This is the accessible text file for GAO report number GAO-04-225 entitled 'Corporation For National And Community Service: Better Internal Control and Revised Practices Would Improve the Management of AmeriCorps and the National Service Trust' which was released on February 20, 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 Congressional Requesters: United States General Accounting Office: GAO: January 2004: Corporation for National and Community Service: Better Internal Control and Revised Practices Would Improve the Management of AmeriCorps and the National Service Trust: GAO-04-225: GAO Highlights: Highlights of GAO-04-225, a report to Congressional Requesters Why GAO Did This Study: The Corporation for National and Community Service (the Corporation) was created to help meet community needs and expand educational opportunity by providing education awards to participants. The Corporation oversees and funds the AmeriCorps program as well as the National Service Trust (the Trust), which pays the education awards. From November 2002 to March 2003 the Corporation suspended AmeriCorps enrollments because there would not have been sufficient funds in the Trust to pay education awards. GAO was asked to determine (1) if all AmeriCorps enrollments were accurately recorded, (2) how the Corporation estimated its funding needs, and (3) if the Corporation made changes to prevent another enrollment suspension and to address requirements established in the Strengthen AmeriCorps Program Act. GAO analyzed laws, reviewed documents, interviewed officials, assessed the reliability of the Trust database, examined the model used to estimate funding needs, and surveyed AmeriCorps grantees. What GAO Found: Discrepancies between information in the Trust database and participant documentation indicate that not all AmeriCorps participant information was accurately reflected in the Trust database. An estimated 5 percent (8,300) of about 158,000 enrollments from program years 2000 to 2002 have discrepancies, and about 3 percent (4,400) have discrepancies that could affect estimates of future probable expenditures of the Trust. Further, the users’ manual for the Trust database system had not been updated. In 2003 the Corporation began using a new model with conservative assumptions of participant behavior to develop its funding estimates. Corporation officials explained that they used conservative assumptions because the AmeriCorps program does not have a long history from which to extrapolate participant behavior, and the Corporation wanted to regain credibility after the enrollment suspension in 2002. Using the new model may be prudent until the Corporation gains more experience. However, because the new model increased the Trust’s funding estimates, the Corporation will need to monitor actual experience compared with the model’s assumptions and may need to deobligate unused Trust funds. Further, the new model does not incorporate external factors, such as downturns in the economy, which may affect funding estimates or the Trust’s balances. The Corporation recently formed a team to assess the costs and benefits of adding external factors in its model. The Corporation has made some changes to its operations that minimize the likelihood it will need to suspend enrollments in the future. Corporation officials have been obligating Trust funds when positions are approved since June 2003, and the communication and coordination among officials have greatly improved. Changes have also been implemented and planned to address the Strengthen AmeriCorps Program Act requirements. However, changes to improve oversight of grantees have not been fully implemented, and policies related to refilling vacated positions and converting unfilled positions may limit enrollments, hinder service delivery, and contribute to the accumulation of a larger Trust balance. What GAO Recommends: GAO recommends that the Corporation make improvements to strengthen internal control, enhance the accuracy of its budget estimates, ensure the Trust does not accumulate large balances, and ensure that its policies support efforts to deliver services. The Corporation generally agreed with GAO’s recommendations. www.gao.gov/cgi-bin/getrpt?GAO-04-225. To view the full product, including the scope and methodology, click on the link above. FOR MORE INFORMATION, CONTACT CORNELIA M. ASHBY, (202) 512-8403 (ASHBYC@GAO.GOV) OR SUSAN POLING (202) 512-2667. [End of section] Contents: Letter: Results in Brief: Background: Trust Database and Participant Documentation Discrepancies Could Affect Estimates of Probable Education Award Expenditures: The New Model Increased Funding Estimates and Does Not Consider External Factors: Operational Changes Should Reduce the Risk of Enrollment Suspensions, but Two New Policies May Hinder Service Delivery and May Contribute to Higher Balances in the Trust: Conclusion: Recommendations: Agency Comments and Our Evaluation: Appendix I: Scope and Methodology: Appendix II: The Corporation's Antideficiency Act Violation: Appendix III: Discrepancies between Trust Data and Participant Documentation That Could Affect the Amount Owed by the Trust: Appendix IV: The Service Award Liability Model: Appendix V: Comments from the Corporation for National and Community Service: Appendix VI: GAO Contacts and Acknowledgments: GAO Contacts: Staff Acknowledgments: Tables: Table 1: Estimates of Documentation Availability for AmeriCorps State and National Grantee and VISTA Participants, Program Years 2000-2002: Table 2: Description of Discrepancies for AmeriCorps State and National Cases: Table 3: Description of Discrepancies for AmeriCorps VISTA cases: Figures: Figure 1: Flow of Funds and Information for AmeriCorps Education Awards: Figure 2: AmeriCorps Enrollments by Program Year: Figure 3: Estimate of Discrepancies between Trust Data and Participant Documentation: Figure 4: Percentage of AmeriCorps Participants Who Earned Education Awards Compared with the Percentages used in the SAL and the New Model: Figure 5: Points in AmeriCorps Program Cycle when Deobligation May Occur: Figure 6: The SAL Model: Abbreviations: NCCC: National Civilian Community Corps: OIG: Office of Inspector General: SAL: Service Award Liability: SSA: Social Security Administration: SSN: Social Security number: VISTA: Volunteers in Service to America: United States General Accounting Office: Washington, DC 20548: January 16, 2004: The Honorable Christopher S. Bond: Chairman, Subcommittee on VA, HUD, and Independent Agencies: Committee on Appropriations: United States Senate: The Honorable Charles E. Grassley: United States Senate: The Corporation for National and Community Service (the Corporation) was created in 1993 to help meet community needs in education, the environment, and public safety through activities such as tutoring and mentoring youth, building affordable housing, cleaning parks and streams, and helping communities respond to disasters. The Corporation also helps to expand educational opportunity by providing national service participants who complete a term of service with funds--$4,725 for a full-time participant in 2002--to help pay for their education. AmeriCorps is one of three national service programs the Corporation oversees. The Corporation receives appropriations to fund program operations and the National Service Trust (the Trust), which provides money to pay education awards. In fiscal year 2003 the Corporation was appropriated about $400 million to support AmeriCorps--about $300 million to support program operations and $100 million for the Trust. From November 2002 to March 2003, the Corporation suspended enrollments in AmeriCorps because it concluded that there would not be sufficient funds in the Trust to pay the education awards for all of the 2002 program positions. In April of 2003 we provided a statement for the record to the Subcommittee on Veterans Affairs, Housing and Urban Development, and Independent Agencies of the Senate Committee on Appropriations describing our preliminary observations regarding the causes for the enrollment suspension and the proposed policy changes.[Footnote 1] In that statement, we emphasized that had the Corporation appropriately recorded and tracked its obligations for education awards to program participants, the Corporation likely would not have needed to suspend enrollments. In July 2003, Congress passed the Strengthen AmeriCorps Program Act (Pub. L.No. 108-45). This act requires the Corporation to obligate funds in the Trust at the time AmeriCorps positions are approved and to estimate the value of the education awards based on a formula that considers historical rates of enrollment in the program and for the earning and use of education awards. The act also established several other requirements for the Corporation. Because of concerns about the Corporation's management of AmeriCorps and the Trust, you asked us to answer the following questions: (1) Has all AmeriCorps participant information been accurately recorded in the Trust database? (2) How does the Corporation estimate the funding needed to provide education awards through the Trust? (3) Has the Corporation made management and operational changes that ensure enrollments will not be suspended in the future and that address the Strengthen AmeriCorps Program Act requirements? To answer these questions we used a multifaceted methodology. We performed a detailed reliability assessment of the Trust database by comparing information in the database with original enrollment and exit forms. We obtained enough documentation to compare key information on 363 of 400 cases in our sample, or an estimated 157,045 of the 172,434 enrollments in the 2000 to 2002 program years. We examined the model used by the Trust to estimate its funding requirements and reviewed reports of auditors and contractors related to the model. We surveyed all 148 AmeriCorps grantees for their views on the Corporation's new enrollment and oversight policies as well as the training and technical support provided by the Corporation, and obtained a response rate of 71 percent. We also reviewed applicable laws, analyzed Corporation data, reviewed relevant documents and reports from the Corporation's Inspector General, and interviewed knowledgeable officials. We conducted our work between March and December 2003 in accordance with generally accepted government auditing standards. For more details about our scope and methodology, see appendix I. Results in Brief: Discrepancies between information in the Trust database and participant documentation indicate that not all AmeriCorps enrollment and exit information has been accurately recorded in the Trust database. We estimate that there are discrepancies in about 5 percent (8,300) of 158,000 enrollments from program years 2000 to 2002, and of these, about 3 percent (4,400) could affect estimates of future probable expenditures of the Trust. One of the most frequently occurring discrepancies pertained to cases in which the Trust database showed participants were still serving while their documentation showed they had exited the program without earning an award. For example, the database, as of July 25, 2003, indicated that a participant was still serving and the documentation showed that this participant had completed service on August 24, 2002. Our analysis also found that more than 300 of the participants enrolled between 1999 and 2002 had Social Security numbers that were invalid or had not been issued and about 170 had numbers for persons listed as deceased in the Social Security death master file. In addition to the data discrepancies, we found the documentation for the Trust database difficult and cumbersome to use. Corporation managers told us the users' manual had not been updated since it was first prepared in 1995, although there have been many changes to the system. Documenting systems is an important internal control that helps organizations ensure that data are reliably collected and properly used and helps ensure organizations are positioned to continue operations in the event of a disaster or emergency. The Corporation used one model for several years to estimate the funding needed to provide education awards through the Trust, and in 2003 it developed and used a new model that increased the funding estimates of the Trust. According to Corporation officials, the Corporation used the Service Award Liability (SAL) model from 1996 to 2003 to estimate the future probable expenditures of the Trust for past and current participants, as well as to develop funding estimates for future participants that it used in its budget requests. Following the passage of the Strengthen AmeriCorps Program Act in July 2003, the Corporation developed and used a new model to estimate the funding needed for future participants. This new model used more conservative values as compared with the previous model. Corporation officials stated that they believe their historical data provide a sound basis for Trust funding estimates. However, they chose to use more conservative assumptions because the AmeriCorps program does not have a long history and they wanted to regain credibility after having had to suspend enrollments last year. As a result, the new model increased the funding estimates. For example, the SAL model generated a requirement of about $116 million for fiscal year 2004 for 75,000 participants, and 7,000 scholarships for high school students, while the new model generated funding estimates of about $133 million for the same number of participants and scholarship recipients.[Footnote 2] Our analysis indicates that by using the new model's assumptions, the Trust may accumulate larger balances than if historical rates were used. This accumulation in the balance may occur in addition to the reserve account that is required by the Strengthen AmeriCorps Program Act. While Corporation officials acknowledged that these changes may cause larger balances in the Trust, they said they would be able to deobligate funds at several points in the AmeriCorps program life cycle and could adjust the assumptions in the future. While there may be large balances in the Trust, there are external factors, such as the strength of the economy and the support for volunteerism, that could either reduce these balances or further increase them. The Corporation has reported that external factors have affected its programs, but has not incorporated into its model the possible effects these factors could have on its estimates. The Corporation recently formed a management improvement team to assess the costs and benefits of incorporating external factors into its model. Building in consideration of external factors could further improve the quality of the Corporation's estimates. The Corporation has made changes that minimize the likelihood of a need to suspend enrollments in the future, and changes have been implemented and planned to address the Strengthen AmeriCorps Program Act requirements. However, some operational changes have not been fully implemented, and two new policies are of concern to grantees. Prior to the suspension of enrollments, the Corporation did not record an obligation to the Trust fund at the time it created positions, officials did not communicate regularly about the number of positions created or the number of participants enrolled in AmeriCorps relative to Trust resources, and the Corporation allowed grantees various flexibilities concerning the enrollment of participants. Beginning in July 2003, Corporation officials began obligating Trust funds when they approved positions in accordance with the Strengthen AmeriCorps Program Act. Corporation officials have placed greater emphasis on monitoring the availability of funds in the Trust, and the communication and coordination between officials responsible for managing the Trust and officials responsible for creating positions have greatly improved. Since March 2003, Corporation managers have been tracking AmeriCorps enrollments on a biweekly basis, investigating discrepancies in counts between its enrollment system and the Trust database, and monitoring enrollments in relation to the availability of funds in the Trust. The Corporation also tightened grantee program rules and now requires grantees to provide timely enrollment information. For example, grantees can no longer enroll more participants than are specified in their grant award, and grantees are now required to enroll participants within 30 days of their starting work. The Corporation has plans to improve the process for monitoring enrollments, to improve grantee oversight, and to meet the requirements of the Strengthen the AmeriCorps Program Act. However, two new policies are of concern to grantees because they believe these policies may limit enrollments and hinder service delivery. If these policies reduce enrollments, balances in the Trust may be further increased. Under current policy, if a participant enrolls in a program and drops out before earning an award, the grantee cannot offer the position to another applicant for the balance of the term of service. Grantees are also prohibited from exchanging a full-time position for an equivalent value of part-time positions. More than 80 percent of the respondents to our survey stated that the policy prohibiting refilling positions will have a negative effect on the program operations of their subgrantees, and 75 percent of the respondents stated that the policy prohibiting them from converting positions would hurt their ability to provide services. Also, these policies could hinder the Corporation's ability to fulfill its mission and reduce the number of participants and the number of earned education awards, thereby contributing further to the balance of funds in the Trust. We are recommending that the Corporation's chief executive officer review and document the Corporation's data assurance processes, update and improve Trust database system documentation, incorporate external factors into its Trust estimates, and assess enrollment policies to determine whether they have negatively affected service delivery and contributed to the balance of funds in the Trust. The Corporation's chief executive officer provided written comments on a draft of this report and generally agreed with our recommendations. Additionally, he suggested several changes to help clarify the report, which we incorporated as appropriate. These comments are discussed in the report and are shown in appendix V. Background: The Corporation is part of the USA Freedom Corps, a White House initiative to foster a culture of citizenship, service, and responsibility, and help all Americans answer the President's Call to Service.[Footnote 3] The Corporation maintains administrative field offices in almost every state. AmeriCorps was created in 1993 and is one of three national service programs the Corporation oversees: the Senior Corps, AmeriCorps, and Learn and Serve America. AmeriCorps consists of three programs: AmeriCorps State and National (state and national), AmeriCorps VISTA (Volunteers in Service to America), and AmeriCorps NCCC (National Civilian Community Corps). AmeriCorps programs for tribes and territories are included in state and national programs. The Corporation makes grants from its program appropriations to help grant recipients carry out national service programs. These include programs that tutor and mentor youth, build affordable housing, teach computer skills, clean parks and streams, run after-school activities, help communities respond to disasters, and those that are related to homeland security. About 60 percent of the Corporation's fiscal year 2003 grant funds for AmeriCorps programs went to state service commissions, tribes, and national direct grantees[Footnote 4] for AmeriCorps State and National programs, which award subgrants to nonprofit groups, which then enroll the AmeriCorps participants. The remaining grant and administrative funding paid for VISTA and NCCC participants--about 32 percent and 8 percent respectively, in 2003. AmeriCorps is open to U.S. citizens and nationals or lawful permanent resident aliens age 17 and older. Participants in the AmeriCorps program can receive stipends as well as health benefits and child care coverage. For example, about one-half of AmeriCorps participants received a $9,300 living allowance and health benefits in program year 2002. Those participants who successfully complete a required term of service earn a national service education award that can be used to pay tuition, fees, and expenses for undergraduate school, graduate school, or an approved school-to-work program, or to pay back qualified student loans. In exchange for a term of service, full-time AmeriCorps participants earned an education award of $4,725 in program year 2002. VISTA participants can elect to receive a cash stipend instead of an education award. About one-third of VISTA participants chose to take the stipend rather than the education award in 2002. Figure 1 illustrates the flow of the funds and information for the AmeriCorps program. Figure 1: Flow of Funds and Information for AmeriCorps Education Awards: [See PDF for image] [End of figure] According to the Corporation's Trust database, AmeriCorps enrollments more than doubled between 1994 and 2000. Overall AmeriCorps enrollments increased from about 25,000 in 1994 to about 42,000 in 1999, to over 59,000 in 2001. Enrollments fell in 2002 because of the suspension of enrollments and were legislatively limited to 50,000 in 2003. The Corporation requested Trust funding to support 75,000 new enrollments for 2004. AmeriCorps enrolls participants on a full-time and a part-time basis. Part-time participants who serve 900 hours or less annually earn education awards proportional to the hours served. During the first three years when AmeriCorps programs enrolled participants, 60 percent or more of the participants served in full-time positions. By 2000 less than 50 percent of the participants were full-time, and Corporation officials stated that they are planning for about a 50/50 full-time/ part-time mix in 2004. Trends in total enrollment from 1994 to 2002, and projected enrollment for 2003, and the full-time and part-time mix of participants for each year are shown in figure 2. Figure 2: AmeriCorps Enrollments by Program Year: [See PDF for image] Note: Program year 2003 figures are planned positions. [End of figure] Additionally, AmeriCorps enrolls "education award only" participants. AmeriCorps does not pay these participants a living allowance or other benefits, but it provides funding to grantees for administrative purposes only, about $400 annually per participant. However, each education award participant receives an education award equivalent to that earned by a paid AmeriCorps member. More than half of the AmeriCorps enrollment growth has come from grants that provide participants no benefits other than the education award. The Corporation reports there has been high interest in these grants in recent years. Enrollments for these positions increased from fewer than 6,500 in 1999 to almost 16,000 in 2001. All participants--full-time, part-time, and education award--have up to 7 years after they successfully complete their service to use their education awards.[Footnote 5] AmeriCorps also distributes funds to high school students through Presidential Freedom Scholarships. These scholarships are valued at $500 for each participant. For 2003 and 2004, the Corporation planned to award about 7,000 Presidential Freedom Scholarships. The Government Corporation Control Act requires the Corporation to have an annual audit of its financial statements.[Footnote 6] The Corporation prepares its financial statements using generally accepted accounting principles that are used by private sector corporations and federal corporations. The Corporation plans to include a statement of net cost and a statement of budgetary resources as supplemental information in its annual Performance and Accountability Report by fiscal year 2005. The Corporation's financial statement auditor reported that in fiscal year 2002 the Corporation approved AmeriCorps national service positions in excess of the number of positions that the Trust could support. In November 2002 the Corporation suspended enrollments in AmeriCorps. Several factors contributed to the need to suspend enrollments, including a lack of communication among staff responsible for program and Trust operations. The Office of the Inspector General (OIG) reported in April 2003 that the AmeriCorps program approved grants for thousands of positions more than were used in the Corporation's model to estimate the funding needs of the National Service Trust. On July 24, 2003, the OIG reported the Corporation had enrolled more AmeriCorps participants than the Corporation's National Service Trust could support, and as a result, beginning in 2000, the Trust's liabilities exceeded the funds it had available from appropriations and interest earnings.[Footnote 7] In that report, the OIG concluded that the Corporation had violated the Antideficiency Act. The Antideficiency Act prohibits an employee or officer of the United States government from making or authorizing an expenditure or obligation exceeding an amount available in an appropriation.[Footnote 8] The Corporation submitted an Antideficiency Act report to OMB, but as of December 2003, the report, including the amount of the deficiency, had not been submitted to the President and the Congress.[Footnote 9] For more information see appendix II. Trust Database and Participant Documentation Discrepancies Could Affect Estimates of Probable Education Award Expenditures: Discrepancies between the information in the Trust database and participant documentation indicate that not all AmeriCorps enrollment and exit information has been accurately recorded in the Trust database.[Footnote 10] Some of these discrepancies could affect the estimated probable expenditures of the Trust because these data are used to estimate the amount of education awards the Corporation will ultimately pay. Furthermore, the Corporation does not have complete or current user documentation for the Trust database. Without clear documentation explaining the components of the database, the Corporation may be at risk of not being able to properly interpret and analyze its participant data. Trust Database and Participant Documentation Discrepancies Could Affect Estimates of How Much the Corporation Owes in Education Awards: Out of about 172,000 enrollments in program years 2000 to 2002, we obtained sufficient documentation to estimate the extent of discrepancies for about 158,000 enrollments.[Footnote 11] We estimate that about 5 percent (8,300 enrollments) have a discrepancy between the Trust database and participant documentation.[Footnote 12] Of these, about 3 percent (4,400 enrollments) could affect estimates of future probable expenditures of the Trust.[Footnote 13] Figure 3 illustrates these results. Figure 3: Estimate of Discrepancies between Trust Data and Participant Documentation: [See PDF for image] [End of figure] Several of the discrepancies that could affect what is owed by the Trust were similar in nature. One of the most frequently occurring discrepancies pertained to cases in which the Trust database showed participants were still serving while their documentation showed they had exited the program without earning an award. For several other cases, the Trust data showed VISTA participants were still serving while the documentation showed they had elected to receive a cash stipend instead of an education award. According to an AmeriCorps official, cash stipends are not paid from funds in the Trust. In several other cases, the Trust data showed participants were still serving while their documentation showed they had signed up for a shorter term of service and should have exited the program. The Corporation acknowledged it would need to take further action to determine the actual status of these cases. The specific discrepancies we found that could affect the amount owed by the Trust are described in appendix III. Nearly all the discrepancies that will not affect the amount owed by the Trust were for cases without an enrollment date on the form, although an enrollment date was in the database. These cases do not affect the estimate of what the Trust owes because other information in the documentation was consistent with the database, such as whether or not the participant earned an award. Corporation officials told us an enrollment date should be recorded on the forms and that they would issue clarifying instructions to grantees. According to Corporation officials, there are numerous processes designed to help ensure the accuracy and validity of its data. However, the results of our analysis raise questions about the effectiveness of some of these processes. The Corporation has controls over who may enter or change participant information in the Web-Based Reporting System (WBRS), and several edit and data checks are in place to ensure data is entered properly and completely. For example, only a grantee's program director can certify that participants have completed their service and qualified for an award, and WBRS will not allow grantees to enroll more participants than the number of slots they were awarded. Corporation officials also stated that a number of data checks are performed when WBRS data are transferred to the Trust database. For example, after each weekly upload, an error report is generated showing such things as participants who had not been officially enrolled in the Trust and duplicate enrollments. Furthermore, the Corporation's fiscal year 2001 Performance and Accountability Report noted that the Corporation had instituted a procedure to randomly sample and verify enrollment and exit data on an annual basis. However, Corporation officials could not provide any reports documenting the results of these reviews. Our analysis of the Trust data also found that out of 186,000 participants enrolled during program years 1999 to 2002, 313 participants had Social Security numbers (SSNs) that were invalid or that the Social Security Administration (SSA) had not issued, and 169 had SSNs for persons listed in SSA's death master file. In about 75 percent of these last cases, the name and birthday of the individual in the SSA death file differed from the information in the Trust database. In the remaining 25 percent of cases, the name and birthday matched, but either the SSA records show the individual had died more than 30 days before the service completion date shown in the Trust database or no completion date was in the Trust database, indicating these participants may be listed as still serving. Corporation officials told us that some of these cases could be the result of data entry errors. They also said that, on occasion, grantee officials create records for nonexistent individuals, including mock SSNs, to test how the data entry system works. However, these erroneous data are not routinely cleared from the database, according to Corporation officials. Since these cases are in the database, they are considered as enrolled participants. Such records for nonexistent individuals will cause the Corporation to overstate the estimated amount owed by the Trust for education awards. In November, 2001, the Corporation entered into an agreement with SSA to test the validity of the SSNs of newly enrolled participants. Corporation officials provided information showing that they completed a SSN match under this agreement in May 2002 and found that 2,910 participants out of about 58,000 that did not match SSAs records. Corporation officials stated that they were undertaking an internal review to resolve these discrepancies. The Corporation has not done any subsequent matches, but officials told us that because of our findings, they would reconsider this. Corporation officials also said that while the database could be improved, there are safeguards to prevent an unauthorized person from claiming an award. For example, the postsecondary institution the participant is attending must verify that he or she is a student there, and the award funds are sent directly to the institution. Nonetheless, since there are tens of thousands of different individuals joining AmeriCorps every year and given the concerns about how SSNs are used and protected, particularly in light of the rise in identify theft, it may be in the Corporation's best interest to regularly verify the accuracy of the SSNs.[Footnote 14] Without valid SSNs in the Trust database on its participants, the Corporation cannot be certain that it has accurate information on its participants and that all participants meet the eligibility criteria. The Corporation Has Not Updated Its Users' Manual for the Trust Database: We found the system documentation for the Trust database was difficult to use and, in some instances, out of date. As a result, we had to rely on oral testimony and e-mails provided by Corporation officials and the WBRS support contractor for information about the system. The users' manual for the database was prepared in 1995. However, Trust database managers told us that this manual does not reflect all system changes since that time.[Footnote 15] The Corporation has a data definition dictionary for the Trust database intended to describe the data fields used in the database and what information they represent. However, the document we obtained does not provide definitions or labels for the data fields. Without this information, we were not able to identify what the data fields represented, nor would any new users of the system be able to identify the data fields. In commenting on a draft of this report, Corporation officials stated that the system documentation, including the data definition dictionary, was formatted to facilitate use by system developers rather than laypersons. We were also only able to obtain a partial written inventory of edit and data checks used for the database. We had to rely on two Corporation employees knowledgeable about the system's components to explain the type of information in each of the data fields, identify the data fields used in determining education awards, and determine what related data and edit checks were used. Documenting how data systems are to be used is a common, and required, management practice.[Footnote 16] Without clear or up-to-date system documentation explaining the data elements in the Trust database, and procedures for validating the data, the Corporation may be at risk of using incorrect data for its estimates of future probable expenditures. Also, the Corporation would be dependent upon a few employees who are familiar with the system to produce reports or prepare analyses of the data. Documenting systems is an important internal control that helps organizations ensure that data are reliably collected and properly used and helps ensure organizations are positioned to continue operations in the event of a disaster or an emergency. The New Model Increased Funding Estimates and Does Not Consider External Factors: In 2003, the Corporation began using a new model to estimate the funding needed to provide future education awards through the Trust. This new model used conservative values that increased the Trust's funding estimates as compared with the previous model. Corporation officials believe the historical data they possess provide a sound basis for Trust funding estimates. However, they chose to use more conservative values because the AmeriCorps program does not have a long history and they wanted to regain credibility after having had to suspend enrollments in 2002. In addition, the Trust fund now includes a reserve account required by the Strengthen AmeriCorps Program Act. The Corporation's model also does not include a way to consider the possible effects that external factors could have on its estimates. If the Corporation does not ensure its funding estimates for future education awards are as reasonable and complete as possible, millions of federal dollars may accumulate in the Trust and not be available to help support this or other programs. The New Model Increased the Estimated Funding Needed for the Trust: The Corporation used one model, the SAL model, to estimate both the probable expenditures of the Trust for past and current participants and the funding needed to provide education awards for future participants. According to Corporation officials, the SAL model was used from 1996 to 2003. In 2003, the Corporation developed a new model and revised its method for developing the Trust's future funding estimates. The SAL model is still used to estimate probable education award expenditures for past and current participants. The reliability and supportability of the estimates produced by earlier versions of the SAL model have been examined by outside auditors on two separate occasions.[Footnote 17] The auditors asked to review and assess the Trust model determined that the model produced reliable estimates for the period examined. However, they suggested functional enhancements and provided model documentation to the Corporation. Other auditors also reviewed the model as part of the annual audits of the Corporation's financial statements.[Footnote 18] These auditors advised the Corporation that controls and checks on the model's data should be strengthened. They also recommended that automated techniques be periodically used to systematically review the model's database and that the Corporation should consider several minor changes to the model to enhance the reasonableness of its estimates.[Footnote 19] We also reviewed the SAL model and found that the key factors and assumptions used to develop the accounting estimates were generally reasonable. Furthermore, neither the Corporation's IG nor our assessment found that the SAL model was a key factor that contributed to the suspension of enrollments in 2002. For information on the structure and content of the SAL model, see Appendix IV. In the new model, the Corporation used more conservative values for the rates at which education awards are earned and used. The Strengthen AmeriCorps Program Act states that the Corporation shall use a formula for estimating Trust obligations that takes into consideration historical rates of relevant participant behavior. In considering historical rates, the Corporation officials commented that they believe the historical data they possess provides a sound basis for Trust funding estimates and having extra funds available in the Trust--as compared with those estimated by the SAL model--might be prudent, particularly since the AmeriCorps program does not have a long history from which to extrapolate trends in participant behavior--only one cycle has been completed. The officials also noted that the new model's assumptions do not differ significantly from the historical averages. For example, the rate at which participants earn awards in the former model is about 75 percent, while the new model uses a rate of 80 percent. Further, the officials said they chose to use more conservative values because they wanted to regain credibility after having had to suspend enrollments in 2002. These officials said that they wanted to ensure that the Corporation would have adequate funds in the Trust and avoid any possible need to suspend enrollments again. It may be appropriate to gain some experience with the new model and current participant behavior before adjusting the assumptions used. The Strengthen AmeriCorps Program Act also required the Corporation to consult with the Congressional Budget Office (CBO) on its model formula. CBO focused its analysis on the Corporation's discount rate and reported that the Corporation used a discount rate that is more conservative than the one it uses when calculating the costs related to proposed legislation. When the new model's assumptions are used, the Corporation's Trust funding estimates were greater for the same number of participants than if the values in the SAL model were used. We calculated that the cost to provide education awards for up to 75,000 AmeriCorps participants and 7,000 Presidential Freedom scholarship recipients using SAL model assumptions was about $116 million. Using the assumptions in the new model, the Corporation's funding estimate was about $133 million for the same number of participants and scholarship recipients. The total amount in the Trust also may be higher because the act required the Corporation to include a reserve account. Corporation officials told us the reserve account value of 10 percent of the funding estimate was a value reached through discussions with congressional staff, and this value could change in subsequent years. The New Model Assumptions May Increase Balances in the Trust: Our analysis indicates that by using the assumptions in the new model, the Trust may accumulate more funds than have been needed to pay estimated education awards in the past. For instance, the Corporation's data show the Trust funding estimates are $13 million and $17 million more, respectively, when 50,000 and 75,000 participants are assumed, and the new award levels and the award earning and usage rates are used than those used in the SAL model. Not only are the assumptions in the new model greater than those used in the SAL model, but the SAL model assumptions are higher than the actual rates in most years. For example, the percentage of AmeriCorps participants who earned education awards has fluctuated since the beginning of the program. The rate decreased from a high of about 75 percent in 1994 to 68 percent in 1998, and then increased to 73 percent in 2001. The rate used in the SAL model for earned awards was about 75 percent--the highest average percentage rate achieved in the history of the program.[Footnote 20] Figure 4 compares the actual rates at which education awards were earned with the rate used in the SAL model and the rate used in the new model. If future participant behavior reflects the behavior of most past participants, the Trust fund balance may increase more using the new model assumptions than it would using the estimates produced by the SAL model. It may be appropriate for the Corporation to gain some experience with the new model and current participant behavior before adjusting the assumptions used. Figure 4: Percentage of AmeriCorps Participants Who Earned Education Awards Compared with the Percentages used in the SAL and the New Model: [See PDF for image] [End of figure] The Corporation acknowledges that the new model may create Trust balances that are greater than if historical rates were used if future participant behavior mirrors historical behavior. The Corporation chief financial officer said that there are opportunities to periodically deobligate funds during the program cycle. Funds could be deobligated from the Trust if (1) the positions created in the grant awards are less than the full-time-equivalent number of the positions approved, (2) all positions are not filled, (3) participants drop out before earning an education award, and (4) participants who earn awards do not use them. Figure 5 illustrates the points when the Corporation can deobligate funds. Figure 5: Points in AmeriCorps Program Cycle when Deobligation May Occur: [See PDF for image] Note: Shaded boxes are points at which Trust funds may be deobligated. [End of figure] Corporation officials said any deobligated funds would become available for other enrollments. They also said these funds would be used to fund other awards in current and future program years, and would be considered as reductions in subsequent budget requests. Furthermore, Corporation officials stated that the annual financial audit will address whether the size of the reserve account and the assumptions used are prudent. As of December 2003, the Corporation had two studies under way examining AmeriCorps participants' attrition rates and their utilization of earned awards.[Footnote 21] These studies focus on the frequency with which awards are earned and used. Corporation officials said that these studies have provided information that aids them in understanding AmeriCorps attrition and award usage, but they have not resulted in recommendations for policy changes that could affect the assumptions used in the model. External Factors Not Considered in Developing Estimates: Although Corporation reports indicate that external factors have affected program participation levels, these factors are not included in either model, nor have the Corporation officials taken them into account when they submit their funding request for the Trust. Estimating models should account for factors external to the business or entity that can affect the reasonableness of the estimates. Several factors can affect the assumptions used by the Corporation to estimate its budget needs. These factors include the state of the economy, the cost of postsecondary education and the availability of financial aid, and the levels of interest in volunteerism. Neither of the Corporation's models takes into account any external factors that could change future trends when it creates its assumptions or calculates its final budget estimates. Corporation officials stated that the models are not sophisticated enough to account for all of these factors. However, the Corporation has formed a Management Improvement Team to examine the potential costs and benefits of upgrading the model to account for some external factors. External factors could affect the number of AmeriCorps participants, as well as the attrition rate and use of awards once enrolled. For example, if unemployment rates are high, more participants may be willing to enroll, since they could receive a stipend as well as future education awards. Additionally, if postsecondary education costs increase, more participants may be likely to ensure they earn, and afterward use, their education awards. The Corporation has acknowledged that external factors have prevented it from achieving program goals. In its fiscal year 1999 and 2000 Performance Reports, the Corporation said it believed the strong economy was partly to blame for the Corporation's not achieving its program year 1998 and 1999 enrollment goals. In its fiscal year 2002 Performance and Accountability Report, the Corporation said the high level of interest in volunteerism following September 11, 2001, and the President's Call to Service contributed to higher levels of AmeriCorps enrollments than were anticipated. Operational Changes Should Reduce the Risk of Enrollment Suspensions, but Two New Policies May Hinder Service Delivery and May Contribute to Higher Balances in the Trust: The Corporation has implemented and planned substantive changes that should minimize the risk of an enrollment suspension in the future, and it has met or plans to meet the requirements established by the Strengthen AmeriCorps Program Act. Our previous statement and testimony from the Corporation's Inspector General identified several factors that contributed to the conditions surrounding the November 2002 suspension in enrollments.[Footnote 22] The Corporation has made changes to address these factors. However, grantees raised concerns about two new policies because they may limit enrollments and hinder service delivery. If these policies reduce enrollments, balances in the Trust may be further increased. The Corporation Has Improved Aspects of Its Operations: The Corporation has made changes to address the three conditions that we reported contributed to the need to suspend enrollments in AmeriCorps.[Footnote 23] The Corporation had not recognized its obligation to fund participant education awards. There was a lack of communication among program, grants management, and Trust officials with regard to the number of AmeriCorps positions the Trust could support. Finally, because they did not require grantees to provide timely enrollment information, Corporation and AmeriCorps managers could not be certain about the number of AmeriCorps participants and their effect on the Trust. The Corporation's obligation practices comply with the Strengthen AmeriCorps Program Act. In accordance with this act, the Corporation began obligating Trust funds when it approved positions in AmeriCorps grants beginning in June 2003. As of September 2003, the Corporation had obligated fiscal year 2003 appropriated Trust funds for about 46,000 of the 50,000 AmeriCorps positions. Greater emphasis has been placed on monitoring the availability of funds in the Trust, and the communication and coordination between officials responsible for managing the Trust and officials responsible for creating positions have greatly improved. Since March of 2003, the Corporation has been tracking AmeriCorps enrollments on a biweekly basis, investigating discrepancies in counts between its enrollment system and the Trust database, and monitoring the enrollments against the capacity of the Trust. These results are made available to high- ranking program and financial management officials of the Corporation, who meet monthly to discuss the results of these exercises. The Corporation has also tightened the controls and eliminated much of the flexibility it previously gave grantees. Grantees are now prohibited from enrolling more participants than specified in their grant awards, and the Corporation modified WBRS to prevent grantees from enrolling more participants than the number of positions contained in the grant award. To better monitor progress toward its enrollment goals and their effect on the Trust, the Corporation now requires grantees to report certain data about potential participants to the Corporation prior to their actual enrollment. If a potential participant does not enroll within 45 days of the expected start date, the file on the potential participant is deleted. If the individual enrolls as scheduled, the grantee must complete the enrollment process within 30 days of the participant beginning work. The Corporation added a tool to WBRS that allows managers to monitor the average number of days between a participant's start date and enrollment. According to the Corporation, 60 percent of program year 2002-2003 grantees have improved the average time between participants' start dates and the reporting of their enrollment compared to their performance in program year 2001-2002. However, less than 50 percent of the program year 2002- 2003 grantees reported an average time from start date to enrollment of 30 days, or less. In addition, the Corporation provided training and technical support to assist grantees with the system changes and new enrollment requirements. Most of the AmeriCorps grantees who responded to our survey said the training was adequate to meet the new requirements. Of the AmeriCorps grantees who responded, 80 percent said that the Corporation provided enough training and technical support to help them meet these new requirements. However, in written responses some grantees said they would like to receive training more frequently and believed it would be beneficial if the Corporation could provide training directly to subgrantees, rather than just to grantees. Additionally, about 7 percent of the responding grantees said that as a result of the new requirements, they or their subgrantees would have to perform additional tasks in their enrollment procedures. One of the responding grantees said that because it had over 150 part-time positions, and only limited staff, the requirement to enter data on participants twice imposed a significant burden. Some Changes Have Not Been Fully Implemented: Corporation officials stated that they are planning to improve oversight of grantees' performance. The Corporation plans to create consolidated reports in WBRS to facilitate oversight of the performance of its grantees, such as the state commissions and national direct grantees. This will allow them to identify programs having enrollment reporting problems more quickly and allow them to focus their oversight on those most in need of attention. Additionally, the Corporation plans to strengthen grantee oversight requirements to ensure that grantees are overseeing their subgrantees' compliance with enrollment procedures and time frames. The Corporation updated the administrative standards for the state commissions' process for monitoring their subgrantees' compliance with the enrollment procedures. These standards also include the expectation that state commissions will consider a subgrantee's compliance with the enrollment procedures and time frames in their funding decisions. The Corporation issued this guidance to the commissions in November 2003. In commenting on the draft report, Corporation officials said that the revised standards will be piloted in spring 2004 and then submitted to OMB for review in early fiscal year 2005. The Corporation also plans to establish a schedule for its staff to review grantees' enrollment cycle times and provide additional training and technical assistance to grantees. There are two provisions in the Strengthen AmeriCorps Program Act that increase the oversight of Trust operations. The first provision requires that the Corporation's chief executive officer (CEO) certify annually in a report to Congress that the Corporation is in compliance with other sections of the act. The second provision requires an annual audit of the accounts and records supporting the national service positions, and the National Service Trust estimate to fund those positions--referred to in the act as the Corporation's trust obligations. The act requires the CEO to include this annual audit with the CEO certification report forwarded to Congress. Although the audit has been completed, the Corporation has not yet provided its report to Congress.[Footnote 24] The auditor found, after accounting for the model's assumptions and reserve account, the Trust still had about $10 million of its fiscal year 2003 appropriation available for awards as of September 30, 2003. The acting CEO certified the Corporation's compliance with the act in the management representation letter provided to the Corporation's Inspector General on November 13, 2003. Corporation officials said they also plan to include the CEO certification and the audit of the estimated obligations in its 2003 Performance and Accountability report to Congress. Two Policy Changes May Limit Enrollments and Contribute to Balances in the Trust: Prior to the suspension in enrollments, the Corporation allowed grantees to replace a participant who left AmeriCorps before earning a full award and to convert an unfilled position to a different number of positions with an equivalent value of education awards. For example, if a grantee had a difficult time recruiting full-time participants, it could convert the full-time position into two half-time positions. In an effort to help ensure the Trust would not incur education award expenditures greater than its funds, the Corporation established policies that prohibit grantees from replacing participants and converting full-time positions. With these policies, if a participant enrolls in a program but then leaves after 1 week, the grantee cannot replace that person, and if a grantee has one full-time position but can only find people willing to work part-time, the grantee cannot convert that position to two half-time positions but instead can only fill it with one part-time person. The Corporation established these policies because it estimated that the Trust could only support a finite number of 2002 positions. Subsequently, Congress placed a 50,000 limit on the number of total AmeriCorps enrollments that could be filled with the 2003 Trust appropriations. However, as of November 2003, Congress has not included an enrollment limit in the proposed 2004 Trust appropriations. The policies prohibiting grantees from refilling vacated positions and converting unfilled positions helped the Corporation officials to control enrollments and helped them to comply with the legislatively imposed limit, which in turn helped ensure the solvency of the Trust. However, grantees who responded to our survey said that these policies may hinder their ability to provide services. Additionally, these policies may lead to fewer enrollments and coupled with the Corporation's obligation practices and the model assumptions may contribute to a higher balance of funds in the Trust. Our survey of AmeriCorps grantees shows that of those grantees that responded, 80 percent reported that the prohibition regarding the refilling of positions will have a generally or very negative effect on their program operations. Three-quarters of respondents reported that this policy will force changes in their program operations or those of their subgrantees or operating sites. Eighty-one percent reported that that it will limit enrollments and hinder their ability to provide services. Sixty-five percent reported that the policy will require changes to the services provided. Respondents reported that they may have to recruit more carefully, or not report recruitments until the latest possible time, to maximize the likelihood participants will remain in the program. This will affect AmeriCorps participants as well as those being served. One respondent in our survey commented, Our subgrantees must now rethink their member enrollment and termination policies. In the past, programs could "take a chance" on enrolling a potential member who showed promise but who may have also had potential risk factors. They were able to do this because they could refill the slot if the member did not work out. In relation to terminating members, subgrantees will now be somewhat hesitant to release a member who is not performing as expected, because they will not be able to refill the slot. Similarly, most of the respondents reported that the policy on converting unfilled positions will affect program operations and service delivery. About 75 percent of respondents reported that this restriction will negatively affect their operations and require them to change their operations. Seventy-five percent reported that it will hinder service delivery, and 60 percent reported that it will cause changes in services provided. Respondents commented that the authority to convert positions allowed them to respond positively to changing circumstances and better address community needs. One respondent noted that without being able to convert positions, applicants will not be accommodated if their availability does not conform to the slots provided in the grant. Positions, therefore, could go unfilled. Furthermore, if these policies result in fewer enrollments, the balance in the Trust may grow. As previously mentioned, since the passage of the Strengthen AmeriCorps Program Act, the Corporation obligates funds for all AmeriCorps positions when the grants are approved and prior to participants actually enrolling. Also, the new model the Corporation uses for developing funding estimates assumes 100 percent enrollment for all positions created. If enrollments do not reach the approved levels, and if Corporation officials do not regularly and diligently monitor enrollments and periodically deobligate funds, the funds in the Trust may accumulate. Internal Control Weaknesses Remain: Financial statement auditors have reported several internal control weaknesses or conditions at the Corporation and whether the Corporation has improved those previously identified. For example, the auditors reported that the grant approval policies and procedures were a serious weakness in fiscal years 1999 and 2000. The reports for 2001 and 2002 show that the Corporation improved in this area and the auditors did not list this as a concern in the 2003 report. However, the Corporation continues to have some internal control problems. The fiscal year 2003 audit reported a continuing internal control problem regarding the Corporation's monitoring of grantee activities. This problem was also cited in the audit reports for fiscal years 2001 and 2002. Additionally, as previously stated, we found that the Corporation has not fully implemented its efforts to improve oversight of grantees' performance, its procedures do not ensure accurate data in the Trust database, and the Corporation's ability to fully use the data on its participants may be limited because the users' manual for the database has not been kept current. In light of these internal control weaknesses and the concerns related to the problems that lead to the suspension in enrollments, having an auditor review internal control would provide a measure of assurance over the Corporation's accountability and internal control. Auditor opinions on internal control are a critical component of monitoring the effectiveness of an entity's risk management and accountability systems. When an auditor renders an opinion on internal control, the auditor is providing reasonable assurance that the entity has maintained effective internal control over financial reporting (including safeguarding of assests) and compliance such that material misstatements, losses, or noncompliance that are material to the financial statements would be detected in a timely fashion. The auditor also reports on any significant deficiencies or material weaknesses in the internal control over financial reporting. An opinion on internal control is appropriate and necessary for major government entities and in other cases where an opinion on internal control would add value and mitigate risk.[Footnote 25] Conclusion: The suspension of AmeriCorps enrollments had a serious impact on the Corporation's operations, resulting in both internal and external scrutiny. This scrutiny revealed many shortcomings with the Corporation's management of the AmeriCorps program and the Trust. Since the suspension in enrollments, the Corporation has made significant changes to some of its operating procedures and internal controls. Most notably, the Corporation began obligating Trust funds when AmeriCorps positions were created, and key Corporation officials have been much more focused on ensuring adequate funds are in the Trust. However, weaknesses still existing in the Corporation's procedures and internal control could negatively affect the Trust or hinder the Corporation's ability to fulfill its management responsibilities. For instance, discrepancies between the Trust database and the participant data indicate that the Corporation does not have adequate internal control procedures to ensure the accuracy of its data. Accurate and complete participant information in the Trust database is critical. This information plays a significant role in estimating the education awards owed by the Corporation and ensuring they are awarded properly. Inadequate system documentation is an internal control weakness that could limit the Corporation's ability to maximize its use and understanding of the data it possesses. Without valid SSNs in the Trust database on its participants, the Corporation cannot be certain that it has accurate information on its participants and could indicate that not everyone participating in the program meets the eligibility criteria. In addition, other changes are needed to enhance the Corporation's efforts to be good stewards of public funds while fulfilling its mission. Given the Corporation's relatively short history, and fluctuation in its program data, it may be prudent for the Corporation to use factors and value in its model that provide some additional funds in the Trust. Nonetheless, future funding estimates for the Trust should be as reasonable and complete as possible to minimize the accumulation of large balances in the Trust. The Corporation needs to balance its efforts to ensure the Trust does not assume future probable expenditures in excess of its funds with the mission and goals of the grantees and subgrantees that enroll AmeriCorps participants to help meet community needs. Policies that do not balance these goals may also contribute to the accumulation of balances in the Trust. If the Corporation does not ensure its funding estimates are as reasonable and complete as possible and does not regularly and diligently monitor enrollments and periodically deobligate funds, millions of federal dollars may accumulate in the Trust and not be available to help support other federal programs. Finally, obtaining an auditor's opinion on internal control over financial reporting as part of its annual financial statement audits would provide additional accountability and assurance. Recommendations: To improve the management of AmeriCorps and the National Service Trust, we recommend that the chief executive officer of the Corporation take the following nine actions: To ensure the Trust receives accurate data for use in its model estimates and Trust database, * implement a strategy to correct discrepancies between the Trust database and the enrollment and exit forms, * review and document the effectiveness of its data assurance processes, * regularly verify the accuracy of the SSNs of its participants. To better ensure that the Corporation has data that are readily available and is positioned to continue operations in the event of a disaster, emergency, or employee turnover, and: * update the users' manual for the Trust database and develop an inventory of edit and data checks used for the database. To provide additional assurance over internal control and to minimize the related risks, * obtain an auditor's opinion on the adequacy of the internal control over financial reporting as part of the annual financial statement audit. To enhance the accuracy of Trust budget estimates and ensure the Trust does not accumulate large balances, * create a means to take into account the possible impact that external factors may have on participant behavior in its funding estimates and budget requests, * establish and execute a periodic deobligation schedule for unused Trust obligations, and: * review the assumptions being used in the new funding model after the Corporation gains more experience with the new model and current participant behavior. To ensure its policies support its mission and grantees' efforts to deliver services while also providing adequate management controls, * evaluate the enrollment policies regarding refilling and converting positions. Agency Comments and Our Evaluation: We received written comments from the chief executive officer for the Corporation for National and Community Service. These comments are reprinted in appendix V. The chief executive officer agreed with eight of the nine recommendations and identified the actions planned to address them. As for the other recommendation--to obtain an auditor's opinion on the adequacy of the Corporation's internal control over financial reporting as part of the annual financial statement audit-- the chief executive officer stated that he would refer it to the Corporation's Office of Inspector General, since that office contracts for the annual financial audit. Additionally, the chief executive officer identified several areas that needed further clarification. He pointed out that the system documentation for the Trust module of eSPAN was not outdated, as we said in our draft, and informed us that the documentation is up to date but neither it nor the data dictionary is maintained in a laypersons' format. We modified the report to better reflect this information. The chief executive officer stated that the Corporation believes that the historical data currently used to estimate Trust funding provides a sound basis for the estimates. We modified the language in the report to reflect the Corporation's views. Further responding to the chief executive officer's comments, we (1) deleted the statement that the Corporation planned to begin using Generally Accepted Accounting Principles for federal government entities, (2) added program year information to better describe the comparison of the average time between participants' start dates and reporting of their enrollment, (3) made it clear that the Corporation's oversight efforts will focus on its grantees--the state commissions and national direct grantees, (4) updated the status of the Corporation's revised administrative standards, and (5) changed the date when the CEO certification was provided. Further, the chief executive officer expressed the view that our discussion of the Antideficiency Act violation focused on the amount of the deficiency at a specific time in the past and does not calculate the amounts deobligated over time. He also states that it is the Corporation's view that the deficiency is the amount needed at this time to liquidate obligations. We disagree and we revised this section of the report to explain that the Antideficiency Act requires an agency to report the amount of the violation at the time the violation occurred. The chief executive officer also stated that the $64 million deficiency appropriation should be sufficient. The report does not reflect nor did we calculate the amount the Corporation needs to liquidate its obligations. As we agreed with your office, unless you publicly announce the contents of this report earlier, we plan no further distribution of it until 30 days from the issuance date. We will then send copies to the Chief Executive Officer of the Corporation for National and Community Service and make copies available to others who request them. At that time, the report will also be made available at no charge on GAO's Web site at http://www.gao.gov. If you have any questions about this report, please call me on (202) 512-8403 or Carolyn Taylor on (202) 512-2974. Other contacts and staff acknowledgments are listed in appendix VI. Cornelia M. Ashby: Director, Education, Workforce and Income Security Issues: Signed by Cornelia M. Ashby: [End of section] Appendix I: Scope and Methodology: To determine if the activities of AmeriCorps participants were accurately recorded in the Trust database, we performed a data reliability test comparing information in the Trust database with information from original enrollment and exit forms completed by the participants and their AmeriCorps program managers. We obtained data files from the Web-Based Reporting System (WBRS) and the System for Programs, Agreements, and National Service Participants (SPAN). SPAN is the database used by the Trust. We performed general checks on these files to look for invalid data such as Social Security numbers that are invalid or have not been assigned. Because of a lack of documentation describing the data fields in detail, particularly for SPAN, we held several discussions with Corporation officials and contractor staff to resolve technical issues associated with the data files and to identify data fields to use in our reliability test. Based on these discussions, we developed a list of WBRS data fields that would document participants' enrollment, exit, and award eligibility. At our request, Corporation staff provided a list of corresponding SPAN data fields. The population we performed the reliability test for consisted of 172,434 state and national direct grant program and VISTA individual enrollments that were in the SPAN database as of July 25, 2003, and enrolled during the 2000, 2001, and 2002 program years. Each individual enrollment consisted of a discrete period of time a specific individual was an AmeriCorps participant at a single grantee. We did not sample National Civilian Conservation Corps (NCCC) participants because they constitute a small percentage of AmeriCorps enrollments (less than 3 percent). We drew a stratified random sample of 400 enrollments from this SPAN database. The population was stratified into four groups on the basis of program enrollment (i.e., state and national direct grant program or VISTA) and exit status (i.e., had exited from the program or not). With this statistically valid probability sample, each enrollment in the study population had a nonzero probability of being included, and that probability could be computed for any enrollment. Each sample element was subsequently weighted in the analysis to reflect the sample design. The weighting factors were computed as the ratio of the population to the sample within each stratum. We express confidence in the precision of our estimates as a 95 percent confidence interval.[Footnote 26] Corporation regulations require that participant documentation be kept on file for 3 years from the date a participant finishes his or her term of service.[Footnote 27] Documentation for VISTA participants is maintained at the Corporation's state field offices and for state and national participants the documentation is maintained at the grantee's or subgrantee's office. The Corporation sent requests for the participant documentation to the appropriate grantees for each enrollment in our sample. The grantees sent the documents to the GAO headquarters in Washington, DC. We reviewed the documentation to make sure all applicable forms were submitted. Because of a larger than expected number of nonresponses and incomplete document submissions, we made a second request to the grantees that had not provided complete documentation. In total we allowed program officials 6 weeks to provide the documents. Not all grantees provided us with complete participant documentation for the cases we randomly selected. We did not receive any of the documentation needed to assess whether enrollment information was accurately recorded for an estimated 14,055 of the 172,434 enrollments in the 2000 to 2002 program years. Although we received incomplete documents for an estimated 30,176 enrollments, we were able to compare key information with the SPAN data. Corporation officials expressed surprise that we did not receive all of the documents requested. Table 1 shows estimates of documentation availability for all enrollments during program years 2000-2002. Table 1: Estimates of Documentation Availability for AmeriCorps State and National Grantee and VISTA Participants, Program Years 2000-2002: Ninety-five percent confidence intervals shown in parentheses. AmeriCorps state and national; Number of enrollees: 157,045; No documentation available: 7.6% (4.3 - 12.4%); Incomplete documentation available: 15.9% (11 - 21.8%); Complete documentation available: 76.5% (70.5 - 82.5%). AmeriCorps VISTA; Number of enrollees: 15,388; No documentation available: 13.4% (8.5 - 19.6%); Incomplete documentation available: 34.3% (27.1 - 41.6%); Complete documentation available: 52.3% (44.7 - 59.9%). Total; Number of enrollees: 172,434; No documentation available: 8.2% (5 - 12.4%); Incomplete documentation available: 17.5% (13 - 22.8%); Complete documentation available: 74.3% (68.9 - 79.8%). Notes: Totals may not add because of rounding. [End of table] Grantee officials are responsible for entering and updating all information about participants. Participants complete enrollment and exit forms at the start and completion, respectively, of their terms of service. Grantee officials complete and sign other portions of these forms certifying the participant's enrollment and exit dates and whether or not the participant earned an education award. The grantee officials enter the participant's information into WBRS.[Footnote 28] It is then transferred to the Trust database on a weekly basis. Since participant data from the Trust database were used to calculate the Corporation's future probable expenditures of the Trust for past and current participants as well as develop its funding estimates for future participants that it uses in its annual budget requests, we focused our comparison on key fields that would affect the estimates. According to Corporation officials, these include the enrollment date, the number of hours the participant committed to serve, the number of hours actually served, and the participant's termination status-- whether or not the participant completed the term of service and earned an award. We recorded as a discrepancy any case where the information for these fields in the documentation did not match the information recorded in SPAN or the information appeared to be inconsistent with enrollment period rules such as participants still serving when other information indicates they should have finished their term. We assessed whether the discrepancies would affect what the Trust owed. We provided a list of the discrepancies to Corporation officials for further investigation. In several cases, they were able to explain why we found a discrepancy and document that the case was accurately recorded in the Trust database. We did not include these cases in our discrepancy count. To answer the question of how the Corporation develops its funding estimates for the Trust, we reviewed the models used by the Trust to develop its funding estimates. We obtained electronic versions of the models and input data to assess whether the models worked as described by Corporation officials. We gained an understanding of the controls over the preparation of the estimates and supporting data. We also learned about the controls over the sources of data and external factors that could affect the estimates. We verified the consistency of assumptions with historical data and the reliability of historical data. We considered whether changes in business or industry would cause other factors to become significant, and questioned Corporation officials regarding the possible impact of changes in Corporation operations on the factors. We also performed a test of calculations in the model. We obtained and reviewed revisions to the model recommended by auditors and contractors, and the changes planned for future budget submissions. We discussed all these issues with knowledgeable staff in the chief financial officer's office and conducted several hands-on sessions during which the Corporation staff demonstrated how the model worked and was used. Additionally, we obtained and reviewed financial statement reports from the Corporation's IG and assessed the sections of those reports that discussed the models and the supporting data. To determine if the management and operational changes to controls put in place following the suspension in enrollments in November 2002 would be effective, we reviewed Corporation memos and related data to check the progress of and compliance with these reforms and we interviewed knowledgeable staff. We also surveyed 148 AmeriCorps grantees--state commissions, national direct parent organizations, education award programs, tribes and territories--and asked their views on the new enrollment and oversight policies as well as the adequacy of the training and technical assistance provided by the Corporation. In developing the questionnaire, we reviewed memos implementing the new policies and met with Corporation officials to gain a better understanding of the new policies. We also obtained comments from Corporation officials on a draft of the questionnaire. We pretested the questionnaire in person or by telephone with the four grantee groups-- state commissions, national direct grantees, education award programs, tribes and territories. Guided by the pretest results and comments from the Corporation, we revised the questionnaire to ensure that all questions were fair, relevant, and easy to understand and answer. In addition, we tested the questionnaire to ensure that completing it would not place too great a time burden on grantees. Overall, we received responses from about 71 percent of the grantees (105 of the 148 surveyed). Response rates for each type of grantee were: state commissions (81 percent), national direct (73 percent), education award programs (60 percent), tribes and territories (54 percent). To establish whether the Corporation obligated funds in excess of the amount available in the Trust, we reviewed applicable statutes concerning how federal funds should be obligated and the 1993 Corporation legislation. We also reviewed reports related to the violation issued by the Corporation's OIG and its financial statement auditor. We clarified legal issues through correspondence with legal counsel from the Corporation and the Office of Management and Budget (OMB). We reviewed the internal control processes related to various Corporation activities using the Standards for Internal Control in the Federal Government.[Footnote 29] We conducted our work between March and December 2003 in accordance with generally accepted government auditing standards. [End of section] Appendix II: The Corporation's Antideficiency Act Violation: In July 2003, the Corporation's Office of the Inspector General (OIG) reported that the Corporation had obligated funds in excess of the amount available to it in the National Service Trust. In that report the OIG noted that the Corporation had suspended participant enrollment on November 15, 2002, because it was concerned that the Trust would not have sufficient funds to cover education awards. At the time, in November 2002, Corporation did not record education award obligations in the Trust Fund until it paid education awards to eligible participants, so it had no assurance that adequate funds were available. The Corporation amended AmeriCorps grants to suspend enrollments and did not permit any additional enrollments until Congress appropriated additional funds to the Trust in March 2003. However, the suspension came too late. In February 2003, the Corporation's financial statement auditor reported that, in fiscal year 2002, the Corporation had approved AmeriCorps national service positions in excess of the number of positions that the Trust could support.[Footnote 30] The July 2003 OIG report stated that the Corporation had enrolled more AmeriCorps participants than the Trust could support, and as a result, the "[t]rusts' liabilities, based on appropriations and interest forbearance[Footnote 31] expected to be paid, exceeded the Trust's appropriations and interest earnings beginning in 2000."[Footnote 32] The report also stated that the Corporation had violated the Antideficiency Act.[Footnote 33] Additionally, the OIG report stated that the Corporation estimated that the violations resulted in a Trust deficiency of approximately $64 million. This $64 million was derived from a reconstruction of the Trust's financial status that the Corporation prepared at the Office of Management and Budget's (OMB) request. In its July 25, 2003, response to the OIG's report, the Corporation conceded that it had violated the Antideficiency Act but disagreed with the amount of the violation reported by its OIG and stated that the financial reconstruction was a draft that was prepared "while the legal landscape was still unfolding."[Footnote 34] We suspect that the $64 million amount of the violation may be understated. In two legal opinions, in April and June 2003, we explained that the Corporation incurred, and was required to record, an obligation at the time it awarded a grant approving a new participant slot.[Footnote 35] We concluded that the Corporation, by waiting to record an obligation until it paid an education award, was under recording its obligations. In addition, we explained that the Corporation could not, without specific statutory authority, record an obligation on the basis of estimates of what it would have to pay when education awards are earned (its probable accounting liability). We stated that the Corporation should record its maximum potential liability. As our legal opinions established, the Corporation under recorded its obligations by recording (1) obligations at the time of drawdown rather than at grant award and (2) estimates of what it would have to pay rather than the full potential costs of the education awards. In 2002, OMB advised the Corporation that it should record obligations as the grantees enrolled new participants. OMB also advised the Corporation that it should record the amount of the obligation based on estimates of what enrolled participants will draw down in the future, using historical information, such as attrition rates and actual usage by participants who complete a term of service and earn an education award. The Corporation had already been developing these estimates for its budget request to the Congress. At the time of the enrollment suspension, while the Corporation had not yet instituted the practice of recording obligations as the grantees enrolled participants, it took into consideration the consequences of recording obligations at the time of enrollment in its financial reconstruction for OMB and concluded that if the Corporation had been recording obligations at the time of enrollment, the deficiency would be $279 million. We suspect that $279 million might also understate the amount of the violation at the time the violation occurred because the Corporation still would have recorded estimates, as opposed to the maximum potential liability, and still would have failed to record obligations when incurred---at time of grant award. The Corporation submitted an Antideficiency Act report to OMB, but as of December 2003, the report, including the amount of the deficiency, had not been submitted to the President and the Congress.[Footnote 36] In our June 6, 2003, legal opinion we noted that the Corporation could seek legislation that would permit it to use an estimation model for recording its obligations, and that this model could be similar to the process that would be used to determine the Corporation's probable accounting liability. [Footnote 37] On July 3, 2003, the Congress passed the Strengthen AmeriCorps Program Act,[Footnote 38] permitting the Corporation to record obligations based on estimates, as it had been doing without statutory authority prior to the passage of the act. The act redefined what an obligation is for purposes of the AmeriCorps program by authorizing the Corporation to "record as an obligation an estimate of the net present value of the national service educational award associated with the position, based on a formula that takes into consideration historical rates of enrollment in such a program, and of earning and using national service educational awards for such a program."[Footnote 39] With regard to when the Corporation incurs a liability, the act requires the Corporation to change its obligation practices by specifying that the Corporation obligate funds from the Trust at the time it awards a grant approving a new participant slot, rather than at time of enrollment.[Footnote 40] Nevertheless, the Corporation must report the amount of its Antideficiency Act violation based on the legal requirement in place at the time the violation occurred. According to the report by the Corporation's OIG, and conceded by the Corporation, the Corporation's violation occurred beginning in fiscal year 2000, before the Corporation obtained legal authority to alter its obligation practices. Had the Corporation reported its violation at the time it occurred, the Antideficiency Act would have required it to report a violation calculated in consideration of the cost of new participant slots approved at the time the Corporation awarded new grants, and not based on an estimate of what these participants would cost the Corporation in the future. The Corporation should calculate its Antideficiency Act violation for purposes of its report to the President and Congress based on the requirement in place at the time the Corporation incurred the overobligations. The Antideficiency Act requires an agency to report all relevant facts and a statement of actions taken.[Footnote 41] Accordingly, an agency, among other things, must inform the President and Congress if a violation occurred and in what amount, and request from the Congress, if needed, a deficiency appropriation. Even though an agency may cure a violation, it is still required to report the violation to the President and Congress. Because the Corporation's legal requirement for recording obligations differs today from the requirement that was in place when this violation occurred, the Corporation, in fact, may not need additional appropriations to cure the violation. Regardless, the Corporation should report the actual amount of the violation, but advise Congress of the changed circumstances. We have not assessed and do not address whether the Corporation will need additional funds to meet its current potential liabilities. To the extent that the Corporation identifies a need for additional appropriations as a result of its Antideficiency Act violation, the Corporation should include that request in its Antideficiency Act report. [End of section] Appendix III: Discrepancies between Trust Data and Participant Documentation That Could Affect the Amount Owed by the Trust: Table 2: Description of Discrepancies for AmeriCorps State and National Cases: State, national participants case number: 1; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. WBRS report provided in lieu of exit form indicates member was to complete service on 4-6-03. State, national participants case number: 2; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant completed term and earned full award. Award eligibility status was left blank on exit form. State, national participants case number: 3; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. Documentation provided shows participant completed service and earned award. State, national participants case number: 4; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. Enrollment form shows participant signed up for summer term on 8-15-02 with expected completion date of 8-14-03. State, national participants case number: 5; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. Enrollment form shows participant signed up for summer term on 8-15-01 with expected completion date of 8-14-02. State, national participants case number: 6; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. Enrollment form shows participant signed up for summer term on 8-15-02 with expected completion date of 8-14-03. Source: GAO analysis. [End of table] Table 3: Description of Discrepancies for AmeriCorps VISTA cases: VISTA participants case number: 1; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. Documentation shows participant exited on 3-18-02 as an early termination and was not eligible for an award. VISTA participants case number: 2; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. Exit form shows participant completed service as scheduled on 8-24-02 and is eligible for a full award. VISTA participants case number: 3; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. Documentation shows participant resigned on 3-21-03 and is not eligible for an award. VISTA participants case number: 4; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. Documentation shows participant resigned on 4-30-01 and is not eligible for an award. VISTA participants case number: 5; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. Documentation shows participant exited on 2-28-01. VISTA participants case number: 6; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant served 305 days with completion date of 10-9-00. Exit form shows completion date of 12- 9-00. (Enrolled 12-10-99.). VISTA participants case number: 7; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. Exit form shows participant exited early, 1-13-03, and is not eligible for award. VISTA participants case number: 8; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. Documentation shows enrollment period ended 2-4-03 and on 2-19-03, participant changed benefit to cash stipend. VISTA participants case number: 9; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant completed as scheduled and is eligible for award. Exit form shows participant not eligible for an award because cash stipend was chosen. VISTA participants case number: 10; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. Documentation shows participant changed benefit to cash stipend before completing term. VISTA participants case number: 11; Description of discrepancy (Trust data as of July 25, 2003): Trust data show participant still serving. Documentation shows participant changed benefit to cash stipend before completing term. Source: GAO analysis. [End of table] [End of section] Appendix IV: The Service Award Liability Model: The Service Award Liability (SAL) model used SPAN data on the actual behavior of past and current AmeriCorps participants as the basis for assumptions of future participant behavior. These data included the proportion of participants who earned awards, the percentage of earned awards used, the period when awards were used, and the size of the average award. Also included in the calculations were past and anticipated appropriations, net of rescissions, as well as the actual and expected interest earnings of the Trust balance in order to arrive at the estimates of the funding needed for future participants.[Footnote 42] The model used the portfolio of the Trust's investments--the proportion held as short-and medium-term treasury securities--to calculate a discount rate, which in turn is used to convert future funding needs into a net present value.[Footnote 43] Corporation officials would then adjust the resulting estimate to account for any accumulated funds in the Trust. Figure 5 shows how the SAL model works. Figure 6: The SAL Model: [See PDF for image] Note: Shaded boxes represent data used by the SAL model. [End of figure] [End of section] Appendix V: Comments from the Corporation for National and Community Service: CORPORATION FOR NATIONAL & COMMUNITY SERVICE: January 7, 2004: 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: Thank you for the opportunity to review and comment on the General Accounting Office's (GAO) draft report on the National Service Trust (GAO-04-225). As I begin my tenure as Chief Executive Officer at the Corporation, I am pleased that the issues that gave rise to the pause in enrollments have been resolved and that appropriate internal controls have been put in place to prevent a reoccurrence in the future. The new procedures reflect the hard work by Corporation staff to improve operations and implement changes based on valuable guidance from Congress and the Administration. I also appreciate GAO's concern that the improved procedures may result in the National Service Trust accumulating more funds than are needed to pay education awards. However, as noted in the report, the Corporation's new procedures include steps to monitor Trust activity and to deobligate funds for AmeriCorps positions not utilized by grantees. In addition, the adequacy of the obligated and reserve balances in the Trust will be assessed on an annual basis as part of the audit required under the Strengthen AmeriCorps Program Act. These procedures provide ample opportunity to adjust future appropriations requests to prevent significant surpluses from accumulating in the Trust. Corporation staff has also reviewed the report. The Corporation generally agrees with the information presented in the report and is in the process of implementing its recommendations. However, several areas need further clarification. These matters, as well as a brief response to each recommendation, are presented on the following pages. Signed by: David Eisner: Chief Executive Officer: Enclosure: Corporation Response to Draft General Accounting Office Report: Better Internal Controls and Revised Practices Would Improve the Management of AmeriCorps and the National Service Trust (GAO-04-225, Dated December 18, 2003): CLARIFICATIONS: In the Results in Brief section of the draft report (as well as in the body of the report) GAO states that the system documentation for the Trust module of eSPAN is outdated and has not been updated since 1994. This is incorrect. eSPAN system documentation: is maintained in the Oracle Designer repository and is up to date. Since this is system level documentation, it is not maintained in a layperson's format but in a format that is advantageous to the work of the developers. This also holds true for the data dictionary. The Corporation does agree that the user manual for the Trust module needs to be updated. In the Results in Brief section of the report (as well as in the body of the report) GAO states that the Corporation used more conservative assumptions in the new model because the AmeriCorps program does not have a sufficient history from which to extrapolate participant behavior, and the Corporation wanted to regain credibility after having to suspend enrollments. This is incorrect. The Corporation believes that, although trend data will improve over time, the historical data currently used to estimate Trust funding needs does provide a sound basis for the estimates. In addition, while the current assumptions used to estimate Trust obligations are slightly more conservative than those used in the past, they are consistent with historical trends as well as with the assumptions recommended by Senator Bond in his floor statement accompanying passage of the Strengthen AmeriCorps Program Act. On page 10 of the draft report, GAO states that the Corporation plans to begin using Generally Accepted Accounting Principles (GAAP) standards for federal government entities by 2005. However, the Corporation's financial statements are currently prepared in accordance with policy established by the Federal Accounting Standards Advisory Board (FASAB) for government corporations. As such, no changes to the basic financial statements are contemplated absent new guidance from FASAB. However, the Corporation does plan to include a statement of net cost and a statement of budgetary resources as supplemental information in its annual Performance and Accountability Report by fiscal 2005. On page 23 of the draft report, GAO states, "According to the Corporation, over the last year, while 60 percent of all grantees have improved their average time between participants' start dates and the reporting of their enrollment, less than 50 percent of grantees had reported an average time from start date to enrollment of 30 days or less." To clarify, this information is based on a comparison of grantees at a point in time during the 2002-03 program year (which corresponds generally to the 2003 fiscal year) to their performance in the 2001-02 program year (which corresponds generally to the 2002 fiscal year). This information does not reflect grantees' performance in the current program year (2003-04) or the implementation of oversight and management improvements. On page 24 of the draft report, GAO states "The Corporation plans to create consolidated reports in WBRS to facilitate oversight of all grantees' performance. This will allow them to identify programs having enrollment reporting problems more quickly and allow them to focus their oversight on those most in need of attention." This is generally correct, but the Corporation clarifies that its oversight will focus on its grantees-the state commissions and direct grantees such as the national direct and education award programs. In turn, grantees will be responsible for oversight and management of performance at the sub- grantee and individual operating program level. Also on page 24, the Corporation clarifies that it has revised its administrative standards for state commission monitoring of the subgrantees' compliance with the enrollment procedures. These revised standards will be piloted in spring 2004 and then submitted for OMB review after the pilot phase in early fiscal 2005. Also on page 24 of the draft report, GAO states that the Chief Executive Officer's (CEO) certification required by the Strengthen AmeriCorps Program Act has not been completed. This is incorrect. As part of the annual financial statement process the acting CEO certified the Corporation's compliance with the Act. The certification was included in the management representation letter provided to the Corporation's Inspector General on November 13, 2003. The certification is also included in the Corporation's annual Performance and Accountability Report which will be sent to Congress in early February. Finally, Appendix II, in discussing the Antideficiency Act, appears to focus exclusively on the amount of the deficiency attributable to national service positions we approved in excess of available resources in the National Service Trust at a specific time in the past. The appendix calculates this amount without regard to the fact that amounts for lapsed education awards would be deobligated over time; in our view, the deficiency is the amount actually necessary at this time to liquidate obligations related to those national service positions. Based on our analysis, which employs the experience-based obligation approach that Congress provided for in the Strengthen AmeriCorps Program Act, we believe that the $64 million deficiency appropriation that Congress appropriated last spring should be sufficient to liquidate these obligations. RECOMMENDATIONS: GAO Recommendation: Implement a strategy to correct discrepancies between the Trust database and the enrollment and exit forms. Corporation Response: The Corporation agrees with this recommendation. Data supporting member enrollment and exit status is maintained by the Corporation's grantees, which are primarily nonprofit organizations and state governments required to obtain an audit under OMB Circular A-133. The Corporation plans to revise the compliance testing requirements for the AmeriCorps program contained in OMB Circular A-133 to include testing of data submitted by grantees to the National Service Trust database. The Corporation will also review grantee management of enrollment and exit data as part of its grantee monitoring procedures. GAO Recommendation: Review and document the effectiveness of the Corporation's data assurance processes. Corporation Response: The Corporation agrees with this recommendation. The Corporation will better document the data processing that occurs between eSPAN and WBRS and generate standardized reports around the data exchange between the two systems that will produce consistent reliable data. In addition, within the next two years the Corporation plans to begin integrating WBRS functionality into eSPAN, thus eliminating the need to reconcile between the two systems. GAO Recommendation: Regularly verify the accuracy of the social security numbers of participants. Corporation Response: The Corporation agrees with this recommendation. The Corporation entered into a Memorandum of Understanding with the Social Security Administration and tested SSN data in the national Service Trust during fiscal 2002. The Trust office will be required to perform the data verification and resolve discrepancies on an annual basis, or more frequently if warranted by the results. Corporation Response to Draft General Accounting Office Report: Better Internal Controls and Revised Practices Would Improve the Management of AmeriCorps and the National Service Trust (GAO-04-225, Dated December 18, 2003): GAO Recommendation: Update the system documentation for the Trust database, particularly the data dictionary, and develop an inventory of edit and data checks used for the database. Corporation Response: The Corporation agrees with this recommendation. While the Corporation believes that the system documentation is up-to-date, we will continue to strengthen the documentation within Oracle Designer by implementing more rigorous style guidelines and adding additional details to the current data dictionary to enable it to be more user friendly. The Corporation will also assess the reporting that is currently available within Oracle Designer to accommodate the development of an inventory of edit and data checks. GAO Recommendation: Obtain an auditor opinion on the adequacy of the Corporation's internal control over financial reporting as part of the annual financial statement audit. Corporation Response: The Corporation's Office of Inspector General contracts with an independent public accounting firm to perform the annual audit of the Corporation's financial statements and testing of internal controls. We will forward GAO's recommendation to the Inspector General for his consideration. GAO Recommendation: Create a means to take into account the possible impact that external factors may have on participant behavior in funding estimates and budget requests. Corporation Response: The Corporation agrees with this recommendation. The Corporation's Management Improvement Team has begun a project to assess whether unemployment rates, post secondary school enrollment levels, and other external factors can be correlated to the earning and usage rates of education awards. These elements will be factored into the Corporation's budget modeling to the extent practical. GAO Recommendation: Establish and execute a periodic deobligation schedule for unused Trust obligations. Corporation Response: The Corporation agrees with this recommendation. The Corporation's revised Trust procedures include steps to monitor Trust activity and to deobligate funds for AmeriCorps positions not utilized by grantees. GAO Recommendation: Review the assumptions being used in the new funding model after the Corporation gains more experience with the new model and current participant behavior. Corporation Response: The Corporation agrees with this recommendation. The Corporation's procedures require the bi-weekly review of enrollment data and the preparation of quarterly financial reports for the Trust based on the most recent information on the earning and usage rates for AmeriCorps members. As part of these processes the Corporation will review the assumptions being used in the funding model and revise them if necessary. GAO Recommendation: Evaluate the enrollment policies regarding refilling and converting positions. Corporation Response: The Corporation agrees with this recommendation. AmeriCorps program staff will monitor and assess the impact of this policy on grantees and recommend revisions to the policy in the future if necessary. [End of section] Appendix VI: GAO Contacts and Acknowledgments: GAO Contacts: Carolyn M. Taylor (202) 512-2974 or taylorcm@gao.gov Anthony DeFrank (202) 512-4377 or defranka@gao.gov: Staff Acknowledgments: In addition to those named above, the following individuals made important contributions to this report: Joel Marus, Susan Higgins, Elizabeth Lessmann, Patrick diBattista, Hannah Laufe, Thomas Armstrong, James Rebbe, Julie Phillips, LuAnne Moy, Bob Deroy, Jennifer Popovic, and Avrum Ashery. FOOTNOTES [1] U.S. General Accounting Office, Corporation for National and Community Service: Preliminary Observations on the National Service Trust and AmeriCorps, GAO-03-642T (Washington, D.C.). [2] AmeriCorps distributes funds to high school students through Presidential Freedom Scholarships. These scholarships are valued at $500 for each participant. [3] On January 29, 2002, during the State of the Union Address, President George W. Bush said "My call tonight is for every American to commit at least two years--4,000 hours over the rest of your lifetime- -to the service of your neighbors and your nation." [4] State service commissions are governor-appointed public agencies or private nonprofit organizations that distribute AmeriCorps funding within a state. The Corporation reported there are 52 state service commissions, with one each in the District of Columbia, American Samoa, and Puerto Rico, and one in every state except South Dakota. National direct grantees are charitable organizations that operate in more than one state, such as the American National Red Cross. The Corporation provided a list of 48 national direct grantees. [5] Funds deposited into the Trust and obligated for education service awards remain available without fiscal year limitation. For example see Consolidated Appropriations Resolution, 2003, Pub.L.108-7 ("…of which $100,000,000, to remain available without fiscal year limitation, shall be transferred to the National Service Trust for educational awards…") [6] 31 U.S.C. §§ 9105, 9106. [7] Office of Inspector General Report 03-007, The National Service Trust: Internal Control Weaknesses Cause an Anti-Deficiency Act Violation at the Corporation for National and Community Service, July 24, 2003. [8] 31 U.S.C. § 1341(a). [9] The Antideficiency Act requires that if an officer or employee of the United States government violates the act, the agency head "shall report immediately to the President and Congress all relevant facts and a statement of actions taken." 31 U.S.C. §1351. OMB Circular A-11 instructs agencies to submit the transmittal letter and report from the agency head to the President through OMB. Circular A-11 § 145.7. [10] The database used by the Trust is called SPAN, for System for Programs, Agreements, and National Service Participants. [11] Some participants were enrolled for more than one term of service during this period. [12] With a 95 percent confidence interval between 3 and 9 percent. [13] With a 95 percent confidence interval between 1 and 6 percent. [14] U.S. General Accounting Office, Social Security Numbers: Government Benefits from SSN Use, but Could Provide Better Safeguards. GAO-02-352 (Washington, D.C.) May 31, 2002. [15] An updated e-SPAN operator's guide was issued in August 2002 for use by VISTA programs. [16] Agencies are required by OMB to follow Federal Financial Management Systems Requirements issued by the Joint Financial Management Improvement Program (JFMIP). These include maintaining comprehensive and up to date systems documentation. JFMIP is a joint undertaking of the U.S. Department of Treasury, GAO, the Office of Management and Budget and the Office of Personnel Management, working in cooperation with one another and other agencies to improve financial management practices in Government. [17] See OIG Audit Report Number 01-49, June 15, 2001, Assessment of the Service Award Liability Calculation, Price Waterhouse Coopers, December 10, 2001. [18] See OIG Audit Report Number 03-01, February 4, 2003, Audit of the Corporation for National and Community Service Fiscal Year 2002 Financial Statements. [19] Audit Report Number 03-02. Recommended Improvements to the Corporation's Internal Controls, Fiscal Year 2002 Management Letter. January 24, 2003. [20] We are using the SAL model from June 30, 2003. The rate shown is for the 1994 program year. [21] These studies are called the AmeriCorps Attrition Overview Study and the AmeriCorps Education Award Utilization Survey. [22] Testimony of J. Russell George, Inspector General of the Corporation for National and Community Service, Before the Subcommittee on VA, HUD, and Independent Agencies of the Senate Committee on Appropriations. April 10, 2003. [23] U.S. General Accounting Office, Corporation for National and Community Service: Preliminary Observations on the National Service Trust and AmeriCorps. GAO 03-642T (Washington, D.C.) April 10, 2003. [24] Audit Report 04-03, Audit of the Corporation for National and Community Service's National Service Trust Schedule of Fiscal Year 2003 Budgetary Resources and Obligations, October 31, 2003. [25] U.S. General Accounting Office, Government Auditing Standards. GAO-03-673G (Washington, D.C.) June 2003. [26] Because we used a sample (called a probability sample) to develop our estimates, each estimate has a measurable precision, or sampling error, that may be expressed as a plus/minus figure. A sampling error indicates how closely we can reproduce from a sample the results that we would obtain if we were to take a complete count of the universe using the same measurement methods. By adding the sampling error to and subtracting it from the estimate, we can develop upper and lower bounds for each estimate. This range is called a confidence interval. Sampling errors and confidence intervals are stated at a certain confidence level--in this case, 95 percent. For example, a confidence interval at the 95-percent confidence level means that in 95 out of 100 instances, the sampling procedure we used would produce a confidence interval containing the universe value we are estimating. [27] In certain circumstances, paper forms are not required. For example VISTA summer associates do not need to complete an enrollment form. Also, VISTA has developed a practice of not requiring programs to complete the certification section of the paper exit form if the program official who certifies the award status of the participant is the same individual who would complete this section in e-Span. Several cases for which paper forms are not required were included in the sample and we considered these cases as ones for which complete documentation was provided. [28] VISTA participants do not enroll through WBRS. They are enrolled by Corporation officials at AmeriCorps state offices into the Trust database through eSPAN, an enrollment program that feeds into SPAN directly. [29] U.S. General Accounting Office, Standards for Internal Control in the Federal Government. GAO/AIMD-00-21.3.1 (Washington, D.C.) November 1999. [30] Audit of the Corporation for National and Community Service's Fiscal Year 2002 Financial Statements, Audit Report 03-01 at 24, KPMG, Feb. 4, 2003. [31] Interest forbearance refers to the payment of student loan interest by the Corporation for the participant during the period of service. [32] Office of Inspector General Report 03-007, The National Service Trust: Internal Control Weaknesses Cause an Anti-Deficiency Act Violation at the Corporation for National and Community Service, July 24, 2003. [33] The Antideficiency Act prohibits an employee or officer of the United States government from making or authorizing an expenditure or obligation exceeding an amount available in an appropriation. 31 U.S.C. § 1341(a). [34] Preliminary Response by Corporation for National and Community Service to the July 24, 2003, Office of Inspector General Report 03- 007, July 25, 2003. [35] See B-300480, April 9, 2003 and B-300480.2, June 6, 2003, for a more detailed discussion of the obligation practices of the Corporation. [36] The Antideficiency Act requires that if an officer or employee of the United States government violates the act, the agency head "shall report immediately to the President and Congress all relevant facts and a statement of actions taken." 31 U.S.C. §1351. OMB Circular A-11 instructs agencies to submit the transmittal letter and report from the agency head to the President through OMB. Circular A-11 § 145.7. [37] B-300480.2, June 6, 2003. [38] Pub. L. No. 108-45, 117 Stat. 844 (2003). [39] Id. at § 2(b)(1)(B), 117 Stat. at 844. [40] Id. at § 2(b)(A)(ii), 117 Stat. at 844. [41] 31 U.S.C. §1351. [42] Trust funds are held in Treasury securities. [43] Net present value refers to the amount required today to provide a stream of benefits over a number of years. In the case of AmeriCorps, the period of eligibility is about 9 years from the start of the program year. The Corporation uses the weighted average interest earnings of the Trust's portfolio as its discount rate. 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: