This is the accessible text file for GAO report number GAO-04-377 entitled 'Child Support Enforcement: Better Data and More Information on Undistributed Collections Are Needed' which was released on April 15, 2004. This text file was formatted by the U.S. General Accounting Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Report to the Chairman, Committee on Finance, U.S. Senate: United States General Accounting Office: GAO: March 2004: Child Support Enforcement: Better Data and More Information on Undistributed Collections Are Needed: GAO-04-377: GAO Highlights: Highlights of GAO-04-377, a report to the Chairman, Committee on Finance, U.S. Senate Why GAO Did This Study: Congress established the child support enforcement program in 1975 to ensure that parents financially supported their children. State agencies administer the program and the Office of Child Support Enforcement (OCSE) in the Department of Health and Human Services oversees it. In 2002, state agencies collected over $20 billion in child support, but $657 million in collections from 2002 and previous years were undistributed—funds that were delayed or never reached families. One method used to collect child support, intercepting federal tax refunds, involves all state agencies, OCSE, and two Department of the Treasury agencies—the Internal Revenue Service (IRS) and the Financial Management Service (FMS). GAO was asked to address (1) how the total amount of undistributed collections changed over the years, (2) the causes of undistributed collections, (3) states’ efforts to reduce these funds, and (4) OCSE’s efforts to assist states. GAO analyzed OCSE data, administered a survey, visited 6 state agencies and interviewed officials. What GAO Found: OCSE reported that the amount of undistributed collections for fiscal year 1999 was $545 million and $657 million for fiscal year 2002; however, these amounts may not be accurate. State agencies had different interpretations of what comprised undistributed collections and data reported by several state agencies were found to be unreliable throughout this time period. OCSE revised the reporting form, but data accuracy concerns remain, in part, because OCSE does not have a process to ensure the accuracy of undistributed collections data. Federal law, some state policies, and inaccurate or missing information were the underlying causes of nearly all types of undistributed collections. State agencies determined how long they held collections from joint tax refunds and if they held collections received before they were due. Federal law allows collections intercepted from joint tax refunds to be held for up to 180 days and in response to GAO’s survey, 34 state agencies reported holding them for 180 days. Missing or inaccurate information, such as invalid addresses, also leads to undistributed collections. Based on state agencies’ survey responses, GAO determined the median value of the undistributed collections from joint tax refunds was about $1.8 million and the median value of four other types of undistributed collections exceeded $350,000. State agencies GAO visited took steps to better understand and reduce undistributed collections. Of the 6 state agencies visited, 5 had analyzed their undistributed collections cases, 4 adopted performance goals, and officials from all 6 state agencies stressed the importance of researching collections that were missing information. In addition, officials stated that using automated processes to receive and distribute collections helped reduce the number of collections with missing or inaccurate information. OCSE has provided some assistance to help state agencies reduce their undistributed collections. However, the Department of the Treasury has not provided OCSE information that would allow state agencies to distribute collections from joint tax refunds to families sooner. Further, OCSE’s efforts to obtain this information have been minimal. What GAO Recommends: GAO recommends that OCSE periodically review undistributed collections data and that OCSE, IRS, and FMS work together to identify a way to share information on collections held from joint tax refunds. OCSE did not explicitly agree or disagree with the recommendations. IRS and FMS agreed with the recommendation. www.gao.gov/cgi-bin/getrpt?GAO-04-377. To view the full product, including the scope and methodology, click on the link above. For more information, contact Cornelia M. Ashby at (202) 512-7215 or ashbyc@gao.gov). [End of section] Contents: Letter: Results in Brief: Background: OCSE Reported Millions in Undistributed Collections, but Data Were Unreliable: Federal Law, Some State Policies, and Inaccurate, or Missing Information Delayed or Prevented Distribution of Certain Collections: State Agencies Took Steps to Reduce Undistributed Collections: OCSE Has Assisted States' Efforts to Reduce Undistributed Collections, but the Department of the Treasury Has Not Provided Information That Would Help States Distribute Collections from Some Joint Tax Refunds Sooner: Conclusion: Recommendations for Executive Action: Agency Comments and Our Evaluation: Appendix I: Scope and Methodology: Appendix II: Survey of State Directors of Child Support Enforcement: Appendix III: Comments from the Department of Health and Human Services: Appendix IV: Comments from the Department of the Treasury: Appendix V: GAO Contacts and Staff Acknowledgments: GAO Contacts: Staff Acknowledgments: Tables: Table 1: Amount of Tax Refund Offset Collections from 1999 to 2002: Table 2: Categories and Types of Collections to Be Reported as Undistributed: Table 3: Overestimates of Undistributed Collections and Year Reported: Table 4: Time Period Collections from Joint Tax Refunds Are Held: Table 5: Amounts for Nine Types of Undistributed Collections Reported by State Agencies as of June 30, 2003: Table 6: Summary of State Agencies' Responses Regarding the Helpfulness of OCSE Efforts: Figures: Figure 1: Types of Collections and Number of State Agencies That Included Them in Undistributed Collections Reported to OCSE: Figure 2: Number of Years State Agencies Reported Holding Collections before Declaring Them Abandoned Property: Figure 3: Portion of Collections Distributed to Parents Electronically: Abbreviations: CSE: Child Support Enforcement: EFT: Electronic Funds Transfer: FMS: Financial Management Service: FPLS: Federal Parent Locator Service: HHS: Department of Health and Human Services: IRS: Internal Revenue Service: OCSE: Office of Child Support Enforcement: PRWORA: Personal Responsibility and Work Opportunity Reconciliation Act: TANF: Temporary Assistance to Needy Families: TOP: Treasury Offset Program: United States General Accounting Office: Washington, DC 20548: March 19, 2004: The Honorable Charles E. Grassley: Chairman: Committee on Finance: United States Senate: Dear Mr. Chairman: In 2002, the Office of Child Support Enforcement (OCSE), in the Department of Health and Human Services, reported that billions of dollars in child support were collected but that payments totaling $657 million were delayed or never reached the families for whom they were intended. These undistributed child support payments are a concern because child support is an important source of income for many families. According to a 2003 report, for 36 percent of poor children living in families headed by single mothers, child support payments comprised almost one-third of the family's income in 2001. The 1996 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA)[Footnote 1] generally requires state child support enforcement agencies to disburse child support collections within 2 business days, if sufficient information identifying the recipient is provided. In addition, portions of child support collections must be distributed to state government programs, such as Temporary Assistance to Needy Families (TANF), to reimburse them for cash assistance provided to families. Although state child support enforcement agencies administer the child support program, the federal government plays a major role.[Footnote 2] OCSE funds two-thirds of the program's administrative costs; establishes policies and guidance; provides technical assistance, such as designing curricula and providing support for staff training; and oversees and monitors state agencies. Additionally, OCSE is responsible for taking the necessary steps to help resolve issues at the federal level that affect the child support program such as processes that prevent child support payments from reaching families in a timely manner. OCSE and state agencies collect child support through various methods, such as intercepting the federal tax refunds of noncustodial parents--parents who do not have primary care, custody, or control of their children--who are delinquent in paying their child support.[Footnote 3] If the noncustodial parent has a new spouse and files a joint tax return, generally, only the portion of the refund due to the noncustodial parent should be intercepted. Federal law allows state agencies to hold collections from certain intercepted federal joint tax refunds for up to 6 months to provide time for the new spouse to ask for his or her share of the refund by filing an "injured spouse" claim with the Department of the Treasury. In an effort to understand the issues associated with undistributed child support collections, you requested that we address the following questions: (1) How has the total amount of undistributed collections changed in the last few years? (2) What are the causes of undistributed collections? (3) What are state agencies doing to reduce undistributed collections? (4) How has OCSE assisted state agencies' efforts to reduce the amount of undistributed collections? We obtained information from several sources that provided some data for all of the objectives. We conducted a mail survey of all 54 IV-D child support enforcement agencies. Forty-eight state agencies responded to the survey. We did not assess the reliability of the data the state agencies reported in response to our survey. However, we reviewed the data for completeness and reasonableness. We conducted site visits to 6 state agencies. We selected California, Florida, Iowa, New York, Texas, and Virginia because they represented diversity in amounts of undistributed collections, and geographical location, and they provided examples of statewide and county administered programs. We interviewed federal and state officials, reviewed related reports and analyzed applicable laws and regulations. In addition, we took specific steps to address each of the objectives. To address how undistributed collections have changed, we analyzed OCSE data for fiscal years 1999 to 2002 reported by state agencies. The fiscal year 1999 data was the earliest year with data comparable to fiscal year 2002; the most current data available at the time of our review. We assessed the reliability of undistributed collections data reported to OCSE and found that the data were inconsistent and unreliable. We also reviewed state and federal reports that discussed issues associated with calculating and reporting undistributed collections. To identify the causes of undistributed collections, steps state agencies have taken to reduce undistributed collections, and assistance OCSE has provided to state agencies, we reviewed relevant research and audit reports; examined OCSE guidance and documents related to funded projects; and interviewed child support advocates and experts. We conducted our work between May 2003 and March 2004 in accordance with generally accepted government auditing standards. See appendix I for more details on our scope and methodology and appendix II for a copy of our survey. Results in Brief: OCSE reported that the amount of undistributed collections for fiscal year 1999 was $545 million and $657 million for fiscal year 2002; however, these amounts may not be accurate because state agencies had different interpretations of what comprised undistributed collections and data reported by several state agencies were found to be unreliable throughout this time period. A little more than half of the 48 state agencies that responded to our survey reported that they included collections they would distribute in the next 2 business days and collections to be distributed to other government programs, while the others did not. In its 2002 preliminary annual report, OCSE acknowledged the limitations of its data due to reporting discrepancies and has revised the reporting form and related instructions to provide state agencies with uniform definitions of undistributed collections. However, even with a new form and uniform definitions, concerns remain about data accuracy within state agencies. For example, 1 state reported in 2003 that it had overestimated its undistributed collections amount by more than $160 million due to accounting errors. Data accuracy concerns remain, in part, because OCSE does not verify or periodically review undistributed collections data as it does other performance and financial data. Federal law and some state policies as well as inaccurate or missing information were underlying causes for nearly all undistributed collections. Federal law allows collections from joint tax refunds to be held for up to 180 days. In response to our survey, 34 state agencies reported that they held these collections for 180 days. Inaccurate or missing case information also contributed to undistributed collections. For example, information needed to determine the correct amount of the collection to be distributed to custodial parents who are current or former TANF recipients was sometimes incorrect or missing, which caused payments to be delayed. Additionally, invalid addresses for custodial parents caused payments to be returned. Our analysis of state agency survey data found that many state agencies reported holding more that $1 million from joint tax refunds and several hundred thousand dollars in other types of collections. To better understand and reduce undistributed collections, state agencies we visited devoted resources to analyzing and resolving their undistributed collections, established performance goals, and encouraged the use of automated payment processes. Nearly all of the state agencies we visited conducted extensive analyses of the specific causes of undistributed collections. For example, 1 state agency commissioned a study with a private firm to determine the causes of undistributed collections based on a review of thousands of cases. Four state agencies adopted specific performance goals and 1 state agency had a goal to maintain an undistributed collections balance of no more than 1 percent of total collections. State agencies we visited highlighted the importance of dedicating staff to resolving cases with missing information and invalid addresses for custodial parents. Additionally, state agencies reported that the increased use of electronic processes for receiving and distributing funds reduced the amount of undistributed collections because fewer collections had missing or inaccurate information. Thirty-seven state agencies reported using direct deposit to distribute child support payments and 9 state agencies reported using debit cards, which work like other automated teller machine cards, credited with the child support payments. Officials from 2 state agencies we visited told us that these processes reduced the cost of operating the program. For example, officials from 1 state agency reported saving over $300,000 per month with the use of direct deposit. OCSE has provided some assistance to help state agencies reduce undistributed collections, but does not have the information needed to help reduce undistributed collections from some joint tax refunds. OCSE has focused attention on reducing undistributed collections, shared best practices and information with state agencies at child support conferences, and funded several projects related to undistributed collections. To address one of the causes of undistributed collections, in March 2003, OCSE issued guidance informing the state child support enforcement program directors that they could use the Federal Parent Locator Service (FPLS)--a computer matching system with access to federal information sources--to help locate custodial parents. However, the Department of the Treasury has not provided OCSE with some information on "injured spouse" claims filed with tax returns and resolved, that would allow state agencies to more quickly distribute collections from joint tax refunds. OCSE's efforts to obtain more information on "injured spouse" claims have been minimal. In this report, we are recommending that the Secretary of Health and Human Services (HHS) direct the Commissioner of the Office of Child Support Enforcement to periodically review undistributed collections data from state agencies to help improve the accuracy of the data; and we are also recommending that the Secretary of Health and Human Services and the Secretary of the Treasury work together to identify a cost-effective approach for obtaining information related to intercepted tax refunds so that these collections can be distributed to families sooner. The Department of Health and Human Services and the Department of the Treasury provided written comments on a draft of this report. In commenting on the draft, the Department of Health and Human Services did not explicitly agree or disagree with our recommendations, but raised concerns about implementing them. The Department of the Treasury agreed with our recommendation. Copies of the written comments from the Department of Health and Human Services are in appendix III and the comments from the Department of the Treasury are in appendix IV. Background: Child support is not legally owed until a child support order is issued that stipulates the amount the legally identified noncustodial parent- -the parent who does not have primary care, custody, or control of the child--is required to pay and when these payments are due. The Child Support Enforcement (CSE) program, established in 1975 under Title IV- D of the Social Security Act,[Footnote 4] established federal standards for state CSE programs to ensure that parents provide support to their children. Services provided through the CSE program include locating absent noncustodial parents, establishing paternity and support orders, and collecting and distributing child support payments. All 50 states, the District of Columbia, Guam, Puerto Rico, and the Virgin Islands operate CSE programs.However, many aspects of implementing the child support program are generally under the purview of the state rather than the federal government. For fiscal year 2002, OCSE reported over 16 million child support cases and collections of more than $20 billion. One method for collecting child support is intercepting federal tax refunds. Congress enacted the Federal Tax Refund Offset program in 1981.[Footnote 5] Initially the Internal Revenue Service (IRS), an agency within the Department of the Treasury, OCSE, and the state agencies operated the program. In 1998 the Financial Management Service (FMS), another agency within the Department of the Treasury, assumed primary responsibility for the program. Now, all state agencies, OCSE, IRS, and FMS play a role in the program. Table 1 shows the amount of collections offset from tax refunds from 1999 to 2002. Table 1: Amount of Tax Refund Offset Collections from 1999 to 2002: Calendar year: 1999; Child support offsets (in billions): $1.3. Calendar year: 2000; Child support offsets (in billions): $1.4. Calendar year: 2001; Child support offsets (in billions): $1.7. Calendar year: 2002; Child support offsets (in billions): $1.5. Source: OCSE. [End of table] To start the offset process, state agencies identify those noncustodial parents that meet the program criteria. For example, for non-TANF cases, the amount of past due support the noncustodial parent owes must be at least $500. The state agencies, or OCSE on behalf of the state agencies, must send a written notice to the noncustodial parent at least 30 days in advance of sending his or her name to OCSE for the offset program. This notice includes, among other things, information about filing "injured spouse claims."[Footnote 6] State agencies send OCSE the names of the eligible noncustodial parents and the amount owed and OCSE transmits this information to FMS. FMS adds this information to its debtor file that includes information on those who owe child support as well as those who owe other federal debts.[Footnote 7] IRS processes the tax returns, and then forwards information to FMS on those individuals who are due refunds. FMS compares this information to its debtor file and, if there is a match, offsets the refund by the amount of child support owed. After the refunds have been offset, FMS notifies the individual, and transfers the offset funds and information to OCSE. OCSE then distributes the funds and information to the appropriate state agency. When a noncustodial spouse has filed a joint tax return, FMS may offset the refund if either person owes child support. FMS, acting on behalf of the Secretary of the Treasury, must notify OCSE, which notifies the state agencies when the withholding is being made from a refund based on a joint return, and provides the names and addresses of each taxpayer. Under these circumstances, the state agency is permitted to delay distribution of the amount withheld until the Secretary of the Treasury notifies the state agency that the spouse has received his or her proper share of the refund. This delay may not exceed 6 months. However, the law allows a spouse 6 years to file an "injured spouse" claim. If the spouse files a timely "injured spouse" claim and is found to be entitled to a portion of the withheld amount, IRS will process the claim and allocate the appropriate amount to each person. If the spouse's claim is filed after the funds have been forwarded to the state agency, and the spouse is found to be entitled to a portion of the withheld amount, FMS sends the spouse his or her portion and the state agencies must reimburse the Treasury. OCSE data shows that from March to August 2003, almost $200 million dollars were collected from offsets of non-TANF joint tax refunds. PRWORA amended portions of the Social Security Act, including some provisions that pertained to child support enforcement. One provision required state agencies to establish a state disbursement unit to centralize collection and disbursement of child support payments in order to receive federal funds.[Footnote 8] In addition to distributing collections from noncustodial parents to custodial parents, state agencies also distribute collections to: other government programs such as TANF[Footnote 9] and Foster Care programs[Footnote 10] as reimbursement for benefits provided to families. For example, as a condition of receiving temporary cash assistance, parents must apply for child support enforcement services and agree to give all or a portion of their child support to the state. In fiscal year 2002, OCSE reported that custodial parents who were receiving public assistance comprised about three million cases, and those who formerly received public assistance comprised about seven million cases. State agencies are required to report collection and distribution information quarterly to OCSE through the Quarterly Report of Collections, Form OCSE-34A. On the form, net undistributed collections equal the total amount of undistributed collections, less those considered undistributable or abandoned property according to state laws. OCSE Reported Millions in Undistributed Collections, but Data Were Unreliable: OCSE has reported millions in undistributed collections from fiscal year 1999 to 2002, but the amounts may not be accurate. State agencies had different interpretations of what comprised undistributed collections and some state agencies reported data that were found to be unreliable throughout this time period. Although OCSE revised the reporting form for fiscal year 2004, which should improve consistency in the types of collections reported as undistributed, data accuracy remains a concern. OCSE does not audit or periodically review undistributed collections data as it does other performance and financial data. Types of Collections Reported as Undistributed Differed: Although OCSE reported that the amount of undistributed collections for fiscal year 1999 was $545 million and $657 million for fiscal year 2002, OCSE also reported and our survey results indicated that state agencies varied in the types of collections they reported as undistributed. In its preliminary annual report for fiscal year 2002, OCSE acknowledged that interpretation of the data on undistributed collections was limited due to the variation in the types of collections included.[Footnote 11] This report indicated that some state agencies included collections to be distributed within 2 business days; collections received before they were due; and collections to be distributed to other government programs, while other state agencies did not. In our survey, we asked state agencies if they included these types of collections as well as collections to be distributed pending legal resolution and collections received from intercepting joint tax refunds. Of the 48 state agencies that responded, nearly all indicated that they included collections received before they were due to the custodial parent, collections pending legal resolution, and collections received from joint tax refunds. Thirty of the state agencies responded that they included collections to be distributed in 2 business days and 28 state agencies included collections to be distributed to other government programs. Figure 1 illustrates the state responses. Figure 1: Types of Collections and Number of State Agencies That Included Them in Undistributed Collections Reported to OCSE: [See PDF for image] [End of figure] OCSE has taken steps to improve data consistency. OCSE revised its 34A form and related instructions in fiscal year 2003 to improve the consistency of undistributed collections data. According to OCSE officials, this change was also made to provide more information about the composition of undistributed collections. Prior to the changes, all collections were reported under one category, and it was not possible to distinguish between collections pending distribution and collections requiring further research in order to distribute them. As of fiscal year 2004, state agencies are required to report total collections under two categories. The first category, "undistributed collections pending distribution," includes collections that state agencies reasonably expect to distribute through typical business processing in the future. For example, collections held because of a legal dispute over the amount of support owed will be distributed as soon as the matter is resolved. The second category, "undistributed collections unresolved," includes collections that require the child support staff to obtain more information before they can be distributed. These include but are not limited to collections (1) lacking information to match them to a case, (2) missing current addresses for a parent, and (3) from checks that were issued but the time frame for which to cash them had expired. According to OCSE officials, another new reporting form has been developed for use beginning in fiscal year 2005 that will require state agencies to report amounts for 10 types of undistributed collections included under the two categories. These officials informed us that the form is subject to review and approval by the Office of Management and Budget. Table 2 summarizes the two categories and the 10 types of collections to be reported as undistributed collections. Table 2: Categories and Types of Collections to Be Reported as Undistributed: Undistributed collections pending distribution: Collections that were received within the past 2 business days following receipt, and pending distribution within federal timelines; Undistributed collections unresolved: Unidentified collections. Undistributed collections pending distribution: Collections being held for up to 6 months that were offset from non-TANF joint tax refunds; Undistributed collections unresolved: Collections pending the location of the custodial or noncustodial parent. Undistributed collections pending distribution: Collections received before they were due; Undistributed collections unresolved: Collections initially disbursed by check that remain uncashed and considered stale-dated and non- negotiable in accordance with state law and procedures. Undistributed collections pending distribution: Collections held pending resolution of legal disputes and any timely appeal (Examples include contested paternity or dispute over the balance of support owed.); Undistributed collections unresolved: Collections with inaccurate or missing information. (Examples include but are not limited to collections received that do not correspond to the amount owed or collections received for accounts that have been closed or have not yet been opened.). Undistributed collections pending distribution: Collections processed but not yet distributed to other state or federal agencies administering programs such as TANF; Undistributed collections unresolved: Other collections remaining undistributed. Source: Form OCSE-34A. [End of table] Some State Agencies Reported Inaccurate Amounts of Undistributed Collections: In addition to differences in the types of undistributed collections that state agencies reported, 4 state agencies reported data accuracy problems in 2002 and 2003 that were found to be overestimates. The cumulative amounts of these undistributed collections ranged from about $5 million to $168 million and, according to state officials, the actual amounts of undistributed collections were lower than reported. The state agencies and the amounts of the overestimates are summarized in table 3. Table 3: Overestimates of Undistributed Collections and Year Reported: State agency: California; Amount of overestimate: $168,000,000[A]; Year reported: 2003. State agency: Michigan; Amount of overestimate: $25,000,000; Year reported: 2002. State agency: Missouri; Amount of overestimate: $5,000,000; Year reported: 2002. State agency: Nebraska; Amount of overestimate: $5,300,000; Year reported: 2003. Source: State agencies and GAO analysis. Note: Not all state agencies could provide information needed to determine the exact time period associated with these overestimates. [A] OCSE officials reported that this amount was recalculated to be about $163 million and noted that this claim as well as similar claims from other state agencies are subject to federal review. [End of table] Most of the errors were accounting mistakes discovered by the state agencies. Officials from the California Department of Child Support Services reviewed the state's accounting of undistributed collections and found that some local officials had included collections in their totals multiple times because they had misinterpreted a policy. They thought that disbursements could only be reported if collections were also reported in the same period and, as a result, when they disbursed funds they again added these collections into their totals. This resulted in overstating amounts for total collections and for undistributed collections. Nebraska's overestimate occurred partly because collections received before they were due for non-TANF cases were immediately distributed, but the officials did not show them as being distributed until the official due date. In Missouri, nearly all of the overestimate was spousal support collected by the state disbursement unit that was erroneously reported as child support collections. For the remainder, in Missouri, when a collection from a joint tax refund had been distributed to a custodial parent and the amount was later adjusted in the same reporting quarter because of an "injured spouse" claim, the adjusted amount was added as a new collection. An audit of Michigan's child support program revealed that it failed to account for the distribution of payments that noncustodial parents paid directly to custodial parents. The reports related to these overestimates also indicated that having multiple jurisdictions involved with recording collections contributed to the errors. Local agencies in California used forms that did not always include the federal data elements used by the state agency to report undistributed collections. Nebraska discovered errors when its state disbursement unit took over the collection and distribution process from clerks at multiple courts. OCSE Did Not Hold State Agencies Accountable for Accurately Reporting Undistributed Collections: While OCSE is required to audit some child support data, it does not have a process to ensure the accuracy of data on undistributed collections. OCSE is required to audit the reliability of the performance indicators used as the basis for paying financial incentives to state agencies.[Footnote 12] Officials told us they are conducting these audits annually. To ensure the reliability of the data, OCSE selects representative sample cases for a detailed audit and reviews supporting documentation to check for errors. OCSE is also required to conduct financial audits to determine whether federal and other funds used to administer the program are being appropriately expended and properly accounted for. These audits are required to include an examination of collections and disbursements of child support payments for proper processing and accounting treatment. Although OCSE's general instructions for the collection of data used for its annual report reminds state agencies that they should report reliable and complete information, OCSE officials told us they have only reviewed data on undistributed collections in special circumstances. For example, the Department of Health and Human Services and OCSE conducted at least three special reviews of California's undistributed collections data since fiscal year 1994 that revealed problems with the accuracy and reliability of the data. According to OCSE officials, the agency does not have the resources to routinely review data on undistributed collections in the way it reviews other program data. Reviews of undistributed collections data do not have to be done the same way other program data are reviewed. To minimize the impact on OCSE's resources, the reviews of the undistributed collections data could be done in conjunction with one of the other routine audits and could be limited to a portion of the state agencies on a rotational basis, for example, one third of the states each year. Without more accurate data on undistributed collections, OCSE cannot be certain about the amount of collections that are not being distributed to families in a timely manner. Further, without accurate data, OCSE cannot determine whether undistributed collections are more problematic for certain state agencies than for others. Federal Law, Some State Policies, and Inaccurate, or Missing Information Delayed or Prevented Distribution of Certain Collections: The underlying causes for nearly all undistributed collections were federal law, state policies, as well as inaccurate or missing information. Federal law allowed state agencies to hold collections from joint tax refunds and state agencies set polices that guided how long these refunds and other collections were held. Invalid addresses for custodial parents or missing case information also contributed to undistributed collections. For example, information needed to determine the correct amount of the collection to be distributed to families who are or were receiving public assistance was sometimes inaccurate or missing, and delayed distribution of these collections. Our analysis of state agency survey data found that many state agencies reported holding more that $1 million from joint tax refunds and several hundred thousand in other types of collections. Federal Law and State Policies Determined How Long Certain Collections Were Held: Federal law and policies in some state agencies contributed to certain collections being counted as undistributed. While state agencies are generally required to distribute collections in 2 days, federal law allows state agencies to hold collections from joint tax refunds for up to 180 days. Our survey results revealed that the majority of state agencies held collections intercepted from joint tax refunds for 180 days and a few distributed them in 1 to 2 days. Table 4 summarizes this survey data. Table 4: Time Period Collections from Joint Tax Refunds Are Held: Number of state agencies: 4; Number of days collections are held: 1-2. Number of state agencies: 2; Number of days collections are held: 23-30. Number of state agencies: 2; Number of days collections are held: 60[A]. Number of state agencies: 4; Number of days collections are held: 99-170. Number of state agencies: 34; Number of days collections are held: 180[B]. Source: GAO survey. Note: Two state agencies did not answer this question. [A] One state indicated that it distributed these collections 60 days after the end of the month it received them. [B] We converted the responses that were 6 months to 180 days. One state reported that it distributed collections from tax refunds within 1 day following the 180-day hold and another state indicated 2 days, following the 180-day hold. [End of table] In addition, state agencies had policies on whether or not they held child support collections received before they were due.[Footnote 13] Depending on how early collections are received, they may be held for several months and counted as undistributed collections during that time. Twenty-one state agencies reported that they immediately distributed collections that were received before they were due for non-TANF cases and 20 state agencies reported that they held these collections until they were due. Seven state agencies had other policies. For example, 1 state agency reported that it authorized its case managers to distribute the collections based on the preferences of the custodial parent and another state reported it would hold as much as 1 month of support received before it was due and refund the remaining amount to the noncustodial parent. State policies regarding how long undistributed collections are held before they are declared abandoned affects reported amounts of undistributed collections. Eleven state agencies reported that their policies require undistributed collections to be declared abandoned property after 1 year, while 1 state indicated its policy was to hold collections for up to 7 years. Eleven state agencies did not specify a number of years after which undistributed collections were declared abandoned property. Figure 5 shows the range of times reported by 44 state agencies. Figure 2: Number of Years State Agencies Reported Holding Collections before Declaring Them Abandoned Property: [See PDF for image] Notes: Three state agencies did not report whether or not they specified number of years. One state agency reported that it specified a number of years after which undistributed collections are reclassified as abandoned property, but did not indicate how many years. [End of figure] Inaccurate or Missing Information Contributed to Undistributed Collections: Some undistributed collections were the result of invalid addresses for custodial parents or missing information. State officials told us that custodial parents often moved without informing the child support agency of their new address and that the U.S. Postal Service did not forward collections to new addresses, but rather returned them to the child support enforcement agency. According to state officials, noncustodial parents did not always include identifying information on their payments. For example, state officials said they sometimes received money orders from noncustodial parents that were illegible or lacked information, such as the case number or full name, needed to match them to the right case. Missing or inaccurate information on TANF or former TANF cases contributed to undistributed collections. Officials from 2 state agencies we visited explained that some collections could not be distributed to the custodial parent until they determined the correct amount owed to the government. According to Florida officials, this contributed to nearly 50 percent of its approximate $28 million in undistributed collections. The state cited two major reasons for this problem: (1) data inaccuracies due to the conversion of cases from the former statewide system to the state agency's new automated system and (2) difficulties exchanging information between the automated systems of the child support agency processing the payments and the TANF agency. A 2001 report issued by the Department of Health and Human Services' Office of Inspector General found that 11 state agencies experienced difficulties in distributing child support to families leaving TANF.[Footnote 14] Among the reasons state agencies cited were problems exchanging information with TANF agencies and inaccurate addresses for custodial parents receiving TANF benefits. The report also noted that 28 of 51 state agencies surveyed reported problems with the automated exchanges of information between the child support enforcement and the TANF agencies. These problems included incompatible design of state TANF and child support enforcement agencies' automated systems and timing of information exchanges, which, according to the report, could have caused child support payment delays and underpayments after clients left the TANF program. State officials also said that missing information from employers as well as inaccurate payments contributed to the amount of undistributed collections. For example, according to state officials, employers, including some federal agencies, did not always identify the cases for which the withheld wages were designated or mailed one check for multiple cases with the sum of the withheld wages not matching the total amount of the check.[Footnote 15] Additionally, some employers sent payments for more than was due. In its examination of undistributed collections, a report from a private firm indicated that in New York City, 40 percent of undistributed collections were from payments that were not due. In some instances employers sent inaccurate payments because the state agencies had not notified them that the amount to be withheld had changed. Such changes would be needed in cases such as those in which previously owed child support had been paid and only current support was to be withheld, or if the child had reached the age at which child support payments terminated. Officials from 1 state agency we visited, told us that they waited to notify employers about changes in the amount to be withheld until over- payments had been collected. In other cases, officials told us that employers sent inaccurate payments because they did not always correctly calculate the amount to be withheld. For example, if a child support order stipulated that $100 was to be paid each month and $50 were deducted in each of 26 biweekly pay periods, at the end of the year, the noncustodial parent would have paid $1,300 instead of the $1,200 that was owed. Depending on the state's policy, this $100 overpayment could be returned to the noncustodial parent, distributed to the custodial parent before it was due, or held until it was due. Many State Agencies Reported Holding More than $1 Million from Joint Tax Refunds and Several Hundred Thousand Dollars in Other Types of Undistributed Collections: In response to our survey, 32 state agencies provided dollar amounts for undistributed collections from joint tax refunds. The median value reported for these collections was $1.8 million. Of these 32 state agencies, 19 reported an amount of $1 million dollars or higher with 3 reporting amounts greater than $10 million dollars. In 15 state agencies this was the largest amount reported for any of the nine types of undistributed collections we listed on the survey. For the 9 state agencies that provided values for all nine types, we determined that undistributed collections from joint tax refunds ranged from 27 to 48 percent of total undistributed collections. Our survey requested data as of June 2003, and OCSE officials explained that the amount of undistributed collections from joint tax refunds is generally higher in March through September. Many officials cited the potential financial loss as the primary reason they are unwilling to assume the risk of releasing these collections before 180 days. State agencies are fully responsible for payments made in error and must either attempt to recover money that has been distributed to custodial parents or suffer the financial loss that comes from reimbursing the Treasury for the "injured spouse" claims. One state agency we visited, Texas, reduced the time it held collections from joint tax refunds from 120 days to 90 days after analysis of its data showed that the benefit of distributing these collections outweighed the financial risk of holding them. While high values were consistently reported for undistributed collections from joint tax refunds, our analysis also revealed that the median value of four other types of undistributed collections that state agencies reported exceeded $350,000. These undistributed collections included those received before they were due, pending legal resolution, with an invalid address for custodial parents, and with data problems. Table 5 illustrates the median amounts and ranges reported for the nine types included in our survey as well as the number of state agencies that reported an amount. Table 5: Amounts for Nine Types of Undistributed Collections Reported by State Agencies as of June 30, 2003: Types of undistributed collections: Collections from joint tax refunds; Median amount (rounded): $1,750,000; Minimum amount: 0; Maximum amount (in millions): $13.9; Number of state agencies that reported an amount: 32. Types of undistributed collections: Collections received before they were due; Median amount (rounded): $466,000; Minimum amount: 0; Maximum amount (in millions): $8.0; Number of state agencies that reported an amount: 33. Types of undistributed collections: Collections pending legal resolution; Median amount (rounded): $431,000; Minimum amount: $9,700; Maximum amount (in millions): $10.2; Number of state agencies that reported an amount: 24. Types of undistributed collections: Collections with an invalid address for custodial parents; Median amount (rounded): $399,000; Minimum amount: $1,300; Maximum amount (in millions): $5.2; Number of state agencies that reported an amount: 35. Types of undistributed collections: Collections with data problems (overpayments, no active case, missing or inaccurate data, etc.); Median amount (rounded): $363,000; Minimum amount: 0; Maximum amount (in millions): $14.2; Number of state agencies that reported an amount: 30. Types of undistributed collections: Collections sent to custodial parents that can no longer be cashed; Median amount (rounded): $125,000; Minimum amount: 0; Maximum amount (in millions): $5.8; Number of state agencies that reported an amount: 19. Types of undistributed collections: Collections lacking sufficient information to identify them; Median amount (rounded): $94,000; Minimum amount: $1,100; Maximum amount (in millions): $4.5; Number of state agencies that reported an amount: 35. Types of undistributed collections: Collections to be distributed in 2 business days; Median amount (rounded): $65,000; Minimum amount: 0; Maximum amount (in millions): $4.7; Number of state agencies that reported an amount: 26. Types of undistributed collections: Collections to be distributed to other government programs; Median amount (rounded): $6,200; Minimum amount: 0; Maximum amount (in millions): $3.7; Number of state agencies that reported an amount: 20. Source: GAO survey. Note: The types of undistributed collections are slightly different from the types OCSE uses because we administered our survey before the 34A form was officially revised. [End of table] State Agencies Took Steps to Reduce Undistributed Collections: State agencies we visited took various steps to better understand and resolve undistributed collections. They devoted resources to analyzing their undistributed collections to help identify the specific causes. Additionally, 4 state agencies we visited established specific performance goals and all state agencies we visited emphasized regular monitoring of undistributed collections. State agencies also implemented processes to resolve cases with missing information and used electronic processes that helped to reduce the number of collections with invalid addresses for custodial parents and unidentified collections. State Agencies Analyzed and Monitored Undistributed Collections: To help reduce their undistributed collections, officials from 5 of 6 state agencies we visited stated that they devoted resources to better understanding these collections. Beginning in 2000, California assembled a team from various units within the state child support enforcement agency, such as the fiscal and information technology units, to analyze their undistributed collections. The initiative lasted 3 years and included the design and implementation of a new collections and distribution reporting system, verification of the amount and sources of undistributed collections, and publication of a report in June 2003. Another state agency, New York, awarded a yearlong, million dollar contract to a private firm to analyze its undistributed collections. The contractor developed a sampling plan and reviewed thousands of undistributed collections to determine their sources and how long the money had been held. In addition, the contractor evaluated technological solutions and identified general strategies for reducing the largest types of undistributed collections. Officials from 4 state agencies we visited emphasized that in order to reduce undistributed collections, it is necessary to establish performance goals and measures and compare program results with those goals. Four of the state agencies we visited established performance goals and measures to help them monitor undistributed collections. Florida established a series of performance accountability measures, including resolution of collections requiring additional research and timeliness of disbursements, to track the overall operation of its disbursement unit. Virginia established its goal for undistributed collections at 3 percent of monthly collections and was considering lowering that baseline to 2 percent as well as maintaining less than $50,000 in unidentified collections. An objective in California's strategic plan is to ensure that no more than 1 percent of collections due families remain undistributed at any time. Texas established benchmark amounts for each type of undistributed collections. Officials from all of the state agencies we visited highlighted the importance of monitoring undistributed collections to ensure that goals are being met. Three of the state agencies we visited had detailed automated reports to help them monitor undistributed collections. These reports varied from state to state, but the examples provided included data on the total dollar value of undistributed collections, number of cases, number of collections, and dates collections were received. Also, several state agencies produced data reports for each field or local office. For example, a monthly management report in Virginia enabled the state officials to monitor the performance of its 22 district offices. According to our survey, nearly all state agencies indicated that they routinely produced statistical data on undistributed collections. For example, 27 state agencies produced daily reports and 12 produced weekly reports.[Footnote 16] State Agencies Took Steps to Address Missing Information and Improve Payment Processes: State agencies we visited dedicated staff and focused resources on resolving cases with missing information and invalid addresses. Some state agencies used automated methods to receive and transmit collections. Additionally, 1 state agency automated two processes that helped reduce its undistributed collections. Resolving Missing Information and Invalid Addresses: Officials from all of the state agencies we visited highlighted the importance of dedicating staff to researching collections involving missing information. In Virginia, the state disbursement unit separated collections with missing information, such as a social security or case number, from those with this information. The team contacted employers, courts, or other state agencies to obtain the needed information. A database of the most difficult collections along with contact information and instructions was maintained in order to reduce processing time on similar collections received in the future. The team typically resolved unidentified collections within 72 hours. Other state disbursement units we visited also had specific staff dedicated to researching payments with missing information in order to distribute them as soon as possible. State agencies also focused resources on resolving cases with invalid addresses for custodial parents. State officials cited the importance of allowing staff to access federal, state, and private databases to locate custodial parents. Officials in 1 state said they had a contract with a private vendor because they found this information to be most useful. Virginia's system automatically searched databases to match cases that needed information. Such databases included FPLS,[Footnote 17] state licensing agencies, and credit bureaus. In Texas, an indicator was added to case records to allow field workers to automatically refer cases with invalid addresses to special enforcement investigators. In addition, a new report was created to show all cases with collections held due to missing addresses so that field staff could focus their efforts on these cases. This change resulted in the distribution of almost $1 million in the first 4 months of implementation. Also, state officials from the 4 state disbursement units we visited told us that customer service representatives routinely verified addresses when contacted by custodial or noncustodial parents. Using Automated Processes: Officials from 4 of the 6 state agencies we visited encouraged employers to use Electronic Funds Transfer (EFT) when transmitting collections to the state disbursement unit as a way to improve efficiency and reduce undistributed collections. According to OCSE officials and a representative from the American Payroll Association, converting to electronic payments for child support means having to purchase software or make programming changes to payroll systems, which can initially be costly. As an alternative to EFT, 5 state agencies, including 1 we visited, Florida, developed a process to allow employers to send payments to the state disbursement unit over the Internet. A state agency's Web based payment service operates much like other online banking services in that an employer registers for the service, receives a user identification and password and can then access the Web site each pay period to make the payments for each of its employees. State agencies generally offered these services free of charge to employers. State agencies also provided alternative ways for noncustodial parents to transmit their payments. Florida offered noncustodial parents the option of paying their support over the Internet. Texas had a pilot project with a grocery chain to allow noncustodial parents to pay their child support at their stores, and then the stores would transmit the payment to the state disbursement unit electronically. In another pilot project, New York partnered with a private vendor that sells money orders. Noncustodial parents could provide the payment to the vendor who would then transmit the payment electronically to the state disbursement unit. In addition to receiving collections electronically, many state agencies distributed collections to parents electronically. Thirty- seven state agencies that responded to our survey indicated they offered direct deposit--a process whereby money is directly transferred to a checking or savings account--and 9 reported that they issued a state debit card. Officials from all the state agencies we visited explained that direct deposit is not an option for many of their customers because they do not have bank accounts. For anyone without a bank account, a state debit card is an alternative. The card may be used to purchase goods or services as well as to obtain cash. Safeguards have been included that prevent custodial parents from withdrawing cash or making purchases that would cause an overdraft of the available child support funds. In response to our survey, most state agencies indicated that they send 40 percent or fewer of their collections to parents electronically. Figure 6 shows the portion of collections that state agencies reported distributing to parents electronically. Figure 3: Portion of Collections Distributed to Parents Electronically: [See PDF for image] [End of figure] While most state agencies offered at least one form of electronic distribution to parents, in 2003, Iowa began requiring all parents to receive collections electronically. Exceptions to this requirement were based on individual circumstances.[Footnote 18] Officials told us that as of January 2004, about 95 percent of custodial parents in Iowa were receiving collections electronically. Automated processes to receive and distributed collections can help reduce undistributed collections. According to state agency officials, methods that allow state agencies to receive collections electronically can help reduce undistributed collections due to fewer incorrect or unidentified collections. Electronic methods for distributing collections to families can also reduce the number of undistributed collections. With direct deposit and state debit cards, funds are transferred from the state directly to a financial institution and even if the custodial parent changes addresses, the collections can continue to be deposited. As such, by using direct deposit and state debit cards, undistributed collections due to invalid addresses of the custodial parents and those due to paper checks that can no longer be cashed are reduced. In addition to reducing undistributed collections, officials from 2 state agencies told us that electronic distribution to parents reduces many processing costs and requires limited manual intervention. Iowa officials predicted a combined savings to the state and federal government of about $35,000 a month by sending collections to parents electronically. Officials in Texas reported a cost savings of $1.34 per transaction by using direct deposit versus mailing a check, a savings of over $300,000 in a typical month. Additionally, Texas automated two of its child support processes that will help prevent undistributed collections. In 2000, Texas established an automated mechanism to issue refunds to noncustodial parents for collections that exceeded the amount owed. Previously, manual intervention was required to release these collections. According to officials, this program saved hundreds of hours of staff time and enabled them to quickly process refunds. Also in 2000, Texas developed an automated process to issue new wage withholding orders to employers when circumstances of cases changed, such as a reduction in the amount owed. In the first 60 days after implementation, Texas issued over 15,000 new orders to employers. State Agencies Reported Earning Interest and Income: Many state agencies reported earning interest and program income. OCSE issued a memorandum in 1989 that encouraged state agencies to deposit all child support collections in interest-bearing accounts. According to our survey, 32 state agencies reported that they kept all undistributed collections in interest-bearing accounts, 4 state agencies reported keeping some undistributed collections in interest- bearing accounts and 11 state agencies reported that they did not keep their collections in these accounts. One state agency did not answer the question. State agencies are required to deduct interest earned and income from program costs and can charge fees to help recover some of their administrative costs. For example, state agencies are required to charge non-TANF families an application fee, and can charge fees for tax refund offsets and various services and expenses such as case maintenance fees, or a fee to establish a support order. Our prior reports concluded that most states either did not charge fees or charged minimal fees.[Footnote 19] For fiscal year 2002, state agencies reported earning almost $50 million in interest and income. The amounts reported ranged from $0 to $15 million. OCSE Has Assisted States' Efforts to Reduce Undistributed Collections, but the Department of the Treasury Has Not Provided Information That Would Help States Distribute Collections from Some Joint Tax Refunds Sooner: OCSE has provided some assistance to help state agencies reduce their undistributed collections. OCSE focused attention on reducing undistributed collections, shared best practices, and funded projects. However, OCSE does not have information on some "injured spouse" claims that could reduce the amount of time collections from some joint tax refunds are held. OCSE Provided Some Assistance to State Agencies: In fiscal years 2002 and 2003, OCSE focused on reducing undistributed collections. In its fiscal year 2002 annual report, OCSE stated that it was taking actions to understand the composition of undistributed collections and identify efforts state agencies could take to distribute more collections to families. Additionally, OCSE emphasized that improved customer service by state agencies helps reduce undistributed collections. According to state officials, customer service representatives are encouraged to ask for updated information from parents, thereby reducing the number of cases with incorrect addresses. In fiscal year 2003, OCSE partnered with the National Council of Child Support Directors to refine the categories of undistributed collections and obtain state data and best practices. Also, OCSE issued guidance in March 2003 to reduce the number of cases with invalid addresses. This guidance informed state agencies that they could access FPLS to locate custodial parents as well as to locate noncustodial parents. OCSE funded research and provided technical assistance to state agencies to help them reduce undistributed collections. Between fiscal years 2000 and 2002, OCSE awarded three contracts. The first contract awarded in fiscal year 2000, for about $135,000, funded research to identify approaches for reducing undistributed collections in 11 state agencies with large caseloads or amounts of collections. In addition, this contractor reviewed undistributed collections in two New York counties and identified factors in their business processes and automated systems that prevented them from further reducing these collections. According to OCSE, a second contract was also awarded in fiscal year 2000 for about $112,000 that funded research focused on understanding the extent and causes of undistributed collections across state agencies and highlighting best practices for distributing such collections. Additionally, OCSE officials said that a third contract was awarded in fiscal year 2002 for about $300,000 that funded research to review undistributed collections in 5 state agencies. Beginning in fiscal year 2000, OCSE made projects designed to reduce undistributed collections a priority for demonstration grants, and awarded five grants.[Footnote 20] In fiscal year 2000, OCSE awarded one grant for about $188,000, and in fiscal year 2002, OCSE awarded four grants--one each to Texas, and the District of Columbia and two to Indiana--for a total amount of about $500,000. The goals for each of the fiscal year 2002 projects were different. For example, the goal for one project was to evaluate the use of state debit cards as a way to reduce undistributed collections. Further, OCSE actively encouraged more use of EFT, which as previously stated can help reduce undistributed collections. In July 2003, the OCSE Commissioner sent a letter to 80 private sector employers that employ a large number of noncustodial parents encouraging them to use EFT to pay the child support they withheld from their employees' wages. During several conferences sponsored by employer organizations, OCSE promoted electronic payment by distributing literature and making presentations. Additionally, in 2002 OCSE staff assisted the NACHA-- Electronic Payments Association's Child Support Task Force by helping to identify issues associated with promoting the electronic collection and distribution of child support payments. OCSE also worked with several federal agencies on issues related to electronic payment. For example, OCSE worked with the Defense Finance and Accounting Service to encourage more use of electronic payments. OCSE shared information about initiatives state agencies took to reduce undistributed collections in its publications. Each month OCSE published its "Child Support Report" with information about various child support topics. Descriptions of successful state efforts related to undistributed collections were featured in several editions. OCSE also published an annual compendium of best practices in child support enforcement and several of the entries in its 2002 edition were related to undistributed collections. In addition to these publications, OCSE has discussed or arranged sessions on undistributed collections at conferences, forums, and training sessions. OCSE officials also reported that regional meetings have included sessions focused on reducing undistributed collections. For example, officials from Region VII organized a workshop where perspectives on reporting and best practices were addressed.[Footnote 21] In July 2003, OCSE initiated monthly audio conference calls to foster discussions on implementing best practices such as electronic funds distribution. As of December 2003, OCSE had arranged five calls with a range of 25 to 38 state agencies participating.[Footnote 22] As part of our survey, we asked state agencies how helpful various OCSE efforts related to undistributed collections have been. For the most part, state agencies reported that OCSE's efforts have been helpful and many reported that discussions of best practices at forums were greatly or extremely helpful. However, several state agencies reported that OCSE's efforts were hardly or not at all helpful. Table 6 summarizes the state agencies' responses. Table 6: Summary of State Agencies' Responses Regarding the Helpfulness of OCSE Efforts: OCSE effort: Dissemination of best practices; Hardly or not at all helpful: 5; Somewhat or moderately helpful: 31; Greatly or extremely helpful: 9; Have not participated in effort: 2. OCSE effort: In-person training; Hardly or not at all helpful: 7; Somewhat or moderately helpful: 9; Greatly or extremely helpful: 4; Have not participated in effort: 27. OCSE effort: Discussion of best practices at forums; Hardly or not at all helpful: 7; Somewhat or moderately helpful: 18; Greatly or extremely helpful: 15; Have not participated in effort: 7. OCSE effort: Policy documents; Hardly or not at all helpful: 8; Somewhat or moderately helpful: 30; Greatly or extremely helpful: 6; Have not participated in effort: 2. Source: GAO survey. Note: One state did not answer this question at all and 1 state did not provide a response about the helpfulness of the policy documents. [End of table] The Department of the Treasury Has Not Provided OCSE Information on Collections from Some Joint Tax Refunds: OCSE and state agencies are not receiving information from the Department of the Treasury about approved "injured spouse" claims that are submitted at the time a tax return is filed. Such information is important because state agencies may delay disbursement of tax-offset collections for a period of up to 180 days to allow for the possibility of a reversal based on an "injured spouse" claim. IRS officials explained that while they do not know the type of debt owed when they process the claims, they have estimated that about 55,000 "injured spouse" claims related to the offset of non-TANF joint tax refunds for child support have been filed with the tax returns each year.[Footnote 23] Additionally, these officials commented that when the claims were filed with tax returns and approved, the amount due the "injured spouse" was allocated before the file was sent to FMS. However, the information FMS sent to OCSE, which was then forwarded to the state agencies, did not identify these collections as having had their "injured spouse" claims satisfied. As our survey results show, most state agencies delayed distributing all collections from joint tax refunds. IRS officials explained that federal law and current processes have played a role in determining the information provided to OCSE on approved "injured spouse" claims. IRS officials stated that their disclosure statute[Footnote 24] allows, but does not require, them to provide OCSE and the states with information on specific tax offsets for payment of past due child support. The officials also explained that they include data on the payment file they send to FMS that could enable FMS to determine whether the "injured spouse" claim has been processed. Furthermore, the IRS officials said that it would be very costly to reprogram the IRS data systems to enable them to provide OCSE information on "injured spouse" claims. FMS officials we spoke with stated that until recently, they were not aware that OCSE needed notification on the payment of injured spouse claims[Footnote 25] and they expressed concern about the cost associated with changing their system in order to provide the information to OCSE. The FMS officials added that neither IRS nor OCSE emphasized the need for information on injured spouse claims when FMS became responsible for and set up their systems to support the tax- offset program. Furthermore, in order for FMS to send OCSE more information about injured spouse claims, they would have to reprogram their data files and change at least two data systems. They also stated that it would take at least 3 years to modify the systems and would be very costly. FMS officials suggested that the more cost-effective solution would be for IRS to send OCSE the requested information: Although the offset program has been operating for almost 20 years, OCSE's efforts to obtain the information on "injured spouse" claims from the Department of the Treasury began in late 2001. According to OCSE officials, they have discussed the need for additional information with officials from the Department of the Treasury on several occasions. Further, in February 2003, OCSE's Commissioner sent a letter to the Department of the Treasury requesting that the two agencies involved in the Tax Offset Program, IRS and FMS, provide information that identifies which joint tax refunds involve "injured spouse" claims. FMS and IRS formed a work group to investigate ways to shorten the process related to satisfying these claims. However, according to an FMS official, there has been little movement in response to OCSE's letter and as of January 2004, the Department of the Treasury had not responded to OCSE's letter. If the Department of the Treasury provided OCSE and state agencies with information on the satisfied "injured spouse" claims filed with the initial tax returns, state agencies could immediately release the offset collections to the families. Conclusion: Receipt of child support is critical for many custodial parents and their children. However, no one is certain about the amount of child support collections that are not distributed to families on time, if at all. By revising the quarterly collections form, OCSE has taken the first step to improving data about undistributed collections, but more reliable data are needed in order for OCSE and the state agencies to know more about undistributed child support collections and to be able to take appropriate actions to help reduce them. While the total amount of undistributed child support collections is uncertain, it is clear that millions of dollars being held for months are collections from joint tax refunds. State agencies need more information about those "injured spouse" claims that are filed with the tax returns and approved. As the federal partner and overseer of this program, it is OCSE's role to work with other federal agencies, such as the Department of the Treasury, to remove barriers that hinder fulfilling its mission. Furthermore, since collections held from joint tax refunds represent a large amount of state agencies' total undistributed collections, it is in the best interest of the child support program for OCSE to focus more attention on getting this information. Additionally, the Secretary of the Treasury needs to provide OCSE and state agencies information about satisfied "injured spouse" claims. If the Department of the Treasury provides this information, collections held from some joint tax refunds could reach families sooner. Recommendations for Executive Action: To better measure the amount of and help reduce undistributed collections, we are making three recommendations. We recommend that the Secretary of Health and Human Services direct the Commissioner of OCSE to: * review undistributed collections data from state agencies periodically in conjunction with one of the other routine reviews to help improve the accuracy of the data and: * work closely with the Department of the Treasury to identify a cost- effective approach for obtaining information on "injured spouse" claims in order to enable collections from some joint tax refunds to reach families sooner. We also recommend that the Secretary of the Treasury direct the Commissioner of IRS and the Commissioner of FMS to work together with OCSE to identify a cost-effective approach for providing OCSE information needed to identify those collections that have had their "injured spouse" claims satisfied so that these collections can be distributed to families sooner. Agency Comments and Our Evaluation: We received written comments on a draft of this report from HHS. These comments are reprinted in appendix III. The department did not explicitly agree or disagree with either of our recommendations. In response to our recommendation to review undistributed collections data, HHS stated that conducting these reviews in conjunction with the data reliability audits would substantially increase the time needed to complete them due to the variability of the undistributed collections data and that the data reliability audits must be completed on time. As we stated in our report, such reviews of undistributed collections data could be done in a number of ways and would not necessarily have to be done with data reliability audits. Also, HHS stated that its audit resources are insufficient to routinely audit any other function or area except data reliability. We also noted their concern about limited resources in the report. With regard to the recommendation that OCSE work with IRS and FMS, HHS stated that it has been working with IRS and FMS. We added more detail to this section in the report. HHS also agreed with our finding that information on approved injured spouse claims could significantly reduce undistributed collections and get money to families in a timelier manner. However, the HHS comments also noted concerns about a direct exchange of information between IRS and OCSE because OCSE's current interaction is strictly with FMS, and it would be more complex, time consuming and costly to add the additional interfaces and processing that would be required for a direct exchange with IRS. We did not recommend a specific approach for sharing the information. We also received written comments on a draft of this report from the Department of the Treasury. These comments are reprinted in appendix IV. The Department of the Treasury agreed with our recommendation. In commenting, the Acting Chief Financial Officer stated that although neither IRS nor FMS can readily provide OCSE information on satisfied injured spouse claims, and that significant programming changes would be needed, IRS and FMS would work together with OCSE to develop a cost- effective way to advise OCSE when injured spouse claims have been satisfied. In addition to written comments, officials from HHS, FMS, and IRS provided technical comments. We incorporated these comments in the report as appropriate. As arranged with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days from its issue date. At that time we will send copies of this report to appropriate congressional committees, the Secretary of Health and Human Services, the Secretary of the Treasury, and other interested parties. In addition, this report will be available at no charge on GAO's Web site at http://www.gao.gov. If you have any questions regarding this report, please call me on (202) 512-8403. Other contacts and acknowledgments are listed in appendix V. Sincerely yours, Cornelia M. Ashby: Director, Education, Workforce, and Income Security: [End of section] Appendix I: Scope and Methodology: To accomplish our objectives, we conducted a mail survey of all 54 IV- D child support enforcement directors, conducted site visits to 6 state agencies, interviewed federal and state officials, as well as reviewed related reports and analyzed laws and regulations. We sent each director a self-administered mail-back questionnaire. To ensure that our survey questions were clear, unbiased, specific, and easy to understand, we pretested the survey instruments in 3 state agencies and discussed the appropriateness of the survey questions with OCSE. We electronically mailed follow-up letters and replacement copies of the survey to nonrespondents to encourage response. Forty-eight state agencies participated in the survey. Nonrespondents included Arizona, Indiana, Guam, Mississippi, New Jersey, and Virgin Islands. We requested data as of June 30, 2003, and administered the survey from September to November 2003. We did not assess the reliability of the data reported by state agencies in response to our survey. However, we reviewed the data for completeness and reasonableness. We visited Virginia, New York, Florida, Texas, Iowa, and California. We selected the state agencies so as to obtain diversity in the amount of collections and balances of undistributed collections, geographic location, number of clients served, and whether the child support program was county or state administered. Of the state agencies we visited, 3--Iowa, Texas, and Virginia--had undistributed collection balances lower than the national average of 3 percent and were cited by the Office of Child Support Enforcement (OCSE) as having best practices in reducing and or improving payment distribution practices. Two of the state agencies--California and New York--had balances of undistributed collections that were higher than the national average of 3 percent while 1 state, Florida, had a balance of about 3 percent. In addition to our overall approach, we took specific steps for each of the objectives. To address how the data on undistributed collections has changed in the last few years, we reviewed and analyzed OCSE data on undistributed collections for fiscal years 1999 to 2002 as reported by state agencies. We began with fiscal year 1999 data because that was the earliest year with data comparable to fiscal year 2002; the most current year data were available at the time of our review. As a part of our analysis we (1) reviewed and compared the data for each fiscal year and (2) identified issues associated with the calculation and reporting of undistributed collections through interviews with state and OCSE officials. Additionally, we analyzed survey responses that identified the types of funds state agencies classify and report as undistributed collections. We cited findings from an internal review conducted in California, however, we did not verify the methodology used in this review. To identify the causes of undistributed collections, we reviewed relevant literature and past audit reports on the causes of undistributed collections. We cited relevant findings from two of these reports. The report issued by the Department of Health and Human Services Office of Inspector General was conducted in accordance with Quality Standards for Inspections issued by the President's Council on Integrity and Efficiency. We did not verify the methodology used in the report produced by a private firm. We also reviewed laws, regulations, and policies governing child support distribution practices to gain an understanding of the required time frames for distribution of child support payments. We interviewed child support advocates and experts to obtain their opinions on the causes for undistributed collections. To identify what state agencies are doing to reduce undistributed collections and to determine what guidance OCSE has given state agencies to help reduce undistributed collections, we reviewed relevant documents and agency audit reports. We also reviewed documents issued by the National Council of Child Support Directors and collected information on relevant OCSE-funded research. In addition, we also interviewed child support advocates and experts, as well as federal and state officials to obtain their perspectives on how OCSE has helped state agencies reduce undistributed collections and additional actions that OCSE can take. [End of section] Appendix II: Survey of State Directors of Child Support Enforcement: [See PDF for image] [End of survey] [End of section] Appendix III: Comments from the Department of Health and Human Services: DEPARTMENT OF HEALTH & HUMAN SERVICES Office of Inspector General: Washington, D.C. 20201: MAR 12 2004: Ms. Cornelia M. Ashby Director, Education, Workforce, and Income Security Issues United States General Accounting Office Washington, D.C. 20548: Dear Ms. Ashby: Enclosed are the Department's comments on your draft report entitled, "Child Support Enforcement - Better Data and More Information on Undistributed Collections Are Needed." The comments represent the tentative position of the Department and are subject to reevaluation when the final version of this report is received. The Department provided several technical comments directly to your staff. The Department appreciates the opportunity to comment on this draft report before its publication. Sincerely, Signed by: Dara Corrigan: Acting Principal Deputy Inspector General: Enclosure: The Office of Inspector General (OIG) is transmitting the Department's response to this draft report in our capacity as the Department's designated focal point and coordinator for General Accounting Office reports. OIG has not conducted an independent assessment of these comments and therefore expresses no opinion on them. COMMENTS OF THE DEPARTMENT OF HEALTH AND HUMAN SERVICES ON THE GENERAL ACCOUNTING OFFICE'S DRAFT REPORT, "CHILD SUPPORT ENFORCEMENT: BETTER DATA AND MORE INFORMATION ON UNDISTRIBUTED COLLECTIONS ARE NEEDED," (GAO-04-377): The Department of Health and Human Services (HHS) appreciates the opportunity to comment on the General Accounting Office's (GAO) Draft Report. Recommendation 1: We recommend that the Secretary of HHS direct the Commissioner of the Office of Child Support Enforcement (OCSE) to: * review undistributed collections (UDC) data from State agencies periodically in conjunction with one of the other routine reviews to help improve the accuracy of the data. Response: GCSE revised the 34A reporting form and related instructions. Categories and types of collections to be reported as undistributed will be reported under the headings: 1) "Undistributed collections pending distribution," and 2) "Undistributed collections unresolved," with detailed instructions on what should be included under each category. Use of these headings, with accompanying definitions, will help ensure consistency in reporting across the States. It will also enable GCSE to accurately assess the nature of UDC in the States and better target technical assistance. As States become more experienced in reporting UDC under these headings, GCSE will evaluate the data received in the new reports and determine which States need additional analysis. If the GAO's recommendation intends that UDC reviews be performed in conjunction with the Data Reliability Audits (DRA), then the recommendation would create substantial problems for the States and difficulties for GCSE. Adding a review of UDC to the reliability audits would substantially increase the time needed to complete the audits due to the variability of UDC records and the complexity of auditing a UDC balance for accuracy and reliability. The DRA cannot be delayed because GCSE uses the findings to calculate the award of performance incentives, which the States rely on for budget purposes. GCSE and States also use the findings for corrective actions that need to be made to avoid penalties. GCSE reviews UDC with its discretionary time after the DRAB are complete. In this regard, GCSE plans to look at California UDC later this year after the DRA work is complete, if contract resources are not available to provide timelier audit services. Due to limited resources, discretionary audit staff time is scheduled to areas that will have the greatest overall benefit to the child support enforcement program. This could include UDC, as well as other important areas, particularly audits of administrative costs. However, audit resources are insufficient to routinely audit any other function or area except data reliability. Recommendation 2: We recommend that the Secretary of HHS direct the Commissioner of the OCSE to work closely with the Department of the Treasury to identify a cost- effective approach for obtaining information on "injured spouse" claims in order to enable collections from joint tax refunds to reach families sooner. We recommend that the Secretary of the Treasury direct the Commissioner of the Internal Revenue Service (IRS) and the Commissioner of the Financial Management Service (FMS) to work together and with GCSE to identify a cost-effective approach for providing GCSE information needed to identify those collections that have had their "injured spouse" claims satisfied so that these collections can be distributed to families sooner. Response: GCSE has been working with IRS and FMS to gain additional information on the filing of injured spouse claims in order to be able to share that information with States. Currently, OCSE is notified if the offset is related to a joint return. It is important to know the processing status of an injured spouse claim in order to assist the States in distribution of the offset. For example, when it is known that an injured spouse claim has been filed and processed the States could immediately release the collection to the family, rather than holding the money for 180 days. This information could significantly reduce UDC and get the money to families in a timelier manner. There are concerns regarding the mention of a direct exchange of information between IRS and OCSE, as OCSE's interaction is strictly with FMS. It would be more complex, time-consuming, and costly to add the additional interfaces and processing that would be required to implement this proposition. It is OCSE's understanding that the operation was transferred from IRS to FMS about 5 years ago, at least in part, to improve the efficiency and cost-effectiveness of the program. The proposition of adding a direct interface between GCSE and IRS seems contrary to the original intent of the transfer. We are pleased that IRS does not note any legal barrier to sharing this information with OCSE. As part of our ongoing effort to gain additional information on joint returns, OCSE met with FMS on March 12, 2004. [End of section] Appendix IV: Comments from the Department of the Treasury: DEPARTMENT OF THE TREASURY WASHINGTON, D.C. 20220: MAR 8 2004: Ms. Cornelia M. Ashby Director, Education, Workforce, and Income Security Issues U.S. General Accounting Office 441 G Street, N.W. Washington, D.C. 20548: Dear Ms. Ashby: The Department of the Treasury has received for comment a copy of the draft report (GAO-04-377), entitled Child Support Enforcement: Better Data and More Information on Undistributed Collections Are Needed. The draft report contains one recommendation that relates to the Department of the Treasury's role in the collection of delinquent child support debt. We concur with the recommendation that the Internal Revenue Service (IRS) and the Financial Management Service (FMS) work with the Office of Child Support Enforcement (GCSE) to identify a cost-effective approach for providing GCSE information needed to identify those collections that have had their "injured spouse" claims satisfied so that these collections can be distributed to families sooner. IRS and FMS have enjoyed a positive working relationship for many years. Their mutual interests cross many operational lines and their staffs engage in problem solving and cooperative efforts on a daily basis. As your report notes, IRS and FMS are currently collaborating on a joint project reviewing the processing of injured spouse claims to determine if the processing cycle time can be reduced. We are confident that IRS and FMS will collaboratively develop a feasible, cost effective way to provide the necessary Injured Spouse information to GCSE. Currently, neither IRS nor FMS captures injured spouse claim information in such a way as to allow either agency to readily provide the required information to OCSE, and significant programming changes would be needed by either agency to identify satisfied Injured Spouse Claims to OCSE. Nevertheless, IRS and FMS will continue to work together and with GCSE to develop a mutually satisfactory, cost- effective way to advise GCSE when injured spouse claims impacting the distribution of their collections have been satisfied. In addition, IRS and FMS will jointly develop a corrective action plan to address the work that will be needed to fulfill this requirement. Thank you for the opportunity to respond to this draft GAO report. If you have any questions or wish to discuss these comments further, please contact Floyd L. Williams (IRS, Director, Legislative Affairs) at (202) 622-4725 or Alvina McHale (FMS, Director, Legislative and Public Affairs) at (202) 874-6604. Sincerely, Signed by: Barry K. Hudson: Acting Chief Financial Officer: cc: Donald Hammond Richard Gregg Ronny S. Rhodes: [End of section] Appendix V: GAO Contacts and Staff Acknowledgments: GAO Contacts: Carolyn M. Taylor (202) 512-2974, taylorcm@gao.gov Rebecca A. Christie (312) 220-7720, christier@gao.gov: Staff Acknowledgments: In addition to those named above the following individuals made important contributions to this report: Vernette Shaw, Carolyn Boyce, Jay Smale, Corinna Nicolaou, James Rebbe, and Paul Schearf. FOOTNOTES [1] Pub. L. No. 104-193, § 312(b) (Aug. 22, 1996). [2] In this report we will refer to the state child support enforcement agencies as state agencies. [3] Under some circumstances, the Secretary of the Treasury may collect past due child support by offsetting federal tax refunds and other types of federal payments. Collections from federal joint tax refunds are held only for non-TANF cases. In this report, the use of the word "non-TANF" means that the child support is owed to a custodial parent who is not a TANF recipient. Joint tax refunds subsequently mentioned in this report refer to federal non-TANF joint tax refunds. [4] 42 U.S.C. §§ 651-669b. [5] Omnibus Budget Reconciliation Act of 1981, Pub. L. No. 97-35, § 2331 (Aug. 13, 1981). [6] IRS Form 8379. [7] The Treasury Offset Program (TOP) is a centralized debt collection program developed by FMS. TOP is designed to assist agencies in the collection of delinquent debt owed to the federal government. FMS disburses payments on behalf of over 400 federal agencies. [8] State disbursement units are also responsible for collecting and disbursing all payments under support orders, including spousal support. [9] The 1996 Personal Responsibility and Work Opportunity Reconciliation Act created TANF. TANF emphasizes the importance of work and personal responsibility rather than dependence on government benefits. After 2 years of assistance, or sooner if the state determines that the recipient is ready, TANF adults are generally required to be engaged in work or work-related activities. A lifetime limit of 60 months (or less, at the state's option) is placed on adults' receipt of cash benefits. Families receiving TANF benefits or benefits under the federally assisted foster care program or the Medicaid program automatically receive CSE services free of charge. Under PRWORA, TANF recipients generally must assign their rights to current child support payments to the state. [10] Foster care programs are authorized under Title IV-E of the Social Security Act. [11] OCSE's final annual statistical report for fiscal year 2002 was published November 2003. [12] 42 U.S.C. § 652(a)(4)(C). Moreover, the Child Support Performance and Incentive Act of 1998 requires that states have complete and reliable data for purposes of computing incentives. 42 U.S.C. § 658a(b) (5)(B). The performance measures for incentive payment base amounts are paternity establishment, support order establishment, current payments distributed, and cost effectiveness. 42 U.S.C. § 658a(b)(4). According to HHS, the Secretary is required to conduct an audit for each fiscal year for incentive purposes. [13] These collections are also referred to as future payments. [14] Distributing Collected Child Support to Families Exiting TANF. Department of Health and Human Services, October 2001. OEI-05-01-00220. [15] Wage withholding is a procedure whereby an employer automatically deducts a specified amount from an employee's wages or income to pay a child support obligation. All support orders issued after January 1, 1994, must contain provisions for wage withholding, except when there is a good cause not to require it or an alternative arrangement is reached by both parties. [16] Three of the 27 state agencies that reported producing daily reports also reported that they produced weekly reports. [17] FPLS is a computerized national location network operated by OCSE. FPLS obtains address and employer information, as well as data on child support cases in every state, compares them and returns matches to the appropriate states. This helps child support enforcement agencies locate noncustodial parents. [18] Officials told us that some exceptions were granted to individuals, for example, if they could not reasonably access a financial institution. [19] U.S. General Accounting Office, Child Support Enforcement: Opportunity to Defray Burgeoning Federal and State Non-AFDC Costs, GAO/ HRD-92-91 (Washington, D.C.: June 5, 1992); Child Support Enforcement: Opportunity to Reduce Federal and State Costs, GAO/T-HEHS-95-181 (Washington, D.C.: June 13, 1995); and Child Support Enforcement: Clear Guidance Would Help Ensure Proper Access to Information and Use of Wage Withholding by Private Firms, GAO-02-349 (Washington, D.C; Mar. 26, 2002). [20] Section 1115 of the Social Security Act authorizes OCSE to provide funding to state Title IV-D agencies for demonstration activities intended to add to the knowledge and to promote the objectives of the Child Support Enforcement Program. The four priority areas for fiscal year 2002 were to (1) increase the rate of cases with collections through better use of automation and improved public-private collaboration projects for interstate cases, (2) increase the rate of cases with collections from low-income noncustodial parents, (3) reduce and limit the amount of undistributed collections by having states better track and distribute more of the undistributed child support that they have collected, and (4) further the national goals of the Child Support Program. [21] Region VII includes Iowa, Kansas, Missouri, and Nebraska. [22] OCSE told us that these discussions are recorded and available to state agencies that could not participate in the original call. [23] About 182,500 injured spouse claims were filed related to offsets for child support and about 60 percent, 110,000, were filed with the original return. Approximately half of the claims are for non-TANF cases. [24] 26 U.S.C. § 6103(l)(10). [25] Under the Social Security Act provision on collection of past-due support from federal tax refunds, a state "may delay distribution of the amount withheld until the State has been notified by the Secretary of the Treasury that the other person filing the joint return has received his or her proper share of the refund, but such delay may not exceed six months." 42 U.S.C. § 664(a)(3)(B). While this section of law does not explicitly require Treasury to affirmatively inform OCSE or the states that an injured spouse claim has been paid, we believe that the statute contemplates that Treasury notify the state when the other person filing the joint return has received his or her share of the refund. See H.R. Conf. Rep. No. 98-925, at 56 (1984). GAO's Mission: The General Accounting Office, the investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO's commitment to good government is reflected in its core values of accountability, integrity, and reliability. Obtaining Copies of GAO Reports and Testimony: The fastest and easiest way to obtain copies of GAO documents at no cost is through the Internet. GAO's Web site ( www.gao.gov ) contains abstracts and full-text files of current reports and testimony and an expanding archive of older products. The Web site features a search engine to help you locate documents using key words and phrases. You can print these documents in their entirety, including charts and other graphics. Each day, GAO issues a list of newly released reports, testimony, and correspondence. GAO posts this list, known as "Today's Reports," on its Web site daily. The list contains links to the full-text document files. To have GAO e-mail this list to you every afternoon, go to www.gao.gov and select "Subscribe to e-mail alerts" under the "Order GAO Products" heading. Order by Mail or Phone: The first copy of each printed report is free. Additional copies are $2 each. A check or money order should be made out to the Superintendent of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a single address are discounted 25 percent. Orders should be sent to: U.S. General Accounting Office 441 G Street NW, Room LM Washington, D.C. 20548: To order by Phone: Voice: (202) 512-6000: TDD: (202) 512-2537: Fax: (202) 512-6061: To Report Fraud, Waste, and Abuse in Federal Programs: Contact: Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail: fraudnet@gao.gov Automated answering system: (800) 424-5454 or (202) 512-7470: Public Affairs: Jeff Nelligan, managing director, NelliganJ@gao.gov (202) 512-4800 U.S. General Accounting Office, 441 G Street NW, Room 7149 Washington, D.C. 20548: