of timely regulatory development, clear policy direction and responsiveness to state programs, cooperative planning, and technical assistance. Third, the Secretary should establish an organizational structure and reporting mechanisms for OCSE that support program priorities, set OCSE performance goals, and provide greater federal CSE program accountability among the regional staff and activities. We recognize that during the 103rd Congress the Secretary sought statutory changes to reform OCSE's audit approach and to change the program funding structure within the context of welfare reform. Because we believe changes to the audit approach and funding structure are essential for the program, we recommend that the Secretary continue to pursue the statutory initiatives to these ends, as appropriate. AGENCY COMMENTS ---------------------------------------------------------- Chapter 5:7 HHS provided written comments on a draft of this report (see app. V). HHS stated that our report was comprehensive and presented a balanced appraisal of the CSE program's accomplishments over the past decade. It commented, however, that the funding levels for OCSE training and technical assistance activities in the last decade should be depicted in the context of increasing stringency in the federal budget for nearly all discretionary spending. In addition, HHS stated that our recommendations were well taken. For most of the recommendations, HHS provided additional information about actions it has already taken or is planning that we believe are responsive to our recommendations. However, we remain concerned that HHS' comments do not (1) indicate that OCSE is establishing performance goals for itself, (2) address the organizational accountability weaknesses we have identified, and (3) specify a plan for obtaining statutory changes for the funding structure and audit approach. With respect to our first recommendation regarding issues important to OCSE's planning efforts, HHS commented that OCSE has taken or is planning several actions that we consider responsive. It is planning to develop individual annual performance plans with state CSE agencies that will contain specific performance goals taking into consideration factors we mentioned and others. In terms of improving data quality, OCSE has recently started refining standard definitions for data elements and plans to meet with regional and state staff to achieve consensus on common terminology. In addition, OCSE is seeking to improve federal-state communication and working relationships through state visits by OCSE's deputy director and through developing various forums for more interaction with the states. In response to our second recommendation regarding the establishment of OCSE performance outcome goals that are linked to its mission, OCSE intends to develop an internal performance plan. While HHS provided additional information about the steps OCSE is taking to enhance training and establish staffing standards, it did not indicate that the performance plan will contain goals, either outcome or output, for OCSE. We believe that such outcome goals are essential for OCSE to (1) clarify its role and relationship to state programs, (2) focus on results, and (3) better hold regional resources accountable. We have clarified our recommendation to emphasize that OCSE should establish what outcomes it wants to accomplish with its actions. With respect to our third recommendation on organizational structure and reporting mechanisms, HHS described a recent reorganization, but its comments do not indicate how this reorganization will ensure regional office accountability for federal CSE activities. HHS states that ACF has reorganized so that the 10 regional administrators report directly to a new ACF Director of Regional Operations rather than the Assistant Secretary for ACF. In addition, an Assistant Director for Child Support Regional Operations has been established to assist in the direction and coordination of child support activities by the regions. HHS comments do not indicate, however, what reporting mechanisms it will use to obtain an up-front commitment of regional staff and travel resources to CSE and accountability for results of CSE activities. We continue to believe that such commitment is necessary to ensure that OCSE knows what regional resources are devoted to CSE and how regional actions will support OCSE's performance goals and foster state goal achievement. Finally, HHS' comments do not specify plans for changing the audit to make it more outcome-oriented or for reintroducing legislation to change the program funding structure. We continue to believe that the Secretary should pursue statutory initiatives as appropriate to bring about needed changes. We have clarified the language of our recommendation to this effect. DESCRIPTIONS OF BASIC CSE SERVICES =========================================================== Appendix I Client intake involves opening a case record and compiling data on the custodial family and noncustodial parent. The custodial parent is interviewed and referral (from AFDC or other states' agencies) and application forms are screened to determine if the information provided is complete and accurate. Outside contacts may be necessary to secure all needed information. Noncustodial parent locator services include efforts at local, state, and federal levels to identify a noncustodial parent's address, social security number, place of employment, or the like. This might include efforts to directly contact individuals; contacts with public and private institutions, such as credit bureaus, state and federal income tax agencies; and computer tape matches. Paternity establishment is the identification of the legal father of a child, usually through the courts or expedited through hearings in a quasi-judicial or administrative body. Paternities are established in either of two ways: (1) through voluntary acknowledgment by the father or (2) if contested, through a determination based on scientific and testimonial evidence. Initiation of the paternity process varies depending on whether the mother is an AFDC recipient. Mothers who receive AFDC assistance are required by law to cooperate in locating and identifying the father of the child for whom aid is requested. Mothers who do not receive AFDC, however, are under no legal obligation to establish paternity and may voluntarily seek CSE services. Support order establishment and review and modification involves the development of a support order that legally obliges the noncustodial parent to pay child support and provide medical insurance coverage when available at reasonable cost. It also involves the periodic review and adjustment of the order at least every 3 years in AFDC cases and upon parental request in non-AFDC cases. The CSE agency must assist custodial parents in initiating an action in court or through an administrative or expedited legal process that will produce such an order. The CSE agency helps in the determination of a child's financial needs and the extent to which the noncustodial parent can provide financial support and medical insurance coverage. Enforcement refers to a wide array of techniques at the disposal of CSE agencies to enforce payment on delinquent cases or to ensure regularity and completeness of current accounts. These techniques include bonds and security deposits, federal and state tax intercepts, garnishments, liens, and wage withholding, among others. Collections processing refers to the processing, recording, and distributing of child support collections from noncustodial parents. On AFDC cases, if the monthly collection is insufficient to disqualify the family from AFDC, the family receives its full monthly AFDC grant plus the first $50 of the support payment. The remainder of the support payment is distributed to reimburse the state and federal governments in proportion to their assistance to the family. On non-AFDC cases, and in AFDC cases where the family's income including the support payment is sufficient to make it ineligible for AFDC, support payments are paid directly to the family. MAJOR REQUIREMENTS OF THE 1984 AND 1988 CSE AMENDMENTS ========================================================== Appendix II CHILD SUPPORT ENFORCEMENT AMENDMENTS OF 1984 -------------------------------------------------------- Appendix II:1 In the event of arrearages, automatic wage withholding for IV-D cases and based on application in non-IV-D cases\57 "Expedited processes" to establish and enforce support orders--defined as faster, more efficient and under which presiding officers are not judges Withholding of state tax refunds if parents are delinquent in support payments Imposition of liens against real and personal property for amounts of overdue support Permit establishment of paternity up to a child's 18th birthday Security bonds required of noncustodial parents to secure overdue support payments Consumer credit bureaus receive upon request data on overdue support of any noncustodial parent Nonbinding state guidelines for child support available to judges and other officials Non-AFDC families charged service fees Annual notices of support collections to AFDC recipients Noncustodial parents may be charged late payment fees Federal incentives to local offices Continuation of CSE services after AFDC termination CSE services for children in foster care cases Enforcement of spousal support\58 CSE services publicized Equal provision of CSE services to AFDC and non-AFDC families Inclusion of provision for medical support in orders Withholding federal tax refunds of non-AFDC families Extension of Medicaid eligibility when support collections result in termination of AFDC -------------------- \57 IV-D cases are those enforced under Title IV-D of the Social Security Act. Non-IV-D cases are not enforced under Title IV-D but may apply for services other than enforcement. \58 Enforcement of legal obligations for support of a spouse or former spouse who is living with a child or children to whom an individual owes child support. FAMILY SUPPORT ACT OF 1988 -------------------------------------------------------- Appendix II:2 Immediate wage withholding on orders: new and modified IV-D; delinquent non-IV-D (November 1990); and all new non-IV-D (January 1994) Recording of parents' social security numbers at children's births Paternity establishment performance standards State guidelines for child support as a rebuttable presumption Periodic review of state child support guidelines Time standards for response to requests for services States give information to the Federal Parent Locator Service Civil and voluntary paternity acknowlegment procedures encouraged Extend the paternity statute of limitations up to a child's 18th birthday Genetic testing in contested cases Periodic review and modification of support orders Monthly notices of collections Statewide automated systems Time standards for distribution of collections SELECTED CHARACTERISTICS OF EIGHT STATE CSE PROGRAMS, 1992-1993 ========================================================= Appendix III Massachusett Characteristic Arizona Iowa Kentucky s New York Oregon Texas Virginia ------------------ ------------ ---------- ------------ ------------ ------------------ ----------------- ---------------- ------------------ Caseload 195,189 126,321 242,768 210,594 1,007,058 195,347 696,761 294,148 AFDC 100,197 60,351 141,372 140,748 510,494 105,087 353,056 150,026 Non-AFDC 94,992 65,970 101,396 69,846 496,564 90,260 343,705 144,122 Administrative expenditures (in millions) ----------------------------------------------------------------------------------------------------------------------------------------------------- Federal share $20.9 $11.1 $21.8 $29.4 $101.1 $14.1 $66.5 $34.1 State share 8.7 5.5 9.9 14.9 50.5 7.0 32.6 15.9 Total $29.6 $16.6 $31.6 $44.3 $151.6 $21.1 $99.1 $50.0 Incentive payments $ 1.5 $ 6.6 $ 4.4 $10.2 $ 19.0 $ 4.2 $ 8.9 $ 4.7 to states (estimated) Full-time 900 346 831 831 2,979 410 2,057 935 equivalent staff\a Administration ----------------------------------------------------------------------------------------------------------------------------------------------------- State administered X X X X X X X County \b \c X \d administered Operational organization ----------------------------------------------------------------------------------------------------------------------------------------------------- Centralized X X AFDC X X Decentralized X X X non-AFDC \e Placement of CSE social social social social services social services attorney general social services office in state services services services revenue government Predominant case process ----------------------------------------------------------------------------------------------------------------------------------------------------- Judicial X X X non-AFDC X Administrative X\f AFDC X Mixed or quasi- X X judicial HHS region number 9 7 4 1 2 10 6 3 Selected enforcement tools ----------------------------------------------------------------------------------------------------------------------------------------------------- New-hire reporting no yes\g no yes no yes\h voluntary\i yes\j License revocation yes\k no no yes\l no yes\m no no ----------------------------------------------------------------------------------------------------------------------------------------------------- Notes: Caseload, expenditure, and staffing data reported by the states for fiscal year 1992; other information in this table as of June 1993 unless otherwise indicated. Some numbers may not add to totals shown because of rounding. \a Total number of full-time equivalent staff employed by the state agency and any local agencies as well as the number of full-time equivalent staff employed by an agency (public or private) working under cooperative agreement or a purchase-of-service agreement with the CSE agency. \b Five of 15 programs are operated by local county attorneys under contract with the state. \c In all counties, many services, such as paternity and order establishment and enforcement on cases needing paternity establishment, are provided through county attorneys under contract. \d Under contracts with the state CSE office, AFDC cases are handled by branch offices of the state department of justice and non-AFDC cases are handled by county district attorneys. \e Payment processing is not centralized. \f Iowa adopted administrative paternity establishment effective July 1, 1993. \g Effective January 1, 1994. \h Legislation passed November 1993 for a pilot project limited to eight employment classifications through June 1995. \i Effective September 1, 1993. \j Effective July 1, 1993. \k Trade, professional, sport, and business licenses. \l Trade, professional, and driver's licenses as of August 1994. \m Trade and business licenses. SELECTED CSE PROGRAM REQUIREMENTS IN EIGHT WELFARE REFORM PROPOSALS ========================================================== Appendix IV Provision H.R. 4318 H.R. 4414 H.R. 4126 H.R. 4566 S. 2224 S. 1795 H.R. 3500 H.R. 4476 ---------------------------------- ---------- ---------- -------------- ------------- ------------- ------------- ------------- ------------- Program requirement ----------------------------------------------------------------------------------------------------------------------------------------------------- Revised audits X X X Strengthened paternity X X X X X X X establishment provisions National support order registry X X X X X National directory of newly hired X X employees Nationwide reporting of newly X X X X X X X hired employees and employee support obligations Expanded role of Federal Parent X X X X X X X Locator Service Training programs X X State program databases X X X X X X X Expanded service population X X X X Revised collection distribution X X X priority Revised funding structure X X X State staffing studies or X X X standards New state authority/mandates ----------------------------------------------------------------------------------------------------------------------------------------------------- Driver's license restrictions X X Periodic review and modification X X of orders Professional license suspension X X Support order registry X X X X X X X Centralized collection and X X distribution Administrative process X X Work by noncustodial parents X X X ----------------------------------------------------------------------------------------------------------------------------------------------------- (See figure in printed edition.)Appendix V COMMENTS FROM THE DEPARTMENT OF HEALTH AND HUMAN SERVICES ========================================================== Appendix IV (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) GAO CONTACTS AND STAFF ACKNOWLEDGMENTS ========================================================== Appendix VI GAO CONTACTS David P. Bixler, Assistant Director, (202) 512-7201 Sarah A. Morrison, Project Director Cathy Pardee, Deputy Project Director, (202) 512-7237 ACKNOWLEDGMENTS In addition to those named above, the following individuals made important contributions to this report: Linda Baker co-authored the report; Daniel Bertoni and Dylan Jones led the data collection in four states and co-authored portions of the draft; Barbara Bordelon provided guidance about GPRA implementation; Leslie Albin edited the report; Kevin Kumanga provided guidance from past GAO work; Craig Winslow provided legal assistance; and Dan Brier, Alice Spargo, Gerard Grant, Harry Johnson, and George Scott provided RELATED GAO PRODUCTS ============================================================ Chapter 1 Child Support Enforcement: Federal Efforts Have Not Kept Pace With Expanding Program (GAO/T-HEHS-94-209, July 20, 1994). Child Support Enforcement: Credit Bureau Reporting Shows Promise (GAO/HEHS-94-175, June 3, 1994). Child Support Enforcement: States Proceed With Immediate Wage Withholding; More HHS Action Needed (GAO/HRD-93-99, June 15, 1993). Child Support Assurance: Effect of Applying State Guidelines to Determine Fathers' Payments (GAO/HRD-93-26, Jan. 23, 1993). Child Support Enforcement: Timely Actions Needed to Correct System Development Problems (GAO/IMTEC-92-46, Aug. 13, 1992). Child Support Enforcement: Opportunity to Defray Burgeoning Federal and State Non-AFDC Costs (GAO/HRD-92-91, June 5, 1992). Medicaid: Ensuring That Noncustodial Parents Provide Health Insurance Can Save Costs (GAO/HRD-92-80, June 17, 1992). Interstate Child Support: Wage Withholding Not Fulfilling Expectations (GAO/HRD-92-65BR, Feb. 25, 1992). Interstate Child Support: Mothers Report Less Support From Out-of-State Fathers (GAO/HRD-92-39-FS, Jan. 9, 1992). Interstate Child Support: Better Information Needed on Absent Parents for Case Pursuit (GAO/HRD-90-41, May 24, 1990). Child Support: State Progress in Developing Automated Enforcement Systems (GAO/HRD-89-10FS, Feb. 10, 1989). Interstate Child Support: Case Data Limitations, Enforcement Problems, Views on Improvements Needed (GAO/HRD-89-25, Jan. 27, 1989). Child Support: Need to Improve Efforts to Identify Fathers and Obtain Support Orders (GAO/HRD-87-37, Apr. 30, 1987).