This is the accessible text file for GAO report number GAO-13-112 entitled 'U.S. Postal Service: Status, Financial Outlook, and Alternative Approaches to Fund Retiree Health Benefits' which was released on December 27, 2012. This text file was formatted by the U.S. Government Accountability 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. United States Government Accountability Office: GAO: Report to the Chairman, Committee on Oversight and Government Reform, House of Representatives: December 2012: U.S. Postal Service: Status, Financial Outlook, and Alternative Approaches to Fund Retiree Health Benefits: GAO-13-112: GAO Highlights: Highlights of GAO-13-112, a report to the Chairman, Committee on Oversight and Government Reform, U.S. House of Representatives. Why GAO Did This Study: PAEA required USPS to prefund its future retiree health benefits as part of comprehensive postal reform by establishing the PSRHBF along with an initial target period to fund the unfunded liability in 50 years. This requirement included annual payments to this fund from 2007 to 2016 of between $5.4 billion to $5.8 billion. USPS, its employee groups, and others have argued that this prefunding requirement is a major source of USPS’s financial woes-—reported by USPS as contributing $32 billion toward its $41 billion of net losses over the past 6 years. USPS defaulted on the last 2 years of PSRHBF payments totaling $11.1 billion. As requested, this report addresses the (1) status and financial outlook of the PSRHBF, (2) impact on future annual USPS payments and unfunded liabilities of alternative approaches, and (3) key considerations for policymakers. GAO reviewed and summarized PSRHBF financial data and analyzed and compared current law requirements with five alternative approaches by developing projections based on OPM and USPS data. What GAO Found: The Postal Service Retiree Health Benefits Fund (PSRHBF) covered about 49 percent of the U.S. Postal Service’s (USPS) $94 billion retiree health benefit liability at fiscal year-end 2012. USPS’s deteriorating financial outlook, however, will make it difficult to continue the current prefunding schedule in the short term, and possibly to fully fund the remaining $48 billion unfunded liability over the remaining 44 years of the schedule on which the 2006 Postal Accountability and Enhancement Act (PAEA) was based. The liability covers the projected benefits for about 471,000 current postal retirees and a portion of the projected benefits for about 528,000 current employees; it does not cover employees not yet hired. Under PAEA, USPS is responsible for contributing an additional $33.9 billion to the PSRHBF by fiscal year 2017, including the $11.1 billion USPS has defaulted on over the past 2 years. PAEA also requires the Office of Personnel Management (OPM) to calculate the remaining unfunded liability in 2017 and develop an initial 40-year amortization payment schedule. USPS, however, projects further declines in mail volume and revenues that may continue to limit its ability to prefund the remaining retiree health benefit liability. GAO’s analysis of maintaining current law requirements compared to five alternative approaches showed differing impacts on USPS’s future annual payments and unfunded liabilities. For example, three of the approaches-—1) the Administration’s Approach, 2) Senate Bill (S. 1789) and 3) “Pay-as-You-Go” (no prefunding)-—would reduce USPS’s annual payments in the short term, thereby easing its immediate cash flow problems and financial losses. However, these approaches would increase USPS’s unfunded liability, sometimes substantially, and require larger payments later. Deferring funding could increase costs for future ratepayers and increase the possibility that USPS may not be able to pay for some or all of its liability. Conversely, a fourth approach-—the House Bill (H.R. 2309)-—and the current law requirement would reduce USPS’s unfunded liabilities more aggressively but may result in significantly higher USPS financial losses in the near future. If USPS stopped prefunding and let the existing fund grow with interest, the unfunded liability is projected to significantly increase. Under a fifth approach, if USPS stopped prefunding and used the existing fund to pay current and future premiums, the fund is projected to be exhausted by 2026. Private sector, state, local, and other federal entities are not required to prefund these benefits, though some do so to a limited extent, and most are required to recognize the future costs in their financial reporting. GAO identified several key considerations including: (1) the rationale and consequences of prefunding such benefits; (2) trade-offs affecting USPS’s financial condition, such as sizes of the annual payments and unfunded liability; (3) fixed versus actuarially determined payments; (4) targeted funding levels; and (5) assumption criteria. USPS is intended to be a self-sustaining entity funded almost entirely by postal ratepayers, but its financial losses are challenging its sustainability. GAO has testified that USPS should prefund its retiree health benefit liabilities to the maximum extent that its finances permit, but none of the funding approaches may be viable unless USPS has the ability to make the payments. USPS’s default on its last two required PSRHBF payments and its inability to borrow further make the need for a comprehensive package of actions to achieve sustainable financial viability even more urgent. What GAO Recommends: GAO is not making new recommendations in this report, as it has already reported on strategies and options for USPS to achieve sustainable financial viability. In commenting on a draft of this report, USPS and the USPS Office of Inspector General stated that USPS could not afford to prefund. USPS has stated that any discussion of PSRHBF’s outlook must consider USPS’s proposed health plan. GAO is currently reviewing this proposal. View [hyperlink, http://www.gao.gov/products/GAO-13-112]. For more information, contact Frank Todisco at todiscof@gao.gov or Lorelei St. James at stjamesl@gao.gov or call (202) 512-2834. [End of section] Contents: Letter: Background: Status and Financial Outlook of the PSRHBF: USPS's Future Payments and Unfunded Liabilities Vary Widely Based on Approaches to Fund Retiree Health Benefits: Short-term Effects: Long-term Effects: Key Considerations for Assessing Funding Approaches: Concluding Observations: Agency Comments and Our Evaluation: Appendix I: Objectives, Scope, and Methodology: Appendix II: Projected Results Using Senate Bill Assumption Basis: Appendix III: Comments from the U.S. Postal Service: Appendix IV: Comments from the U.S. Postal Service, Office of Inspector General: Appendix V: GAO Contact and Staff Acknowledgments: Tables: Table 1: PSRHBF Payments and Fund Status as of September 30, 2012: Table 2: Key Features of Current Law and Alternative Approaches for Prefunding USPS Retiree Health Benefits: Table 3: Estimated Annual Payments and Unfunded Liabilities in the Short Term (Fiscal Years 2013 to 2020) under Current Law Assumption Basis: Table 4: Estimated Annual Payment and Unfunded Liability in the Long Term (in Fiscal Year 2040, the Last Year of the Projection) under Current Law Assumption Basis: Table 5: Key Differences between the Two sets of PSRHBF Funding Assumptions, as of September 30, 2011: Table 6: Sensitivity of Fiscal Year-end 2011 Unfunded Liability to One- Percentage-Point Change in Medical Inflation Assumption: Table 7: Estimated Annual Payments and Unfunded Liabilities in the Short Term (Fiscal Years 2013 to 2020) under Senate Bill Assumption Basis: Table 8: Annual Payment and Unfunded Liability in the Long Term (in Fiscal Year 2040, the Last Year of the Projection) under Senate Bill Assumption Basis: Figures: Figure 1: USPS Net Operating Profit and Loss, Fiscal Years 2006 to 2012: Figure 2: USPS's First-Class Mail Volume Forecast, Fiscal Years 2013 to 2020: Figure 3: USPS Retiree Health Benefits Fund's Sources as of September 30, 2012: Figure 4: Estimated Annual Payments for Different Approaches from Fiscal Year 2013 to Fiscal Year 2020 under Current Law Assumption Basis: Figure 5: Estimated Annual Payments (as a Percentage of Compensation) for Different Approaches from Fiscal Year 2013 to Fiscal Year 2040 under Current Law Assumption Basis: Figure 6: Estimated Unfunded Liability (as a Percentage of USPS's Annual Compensation Costs) for Different Approaches from Fiscal Year- end 2012 to Fiscal Year-end 2040 under Current Law Assumption Basis: Figure 7: Estimated Funded Ratios for Different Approaches from Fiscal Year-end 2012 to Fiscal Year-end 2040 under Current Law Assumption Basis: Figure 8: Estimated Annual Payments for Modified Senate Prefunding Approach from Fiscal Year 2013 to Fiscal Year 2020 under Current Law Assumption Basis and Senate Bill Assumption Basis: Figure 9: Estimated Annual Payments (as a Percentage of Compensation) for Modified Senate Prefunding Approach from Fiscal Year 2013 through Fiscal Year 2040 under Current Law Assumption Basis and Senate Bill Assumption Basis: Figure 10: Estimated Annual Payments for Different Approaches from Fiscal Year 2013 to Fiscal Year 2020 under Senate Bill Assumption Basis: Figure 11: Estimated Annual Payments (as a Percentage of Compensation) for Different Approaches from Fiscal Year 2013 to Fiscal Year 2040 under Senate Bill Assumption Basis: Figure 12: Estimated Unfunded Liability (as a Percentage of USPS's Annual Compensation Costs) for Different Approaches from Fiscal Year- end 2012 to Fiscal Year-end 2040 under Senate Bill Assumption Basis: Figure 13: Estimated Funded Ratios for Different Approaches from Fiscal Year-end 2012 to Fiscal Year-end 2040 under Senate Bill Assumption Basis: Abbreviations: CSRDF: Civil Service Retirement and Disability Fund: CSRS: Civil Service Retirement System: DOD: Department of Defense: FASAB: Federal Accounting Standards Advisory Board: FASB: Financial Accounting Standards Board: FEHBP: Federal Employees Health Benefits Program: FERS: Federal Employees Retirement System: FTE: full-time employees: GASB: Governmental Accounting Standards Board: IG: Inspector General: OIG: Office of Inspector General: OPEB: other post-employment benefits: OPM: Office of Personnel Management: PAEA: Postal Accountability and Enhancement Act (2006): PRC: Postal Regulatory Commission: PSRHBF: Postal Service Retiree Health Benefits Fund: PVB: present value of future benefits: SFFAS 33: Statement of Federal Financial Accounting Standards No. 33: S&P: Standard & Poor's: USPS: U.S. Postal Service: [End of section] United States Government Accountability Office: Washington, DC 20548: December 4, 2012: The Honorable Darrell E. Issa: Chairman: Committee on Oversight and Government Reform: House of Representatives: Dear Mr. Chairman: Because of significant financial difficulties, the U.S. Postal Service (USPS) was unable to make its statutorily required payments of $5.5 billion in fiscal year 2011 and $5.6 billion in fiscal year 2012 to prefund retiree health benefits.[Footnote 1] The Postal Accountability and Enhancement Act (PAEA) required USPS to make fixed annual payments to begin prefunding the cost of future retiree health benefits accrued by current employees and retirees.[Footnote 2] The act required annual payments (ranging from $5.4 billion to $5.8 billion per year) from fiscal years 2007 through 2016 to be deposited into a new fund established in the U.S. Treasury and administered by the Office of Personnel Management (OPM) called the Postal Service Retiree Health Benefits Fund (PSRHBF). During this period, USPS would also make annual payments for its share of health benefit premiums for current retirees to the Federal Employees Health Benefits Fund (this payment amounted to $2.5 billion in fiscal year 2012).[Footnote 3] Beginning in fiscal year 2017, USPS's share of health benefit premiums for retirees is to be paid from the PSRHBF rather than by USPS. PAEA further required that beginning in fiscal year 2017, USPS's contributions to the PSRHBF are to be based on actuarial calculations performed by OPM. USPS's actuarially determined annual contribution to the PSRHBF is to consist of a payment for the cost of future benefits attributable to employee service during the fiscal year, plus a payment to fund the remaining unfunded retiree health benefit liability over a statutorily determined amortization schedule that initially extends 40 years to 2056.[Footnote 4] USPS, employee organizations, and other stakeholders have argued that the prefunding requirement is a major contributor to USPS's financial decline--contributing $32 billion towards its $41 billion of net losses over the past 6 years[Footnote 5]--and that Congress should relieve USPS from these prefunding requirements or substantially reduce them. As of the end of fiscal year 2012, OPM estimated that USPS's total health benefit liability for future and current retirees[Footnote 6] was approximately $94 billion--of which $48 billion was unfunded and $46 billion was in the PSRHBF. We have previously reported that USPS cannot be financially viable until Congress and USPS address the cash flow problems that limit its immediate prefunding capability while also addressing how to pay for the long-term cost of USPS's unfunded retiree health benefit liability. Projected declines in mail volumes and revenues will continue to exacerbate USPS's difficulties in paying for the cost of its retiree health benefits. Several approaches proposed by the Administration (Administration) and congressional committees offer alternatives that would revise USPS's prefunding payments and affect the amount of funds deposited in the PSRHBF to make future retiree health premium payments. In addition, USPS's Office of Inspector General (OIG) has analyzed a proposal to suspend prefunding and let the existing fund grow with interest. Some have also suggested that prefunding is unnecessary, inadvisable, or unfair. You requested that we review the PSRHBF and study the effects of multiple proposals to revise the payment structure for funding retiree health benefits, which include reducing or deferring payments in the short term, as well as eliminating prefunding altogether. This report (1) describes the status and financial outlook of the Postal Service Retiree Health Benefits Fund, (2) analyzes how alternative approaches for funding retiree health benefits could affect future USPS payments and unfunded liabilities, and (3) determines key considerations for policymakers assessing the alternative approaches. USPS has also proposed withdrawing from the Federal Employees Health Benefits Program (FEHBP) and administering its own health care plan for its employees and retirees. This report looks at retiree health benefits funding options assuming that USPS continues to participate in FEHBP under current provisions. We will be issuing a separate report on USPS's proposal to administer its own health care plan. To describe the status and financial outlook of the PSRHBF, we reviewed and summarized USPS financial data regarding payments made to the fund, interest earned from such contributions, overall fund balance, and retiree health benefit liability. We also reviewed our prior work and reviewed and summarized reports and data from USPS and others on how USPS's financial condition has changed since 2006. We reviewed relevant statutes, proposed legislation, and sections of the President's budget request for fiscal year 2012 pertaining to USPS's health and pension benefit programs. We also interviewed USPS and OPM officials on the status and financial outlook of the PSRHBF. To analyze how alternative proposals for funding retiree health benefits could affect future USPS payments and unfunded liabilities, we analyzed and compared current law requirements and five alternative approaches. The alternative approaches included an approach contained in a bill passed by the Senate, an approach contained in a bill approved by a House committee, an Administration proposal, an approach analyzed by the USPS OIG, and a pay-as-you-go method with no prefunding. We obtained data from USPS on current and projected full- time employee (FTE) counts and compensation. In addition, we met with OPM officials to discuss actuarial assumptions and the methodology for projecting future actuarial costs and premium payment levels under different sets of assumptions, using the workforce projections provided by USPS. We used the projections provided by OPM and USPS to calculate, for each of the alternative approaches to prefunding, USPS's required payments and the PSRHBF's unfunded liability through fiscal year 2040. To determine key factors for policymakers to consider when assessing alternative approaches, we used our own actuarial judgment and expertise. In addition, for comparison purposes we examined the prefunding requirements and prefunding behavior of private-sector entities, state and local governments, and other federal entities. We also looked at how these other entities, as well as USPS, recognize the cost of these benefits in their financial reporting based on relevant accounting standards promulgated by the Financial Accounting Standards Board (FASB), Governmental Accounting Standards Board (GASB), and Federal Accounting Standards Advisory Board (FASAB). For more information on our scope and methodology, see appendix I. We conducted this performance audit from May 2012 to December 2012 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. Background: USPS, an independent establishment of the executive branch, is intended to be a financially self-sufficient entity that covers its expenses almost entirely through postal revenues. In April 2001, we placed USPS on our high-risk list for two reasons.[Footnote 7] First, in the short term, USPS's ability to continue to fulfill its mission on a self-supporting basis was threatened because of projected annual losses of $2 billion to $3 billion, severe cash flow pressures, and debt approaching its statutory borrowing limit without any debt reduction plan. Second, in the long term, increasing retirement- related expenses threatened to reduce USPS's future cash flows and place upward pressures on postal rates. We have been reporting on USPS's financial challenges, including those related to funding its retiree health benefit liability, over the past decade. In May 2002, the Comptroller General testified that USPS had about $100 billion in liabilities, including an estimated $49 billion in unfunded retiree health benefit liability.[Footnote 8] Unlike pension liabilities, USPS had been funding its retiree health benefit liability on a pay-as-you-go basis--an approach in which USPS paid its share of premiums for existing retirees, with no prefunding for any future premiums expected to be paid on behalf of current retirees and workers. In May 2003, the Comptroller General testified that USPS's accounting treatment--which reflected the pay-as-you-go nature of its funding--did not reflect the economic reality of its legal liability to pay for its retiree health benefits, and that current ratepayers were not paying for the full costs of the services they were receiving. Consequently, the pension benefits being earned by USPS employees--which were being prefunded--were recovered through current postal rates, but the retiree health benefits of those same employees were not being recognized in rates until after they retired. The Comptroller General testified that without a change, a sharp escalation in postal rates in future years would be necessary to fund the cost of retiree health benefits on a pay-as-you-go basis.[Footnote 9] Two laws, enacted in 2003 and 2006, reformed USPS's pension liabilities and required it to prefund retiree health benefits: * The Postal Civil Service Retirement System Funding Reform Act of 2003:[Footnote 10] - changed USPS funding of its Civil Service Retirement System (CSRS)[Footnote 11] pension liabilities (based on "dynamic assumptions"[Footnote 12]) while retroactively transferring responsibility for funding the cost of CSRS benefits attributable to the military service of postal employees from the U.S. Treasury to USPS; - required USPS to escrow the reduction in annual CSRS payments resulting from the funding changes in the act (about $3 billion); and: - required USPS to report to Congress on how it could use the CSRS savings realized after fiscal year 2005. USPS proposed to Congress in 2003 that the responsibility for funding the cost of CSRS benefits attributable to the military service of postal employees be transferred back to the U.S. Treasury and that it use the resulting savings to prefund its retiree health benefit liability. * PAEA, enacted in 2006,[Footnote 13] - transferred all responsibility for costs related to CSRS military service credit from USPS back to the U.S. Treasury, both retroactively and prospectively; this included all CSRS military service costs for postal employees since the inception of the Postal Service in 1971; [Footnote 14] - established the PSRHBF to begin prefunding the health benefits of current and future postal retirees and transferred about $20 billion of "start-up" funds into the PSRHBF ($3 billion from the discontinued CSRS escrow--as USPS's annual CSRS payment was suspended--and $17 billion from the surplus in the CSRS fund); - required USPS to make annual payments ranging from $5.4 billion to $5.8 billion per year into the PSRHBF from fiscal years 2007 through 2016 to begin prefunding its retiree health benefit liability; and: - required OPM to calculate the remaining unfunded liability in 2017 and each subsequent year, and to calculate an amortization payment based on an amortization period that extends to 2056 or, if later, 15 years from the then-current fiscal year.[Footnote 15] As a result, in 2007 USPS began prefunding its retiree health benefits as its CSRS pension liability was significantly reduced and its annual CSRS payment was suspended. USPS stated in its 2007 Annual Report that such prefunding was a farsighted and responsible action that placed USPS in the vanguard of both the public and private sectors in providing future security for its employees, and augured well for its long-term financial stability, but also acknowledged that the required payments would be a considerable financial challenge in the near term.[Footnote 16] Contrary to statements made by some employee groups and other stakeholders, PAEA did not require USPS to prefund 75 years of retiree health benefits over a 10-year period. Rather, pursuant to OPM's methodology, such payments would be projected to fund the liability over a period in excess of 50 years, from 2007 through 2056 and beyond (with rolling 15-year amortization periods after 2041). However, the payments required by PAEA were significantly "frontloaded," with the fixed payment amounts in the first 10 years exceeding what actuarially determined amounts would have been using a 50-year amortization schedule. We testified in April 2007 that we had removed USPS from our high-risk list due in part to USPS's financial improvements resulting from these congressional actions.[Footnote 17] From fiscal years 2003 to 2005, USPS's annual pension expense declined by $9 billion. USPS had repaid over $11 billion of outstanding debt, reported $5.4 billion in cost savings and record high net incomes, and delayed rate increases from fiscal year 2003 until January 2006. Since fiscal year 2007, however, USPS has experienced significant financial challenges. USPS's gap between expenses and revenues has grown significantly, as shown in figure 1. In addition, USPS's outstanding debt to the U.S. Treasury increased from $2.1 billion at fiscal year-end 2006 to its current statutory-borrowing limit of $15 billion.[Footnote 18] In fiscal year 2009, we returned USPS to our high-risk list due, in part, to a projected loss of $7 billion--and an actual loss of over $8.5 billion--in fiscal year 2010. For fiscal year 2012, USPS had a net loss of almost $16 billion, which included $11.1 billion for required PSRHBF prefunding payments that USPS did not make.[Footnote 19] Furthermore, USPS's future financial outlook is bleak as it projects further declines in mail volume and revenue by fiscal year 2020. Figure 1: USPS Net Operating Profit and Loss, Fiscal Years 2006 to 2012: [Refer to PDF for image: vertical bar graph] Fiscal year: 2006: $900 million; Fiscal year: 2007: -$5.142 billion; Fiscal year: 2008: -$2,806 billion; Fiscal year: 2009: -$3.794 billion; Fiscal year: 2010: -$8,505 billion; Fiscal year: 2011: -$5.067 billion; Fiscal year: 2012: -$15.906 billion. Source: USPS. [End of figure] USPS projects that First-Class Mail--which is highly profitable and generated about 44 percent of USPS's revenue in fiscal year 2012--will decline in volume by about 42 percent by fiscal year 2020, as shown in figure 2. During the economic downturn, there has been an accelerated diversion of business and individual mail to electronic alternatives, and some businesses have left the mail entirely. USPS further projects that an economic recovery will not bring a corresponding recovery in mail volume because of continuing social and technological trends that have changed the way that people communicate and use the mail. USPS has several initiatives to generate new revenue; however, such efforts are unlikely to generate enough revenue in time to offset the projected decline in mail volume.[Footnote 20] Limited increases in revenue require USPS to seek aggressive cost-saving initiatives to achieve financial stability. Figure 2: USPS's First-Class Mail Volume Forecast, Fiscal Years 2013 to 2020: [Refer to PDF for image: vertical bar graph] Pieces: Fiscal year: 2013: 64 billion; Fiscal year: 2014: 59.9 billion; Fiscal year: 2015: 56.1 billion; Fiscal year: 2016: 52.7 billion; Fiscal year: 2017: 49.1 billion; Fiscal year: 2018: 46 billion; Fiscal year: 2019: 42.2 billion; Fiscal year: 2020: 39 billion. Source: USPS. [End of figure] In February 2012, USPS announced a plan that included a goal of achieving $22.5 billion in annual cost savings by the end of fiscal year 2016. The plan included changes in USPS's mail processing and transportation networks and other cost-saving initiatives, as follows: * $9 billion in network operations, of which $4 billion would come from consolidating its mail processing and transportation networks; * $5 billion in compensation and benefits; and: * $8.5 billion through legislative changes, such as moving to a 5-day delivery schedule. At the same time, USPS's plan would also reduce the overall size of the postal workforce by roughly 155,000 career employees, with many of those reductions expected to result from attrition. USPS reports in the plan that half of its current career employees--283,000 employees-- will be retirement eligible by 2016. In March 2010, USPS presented a detailed proposal to the Postal Regulatory Commission (PRC) to move from a 6-day to a 5-day delivery schedule to achieve its workforce- reduction and cost-savings goals.[Footnote 21] USPS projected that its proposal to move to 5-day delivery by ending Saturday delivery would save about $3 billion annually and would reduce mail volume by less than 1 percent. However, on the basis of its review, PRC estimated a lower annual net savings--about $1.7 billion after a 3-year phase-in period--as it noted that higher revenue losses were possible.[Footnote 22] In February 2012, USPS updated its projected net savings from 5- day delivery to $2.7 billion after a 3-year implementation period. As noted earlier, USPS has also proposed withdrawing from the Federal Employees Health Benefits Program (FEHBP) and administering its own health care plan for its employees and retirees. This report looks at retiree health benefit funding options assuming that USPS continues to participate in FEHBP under current provisions. Adoption of any of the funding approaches analyzed in this report would not by itself preclude USPS from continuing to pursue its proposal to administer its own plan. If USPS's proposal was adopted and if it was expected to result in cost savings, these projected savings would be reflected in a lower liability, a lower unfunded liability, and lower prefunding contributions than otherwise.[Footnote 23] We will be issuing a separate report evaluating USPS's proposal to administer its own health care plan. Related to whether USPS should prefund retiree health benefits, some stakeholders have argued that such prefunding is primarily responsible for USPS's dismal financial condition and is unfair, arguing that no other entity is required to conduct such prefunding. According to a 2011 OPM Inspector General (OIG) report, however, postponing prefunding (deferring payments until later) is financially risky. [Footnote 24] OPM's Inspector General reported that future USPS customers (ratepayers) will have to pay for expenses that the USPS is incurring today and added that deferring payments will likely hurt the USPS's ability to compete in the future and affect its ability to improve its financial situation. The report added that USPS would lose the benefit of the interest that its deposits into the funds would have otherwise earned. This interest would have reduced USPS's future unfunded liabilities for these benefits. Consequently, postponing prefunding would require the USPS to make larger contributions in the future. Status and Financial Outlook of the PSRHBF: At the end of fiscal year 2012, OPM estimated that USPS's total retiree health benefit liability was almost $94 billion, whereas the PSRHBF balance was about $46 billion (49 percent), leaving USPS with an unfunded liability of about $48 billion. Approximately half of the $94 billion liability is for retired annuitants and their survivors while the other half is for current career employees. At fiscal year- end 2012, USPS had about 471,000 annuitants and survivors who were receiving retiree health benefit coverage and about 528,000 career- employees who could become eligible for such coverage when they retire.[Footnote 25] The liability for current employees is a portion of the ultimate liability for their future retiree benefits; the liability accrues steadily over their working years, from zero at date of entry into FEHBP to the full liability at retirement. Contrary to some claims, there is no liability held, nor contributions made, for any future employees who have yet to be hired or yet to be born. PSHRBF's balance comes from three sources. USPS's annual prefunding payments have accounted for $17.9 billion, or 39 percent, of the PSRHBF balance as of September 30, 2012.The remaining balance consists of about $20 billion transferred from USPS's excess CSRS funds (referred to as "start-up funds" in figure 3 below) when the PSRHBF was created in 2007 and approximately $7.8 billion in earned interest (see figure 3). Figure 3: USPS Retiree Health Benefits Fund's Sources as of September 30, 2012: [Refer to PDF for image: pie-chart] Start-up funds: $20.0 billion (44%); USPS prefunding: $17.9 billion (39%); Interest earned: $7.8 billion (17%). Source: USPS. Note: Start-up funds included about $3 billion from the CSRS escrow and about $17 billion from a surplus in the CSRS fund. [End of figure] Because of USPS's financial difficulties, however, USPS has not made all of its required prefunding payments. Under PAEA, USPS is still responsible for contributing an additional $33.9 billion to the PSRHBF by fiscal year 2017 as shown in table 1, including $11.1 billion that USPS has defaulted on over the past 2 years. Originally due at the end of fiscal year 2011, USPS's $5.5 billion required retiree health prefunding payment was delayed until August 1, 2012.[Footnote 26] USPS missed that payment as well as the $5.6 billion that was due by September 30, 2012. Table 1: PSRHBF Payments and Fund Status as of September 30, 2012: CSRS start-up[A]; $20 billion. USPS prefunding payments made: Fiscal year 2007; $5.4 billion. Fiscal year 2008; $5.6 billion. Fiscal year 2009[B]; $1.4 billion. Fiscal year 2010; $5.5 billion. Total USPS prefunding; $17.9 billion. PSRHBF earned interest; $7.8 billion. PSRHBF balance; $45.7 billion. Retiree health benefit liability; $93.6 billion. Unfunded liability; $47.8 billion. USPS outstanding mandated prefunding payments to PSRHBF: Fiscal year 2011[C]; $5.5 billion. Fiscal year 2012[D]; $5.6 billion. Fiscal year 2013; $5.6 billion. Fiscal year 2014; $5.7 billion. Fiscal year 2015; $5.7 billion. Fiscal year 2016; $5.8 billion. Total; $33.9 billion. Source: USPS. Note: Totals may not add because of rounding. [A] The start-up included $3 billion from the CSRS escrow and $17 billion from a surplus in the CSRS fund. [B] USPS's required prefunding payment was reduced from $5.4 billion to $1.4 billion in 2009. Pub. L. No. 111-68, § 164(a) 123 Stat. 2023, 2053 (Oct.1, 2009). [C] Originally due at the end of fiscal year 2011, USPS's $5.5 billion retiree health benefit payment was delayed until August 1, 2012. Pub. L. No. 112-74, § 632. USPS, however, did not make this payment. [D] USPS did not make its $5.6 billion required payment in fiscal year 2012. [End of table] While the PSRHBF balance covered about 49 percent of USPS's retiree health benefit liability at fiscal year-end 2012, USPS's deteriorating financial outlook will make it difficult under current requirements for USPS to continue prefunding the remaining unfunded liability in the short term, and possibly to continue funding the remaining unfunded liability over the next several decades, as required under PAEA. USPS's Future Payments and Unfunded Liabilities Vary Widely Based on Approaches to Fund Retiree Health Benefits: We considered current law (PAEA) requirements against five alternative approaches for funding the costs of retiree health benefits, each of which involves tradeoffs that could impact USPS's short-term cash flow, its future financial condition, different generations of postal ratepayers, and over a million postal employees and retirees. We compared the current law prefunding requirements as well as approaches that have been proposed in (1) a bill passed by the House of Representative's Committee on Oversight and Government Reform, ("House Bill"),[Footnote 27] (2) in the President's fiscal year 2012 budget request ("Administration Approach"), and (3) in a bill passed by the Senate, ("Senate Bill")[Footnote 28]. In addition, some postal stakeholders have argued that prefunding is unnecessary or inadvisable altogether, so we also examined the effects of implementing two variations on a "Pay-as-You-Go Approach." Our projections[Footnote 29] showed that some approaches would reduce USPS's annual payments in the short term, thereby easing its cash flow problems and financial losses in the near future, but would increase its unfunded liability, sometimes substantially, and require larger annual payments later. Other approaches would have USPS reduce its unfunded liabilities more aggressively, which could result in significantly higher financial losses in the near future and thereby increase USPS's financial challenges. These options must be considered within the context of USPS's financial situation and prospects. USPS officials told us that given its current financial problems, USPS simply does not have the cash to implement any of the prefunding options and that larger structural solutions must be found to USPS's financial viability as a self-sustaining entity. PSRHBF Funding: Current law (the 2006 PAEA) consists of distinct prefunding requirements that apply first through fiscal year 2016 and then from fiscal year 2017 forward. Through fiscal year 2016, USPS is required to make fixed annual payments[Footnote 30] ranging from $5.5 billion to $5.8 billion to the PSRHBF in addition to making its share of premium payments for existing retirees and beneficiaries, which OPM has estimated will rise from about $2.5 billion to about $3.8 billion per year between fiscal year 2011 and fiscal year 2016.[Footnote 31] Beginning in fiscal year 2017, the current law switches to an "actuarial approach" for the remaining funding, under which USPS's share of premium payments for existing retirees and beneficiaries is paid from the PSRHBF rather than by USPS, and USPS makes annual payments to the PSRHBF consisting of two components: 1. the actuarially determined cost of future benefits attributable to employee service during the fiscal year (known as the annual "normal cost"), and: 2. the actuarially determined amount that, as calculated by OPM, would be projected to fully fund the remaining unfunded liability over an amortization period ending in the later of fiscal year 2056 or 15 years subsequent to the then-current fiscal year. Current law requires OPM to base its actuarial calculations of prefunding requirements on the actuarial assumptions used by OPM for its financial reporting. We will discuss the relevance of this current law assumption basis later in this report. As discussed earlier, USPS did not make the required 2011 and 2012 payments to the PSRHBF, totaling $11.1 billion. We modeled a modified version of current law assuming that these missed payments are eliminated by legislation and that the current law payment schedule resumes with the payment of $5.6 billion due at the end of fiscal year 2013. We refer to this schedule as "Modified Current Law Approach" in our presentation of results. The three alternative prefunding approaches we examined differ from current law in the following respects. * The House Bill (H.R. 2309) reduces the fixed payment due at the end of fiscal year 2011 from $5.5 billion to $1.0 billion, making up the difference in higher fixed payments in fiscal year 2015 and fiscal year 2016. Starting in fiscal year 2017, the House Bill's actuarial approach for determining prefunding is the same as current law. As with our modeling of current law, because the 2011 and 2012 payments have already been missed, we modeled a modified version of the House Bill in which the House Bill's 2011 and 2012 payments are eliminated by legislation and the House Bill's payment schedule commences in fiscal year 2013. We refer to this schedule as "Modified House Approach" in our presentation of results. * The Administration Approach restructures and generally reduces [Footnote 32] the required fixed prefunding payments in each fiscal year from 2011 through 2016. It also calls for USPS's share of premium payments for existing retirees and beneficiaries to begin to be paid from the PSRHBF right away, rather than beginning in fiscal year 2017. As a result, total USPS payments prior to fiscal year 2017 (prefunding plus any required payment of premiums) are significantly lower under the Administration Approach than under current law or the House Bill (and, consequently, would be somewhat greater after fiscal year 2017 to make up for this). Starting in fiscal year 2017, the Administration's actuarial approach for determining prefunding is the same as current law and the House bill. As with current law and the House Bill, we modeled a "Modified Administration Approach" that eliminates its 2011 and 2012 payments by legislation and commences payment in 2013. * The Senate Bill (S. 1789) differs from current law, the House Bill, and the Administration Approach in three key aspects. First, the Senate Bill eliminates the fixed prefunding payments and begins an actuarial approach to prefunding right away (which we modeled to begin at the start of fiscal year 2013).[Footnote 33] Second, the Senate Bill uses a target of funding 80 percent of the liability, instead of the 100 percent funding targeted by current law and the other approaches. Third, the Senate Bill directs OPM to use actuarial assumptions consistent with those used by OPM to determine funding for USPS's share of liabilities in the federal civilian pension programs. These pension-funding assumptions are selected by OPM, with advice from an independent Board of Actuaries. As discussed further in the next section and later in this report, this assumption basis specified in the Senate Bill differs from the assumption basis specified in current law and retained in the House Bill and Administration proposal. We refer to the Senate Bill provisions as "Modified Senate Approach" in our presentation of results.[Footnote 34] Key features of current law and these alternative approaches are summarized in table 2. In addition, we also modeled two variations on a Pay-as-You-Go Approach, discussed in a subsequent section of this report. Table 2: Key Features of Current Law and Alternative Approaches for Prefunding USPS Retiree Health Benefits: Approach (as modified to begin in fiscal year 2013): Modified Current Law Approach; Funding target (percentage of liability): 100; Initial target year for amortization schedule for funding unfunded liability[A]: 2056; Period when premiums for current retirees begin being paid from PSRHBF instead of by USPS: 2017; Assumption basis: Consistent with OPM financial reporting; Total fixed prefunding payments, 2013-2016 fiscal years: $22.8 billion. Approach (as modified to begin in fiscal year 2013): Modified House Approach; Funding target (percentage of liability): 100; Initial target year for amortization schedule for funding unfunded liability[A]: 2056; Period when premiums for current retirees begin being paid from PSRHBF instead of by USPS: 2017; Assumption basis: Consistent with OPM financial reporting; Total fixed prefunding payments, 2013-2016 fiscal years: $27.3 billion. Approach (as modified to begin in fiscal year 2013): Modified Administration Approach; Funding target (percentage of liability): 100; Initial target year for amortization schedule for funding unfunded liability[A]: 2056; Period when premiums for current retirees begin being paid from PSRHBF instead of by USPS: 2013; Assumption basis: Consistent with OPM financial reporting; Total fixed prefunding payments, 2013-2016 fiscal years: $18.1 billion. Approach (as modified to begin in fiscal year 2013): Modified Senate Approach; Funding target (percentage of liability): 80; Initial target year for amortization schedule for funding unfunded liability[A]: 2052; Period when premiums for current retirees begin being paid from PSRHBF instead of by USPS: Upon enactment of Senate Bill[B]; Assumption basis: Consistent with USPS's CSRS and FERS funding; Total fixed prefunding payments, 2013-2016 fiscal years: NA-- Prefunding payments are actuarially determined, not fixed. Source: GAO. [A] As calculated by OPM, the amortization schedule extends to the later of this target year or 15 years from the then-current fiscal year. [B] We modeled this approach based on enactment of the bill in the beginning of fiscal year 2013. [End of table] Short-and Long-term Effects of Alternative Approaches: Our analysis shows that, over the short-term period ending in fiscal year 2020, the Modified Current Law and House Approaches would decrease USPS's unfunded liability for retiree health benefits, while the Modified Administration and Senate Approaches would increase the unfunded liability. This is mainly the result of significantly higher contributions under the Modified Current Law and House Approaches in fiscal years 2013 through 2016. Over the longer term through 2040, there are significant differences in the projected unfunded liability among the various approaches. The Modified Current Law, House, and Administration Approaches are projected to eliminate most of the unfunded liability over that period; the Modified Senate Approach is projected to leave a larger portion of the liability still unfunded because of its lower funding target, while the two Pay-as-You-Go Approaches we examined would lead to very large unfunded liabilities. It should be understood that projections of this type, especially longer term projections, contain a significant degree of uncertainty. Nonetheless, given the magnitude of the retiree health benefit liabilities and the importance of being able to pay for these benefits, reasonable projections of the associated costs and liabilities provide essential information for enabling responsible stewardship of USPS resources. Our comparison of the four prefunding approaches was complicated by the fact that the Senate Bill calls for selecting assumptions based on different criteria than current law, the House Bill, and the Administration Approach. Assumptions represent estimates of future economic and demographic trends, and while initial assumptions may differ, only one scenario can actually occur, and assumptions generally change over time to reflect emerging experience. Accordingly, to compare the four prefunding approaches, we modeled them under uniform assumptions--first using the current law assumption basis, presented in this section, and then using the Senate bill assumption basis, presented in appendix II. We also discuss the underlying differences between these two assumption bases, and present some comparative results, in the section below on "Sensitivity to Assumptions." Our overall findings were not materially affected by the choice between these two assumption bases. Short-term Effects: For a short-term outlook, we projected USPS's required payments (prefunding contributions as well as premium payments for current retirees, when applicable) and the amount of unfunded liability in fiscal years 2013 through 2020.[Footnote 35] Over this 8-year period, USPS's payments under the Modified Current Law and House Approaches would be significantly greater than under the Modified Administration and Senate Approaches. For example, estimated total payments over this period under the Modified Current Law Approach would be 48 percent greater than under the Modified Senate Approach. In particular, payments over the 8 years would total about $58 billion under Modified Current Law and $61 billion under the Modified House Approach, versus $44 billion under the Modified Administration Approach and $39 billion under the Modified Senate Approach. Higher payments mean a lower unfunded liability at the end of the period, and vice versa. Thus, at the end of fiscal year 2020, the Modified Current Law and House Approaches are projected to result in unfunded liabilities of $39 billion and $35 billion, respectively, whereas the Modified Administration and Senate Approaches are projected to result in unfunded liabilities of $59 billion and $64 billion respectively. Thus, in the short term through fiscal year 2020, the unfunded liability is projected to decrease under the Modified Current Law and House Approaches and increase under the Modified Administration and Senate Approaches (see table 3). Table 3: Estimated Annual Payments and Unfunded Liabilities in the Short Term (Fiscal Years 2013 to 2020) under Current Law Assumption Basis: 2013; Modified Current Law Approach: Annual Payments: $8.6 billion; Unfunded Liability (at Year-end): $47.6 billion; Modified House Approach: Annual Payments: $8.6 billion; Unfunded Liability (at Year-end): $47.6 billion; Modified Administration Approach: Annual Payments: $1.3 billion; Unfunded Liability (at Year-end): $55.0 billion; Modified Senate Approach: Annual Payments: $4.4 billion; Unfunded Liability (at Year-end): $51.8 billion;. 2014; Modified Current Law Approach: Annual Payments: $8.9 billion; Unfunded Liability (at Year-end): $44.6 billion; Modified House Approach: Annual Payments: $8.9 billion; Unfunded Liability (at Year-end): $44.6 billion; Modified Administration Approach: Annual Payments: $5.4 billion; Unfunded Liability (at Year-end): $55.9 billion; Modified Senate Approach: Annual Payments: $4.5 billion; Unfunded Liability (at Year-end): $53.6 billion;. 2015; Modified Current Law Approach: Annual Payments: $9.2 billion; Unfunded Liability (at Year-end): $41.2 billion; Modified House Approach: Annual Payments: $11.5 billion; Unfunded Liability (at Year-end): $39.0 billion; Modified Administration Approach: Annual Payments: $5.6 billion; Unfunded Liability (at Year-end): $56.8 billion; Modified Senate Approach: Annual Payments: $4.5 billion; Unfunded Liability (at Year-end): $55.4 billion;. 2016; Modified Current Law Approach: Annual Payments: $9.6 billion; Unfunded Liability (at Year-end): $37.3 billion; Modified House Approach: Annual Payments: $11.9 billion; Unfunded Liability (at Year-end): $32.7 billion; Modified Administration Approach: Annual Payments: $5.8 billion; Unfunded Liability (at Year-end): $57.5 billion; Modified Senate Approach: Annual Payments: $4.7 billion; Unfunded Liability (at Year-end): $57.1 billion;. 2017; Modified Current Law Approach: Annual Payments: $5.0 billion; Unfunded Liability (at Year-end): $37.9 billion; Modified House Approach: Annual Payments: $4.8 billion; Unfunded Liability (at Year-end): $33.3 billion; Modified Administration Approach: Annual Payments: $6.2 billion; Unfunded Liability (at Year-end): $58.0 billion; Modified Senate Approach: Annual Payments: $4.9 billion; Unfunded Liability (at Year-end): $58.9 billion;. 2018; Modified Current Law Approach: Annual Payments: $5.2 billion; Unfunded Liability (at Year-end): $38.4 billion; Modified House Approach: Annual Payments: $5.0 billion; Unfunded Liability (at Year-end): $33.9 billion; Modified Administration Approach: Annual Payments: $6.4 billion; Unfunded Liability (at Year-end): $58.3 billion; Modified Senate Approach: Annual Payments: $5.1 billion; Unfunded Liability (at Year-end): $60.5 billion;. 2019; Modified Current Law Approach: Annual Payments: $5.4 billion; Unfunded Liability (at Year-end): $38.8 billion; Modified House Approach: Annual Payments: $5.2 billion; Unfunded Liability (at Year-end): $34.4 billion; Modified Administration Approach: Annual Payments: $6.6 billion; Unfunded Liability (at Year-end): $58.6 billion; Modified Senate Approach: Annual Payments: $5.3 billion; Unfunded Liability (at Year-end): $62.2 billion;. 2020; Modified Current Law Approach: Annual Payments: $5.7 billion; Unfunded Liability (at Year-end): $39.2 billion; Modified House Approach: Annual Payments: $5.4 billion; Unfunded Liability (at Year-end): $34.7 billion; Modified Administration Approach: Annual Payments: $6.8 billion; Unfunded Liability (at Year-end): $58.7 billion; Modified Senate Approach: Annual Payments: $5.5 billion; Unfunded Liability (at Year-end): $63.8 billion;. Total Payments; Modified Current Law Approach: Annual Payments: $57.7 billion; Unfunded Liability (at Year-end): NA; Modified House Approach: Annual Payments: $61.1 billion; Unfunded Liability (at Year-end): NA; Modified Administration Approach: Annual Payments: $44.1 billion; Unfunded Liability (at Year-end): NA; Modified Senate Approach: Annual Payments: $39.0 billion; Unfunded Liability (at Year-end): NA. Average Annual Payment; Modified Current Law Approach: Annual Payments: $7.2 billion; Unfunded Liability (at Year-end): NA; Modified House Approach: Annual Payments: $7.6 billion; Unfunded Liability (at Year-end): NA; Modified Administration Approach: Annual Payments: $5.5 billion; Unfunded Liability (at Year-end): NA; Modified Senate Approach: Annual Payments: $4.9 billion; Unfunded Liability (at Year-end): NA. Source: GAO. Note: Totals may not add because of rounding. Annual payments consist of any prefunding payments to the PSRHBF, and, where applicable, USPS's share of premium payments for existing retirees and beneficiaries (see descriptions of the different approaches). Unfunded liabilities are the amounts as of the end of the fiscal year, and start at an estimated value of $50.1 billion as of September 30, 2012, under all four approaches. Subsequent to the development of these projections, USPS reported the actual unfunded liability as of September 30, 2012, to be $47.9 billion. [End of table] The higher payments required under the Modified Current Law and Modified House Approaches are concentrated in fiscal years 2013 through 2016, prior to the switch to an actuarial basis in fiscal year 2017. These payments in fiscal years 2013 through 2016 are significantly higher than those that would be calculated under the specified actuarial approach.[Footnote 36] For example, the payments that USPS would have to make in fiscal year 2016 under the Modified House Approach, $11.9 billion, is more than double the following year's payment of $4.8 billion, when the actuarial approach begins. Similarly, under Modified Current Law, the estimated 2016 payments would be almost double the 2017 payments. In contrast, payments under the Modified Administration and Senate Approaches are more affordable and consistent over the period. Figure 4 illustrates these patterns over the fiscal years 2013 through 2020. Figure 4: Estimated Annual Payments for Different Approaches from Fiscal Year 2013 to Fiscal Year 2020 under Current Law Assumption Basis: [Refer to PDF for image: vertical bar graph] Fiscal year: 2013; Modified Current Law Approach: $8.557 billion; Modified House Approach: $8.557 billion; Modified Administration Approach: $1.276 billion; Modified Senate Approach: $4.432 billion. Fiscal year: 2014; Modified Current Law Approach: $8.925 billion; Modified House Approach: $8.925 billion; Modified Administration Approach: $5.423 billion; Modified Senate Approach: $4.459 billion. Fiscal year: 2015; Modified Current Law Approach: $9.213 billion; Modified House Approach: $11.463 billion; Modified Administration Approach: $5.573 billion; Modified Senate Approach: $4.545 billion. Fiscal year: 2016; Modified Current Law Approach: $9.615 billion; Modified House Approach: $11.865 billion; Modified Administration Approach: $5.829 billion; Modified Senate Approach: $4.711 billion. Fiscal year: 2017; Modified Current Law Approach: $5.042 billion; Modified House Approach: $4.777 billion; Modified Administration Approach: $6.205 billion; Modified Senate Approach: $4.902 billion. Fiscal year: 2018; Modified Current Law Approach: $5.230 billion; Modified House Approach: $4.965 billion; Modified Administration Approach: $6.394 billion; Modified Senate Approach: $5.097 billion. Fiscal year: 2019; Modified Current Law Approach: $5.428 billion; Modified House Approach: $5.163 billion; Modified Administration Approach: $6.591 billion; Modified Senate Approach: $5.305 billion. Fiscal year: 2020; Modified Current Law Approach: $5.657 billion; Modified House Approach: $5.392 billion; Modified Administration Approach: $6.821 billion; Modified Senate Approach: $5.549 billion. Source: GAO. [End of figure] In summary, these projections illustrate: (1) the inverse trade-off between payment requirements and unfunded liability (higher payments mean lower unfunded liability; lower payments mean higher unfunded liability) and (2) possible patterns for required payments (e.g., front-loaded, consistent, or back-loaded). These options must be considered within the context of USPS's financial situation and prospects. Long-term Effects: We extended our projection of USPS's required payments and the amount of unfunded liability to fiscal year 2040. While the uncertainty of a projection increases with the length of the projection period, a longer projection period allows potential longer-term implications of different approaches to emerge--effects that might not be observable under a short-term projection. Since dollar amounts in fiscal year 2040 are not fully comparable to dollar amounts today, it is helpful to "normalize" such long-term projections to make the results more comparable across time periods. In table 4 we show projected payments and unfunded liability in fiscal year 2040 in three different ways: (1) as nominal (unadjusted) dollar amounts; (2) in constant (inflation- adjusted) 2012 dollars; and (3) as a percentage of USPS's projected 2040 modified employee compensation costs[Footnote 37] (which for convenience we refer to as "compensation"). We also show the projected funded percentage--or the ratio of PSRHBF assets to USPS's liability for retiree health benefits. Table 4: Estimated Annual Payment and Unfunded Liability in the Long Term (in Fiscal Year 2040, the Last Year of the Projection) under Current Law Assumption Basis: Dollars in billions. Nominal $; Modified Current Law Approach: Annual Payment: $11.8 billion; Unfunded Liability: $9.4 billion; Modified House Approach: Annual Payment: $11.5 billion; Unfunded Liability: $6.5 billion; Modified Administration Approach: Annual Payment: $12.9 billion; Unfunded Liability: $22.1 billion; Modified Senate Approach: Annual Payment: $12.5 billion; Unfunded Liability: $66.5 billion. Constant $; Modified Current Law Approach: Annual Payment: $6.1 billion; Unfunded Liability: $4.8 billion; Modified House Approach: Annual Payment: $5.9 billion; Unfunded Liability: $3.4 billion; Modified Administration Approach: Annual Payment: $6.7 billion; Unfunded Liability: $11.4 billion; Modified Senate Approach: Annual Payment: $6.4 billion; Unfunded Liability: $34.2 billion. Percentage of Compensation; Modified Current Law Approach: Annual Payment: 15%; Unfunded Liability: 12%; Modified House Approach: Annual Payment: 14%; Unfunded Liability: 8%; Modified Administration Approach: Annual Payment: 16%; Unfunded Liability: 28%; Modified Senate Approach: Annual Payment: 16%; Unfunded Liability: 83%. Funded percentage; Modified Current Law Approach: Annual Payment: NA; Unfunded Liability: 96%; Modified House Approach: Annual Payment: NA; Unfunded Liability: 97%; Modified Administration Approach: Annual Payment: NA; Unfunded Liability: 91%; Modified Senate Approach: Annual Payment: NA; Unfunded Liability: 73%. Source: GAO. Note: Annual payment is prefunding payment to the PSRHBF. Unfunded liability is as of the end of the fiscal year. Constant dollar amounts were derived by converting the projected 2040 nominal dollar amounts into 2012 dollars, using an inflation assumption based on the current law assumption basis. Compensation was projected by GAO based on information provided by USPS. Additional information on methodology and assumptions is presented in appendix I. [End of table] Showing payments and unfunded liability amounts as a percentage of compensation provides a sense of the size of USPS's retiree health care costs relative to the size of USPS's operations. Projecting compensation does require an additional assumption regarding compensation growth and therefore introduces additional uncertainty into the projection.[Footnote 38] Nonetheless, increases over time in projected payments and unfunded liabilities as a percentage of compensation can be indicative of a likely greater strain on USPS's resources. For example, unfunded liability as a percentage of compensation will rise to the extent that USPS is operating with a reduced workforce. As seen in table 4, in comparing the projections under Modified Current Law and the three alternative modified approaches, the differences in the projected payment required in fiscal year 2040 are not large, with the dollar amount of the projected payment ranging from $11.5 billion to $12.9 billion across the four approaches ($5.9 billion to $6.7 billion in constant dollars). The more significant differences in the annual payments, across the four approaches, occur in the short-term period covering fiscal years 2013 through 2016. There are, however, significant differences in the projected unfunded liability in fiscal year 2040. The Modified Senate Approach results in a projected unfunded liability of about $67 billion; this compares to unfunded liabilities of $22 billion under the Modified Administration Approach, $9 billion under Modified Current Law, and $7 billion under the Modified House Approach. This projected unfunded liability under the Modified Senate Approach amounts to $34 billion in constant dollars and 83 percent of projected annual compensation. For the Modified Administration Approach, the projected unfunded liability is $11 billion in constant dollars and 28 percent of projected compensation. The corresponding results under the Modified Current Law and House Approaches are significantly smaller. For example, as a percentage of projected compensation, the projected unfunded liability is 12 percent and 8 percent under the Modified Current Law and House Approaches, respectively. A primary reason for these differences is that the Senate Approach uses a target funded percentage of 80 percent, whereas the other three approaches use a target of 100 percent. By fiscal year 2040, the funded percentage is projected to have reached 73 percent under the Modified Senate Approach, versus 91 percent under the Modified Administration Approach, 96 percent under Modified Current Law, and 97 percent under the Modified House Approach. It is important to note that reaching a 100 percent funded percentage-- that is, the unfunded liability is fully paid off and PSRHBF assets equal the liability--would not mean that USPS would have no further prefunding payments to make. USPS would continue to have to pay the "normal cost" each year into the fund, reduced by amortization of any surplus that might develop if experience is more favorable than assumed, or increased if less favorable such that the funded percentage falls back under 100 percent. As mentioned earlier, the normal cost is the actuarially determined cost of future benefits attributable to employee service during the fiscal year, a cost that increases the liability each year. Failure to continue to make such contributions into the fund each year would mean a failure to pay for the cost of then-current employee service; the likely result would be PSRHBF assets again falling short of the liability, thereby creating a new unfunded liability. Under any of the approaches modeled, by fiscal year 2040, roughly 80 to 90 percent of USPS's required payment would consist of this normal cost. Short-and Long-term Effects Combined: Combining the short-term and long-term projection results, figure 5 illustrates projected annual payments, as a percentage of projected compensation, for each fiscal year from 2013 through 2040. The largest differences among the four approaches occur from fiscal year 2013 through fiscal year 2016. Under the Modified House Approach, estimated required payments are in excess of 20 percent of compensation in all 4 of these years, climbing to 28 and 29 percent of compensation in fiscal year 2015 and fiscal year 2016. Under Modified Current Law, estimated required payments are also in excess of 20 percent of compensation in each of these years, peaking at 23 percent of compensation in fiscal year 2016. In contrast, under the Modified Administration Approach, estimated required payments start at just 3 percent of compensation in fiscal year 2013 before climbing to 13 to 14 percent of compensation in the ensuing three years. Under the Modified Senate Approach, estimated required payments round to a steady 11 percent of compensation in each of these first 4 years. From fiscal year 2017 through fiscal year 2040, the estimated required contributions are closer together across the four approaches, ranging from 11 to 17 percent of compensation. The projected payments are somewhat higher under the Modified Administration Approach than under the Modified Current Law or House Approaches, in order to make up for the differences in payments from fiscal year 2013 through fiscal year 2016. The projected payments under the Modified Senate Approach follow a slightly different trajectory because of the approach's 80 percent funding target, but fall within the same range.[Footnote 39] Figure 5: Estimated Annual Payments (as a Percentage of Compensation) for Different Approaches from Fiscal Year 2013 to Fiscal Year 2040 under Current Law Assumption Basis: [Refer to PDF for image: multiple line graph] Fiscal year: 2013; Modified Current Law Approach: 20.4%; Modified House Approach: 20.4%; Modified Administration Approach: 3%; Modified Senate Approach: 10.6%. Fiscal year: 2014; Modified Current Law Approach: 21.6%; Modified House Approach: 21.6%; Modified Administration Approach: 13%; Modified Senate Approach: 10.8%. Fiscal year: 2015; Modified Current Law Approach: 22.4%; Modified House Approach: 27.9%; Modified Administration Approach: 13.5%; Modified Senate Approach: 11%. Fiscal year: 2016; Modified Current Law Approach: 23.3%; Modified House Approach: 28.7%; Modified Administration Approach: 14.1%; Modified Senate Approach: 11.4%. Fiscal year: 2017; Modified Current Law Approach: 12.1%; Modified House Approach: 11.5%; Modified Administration Approach: 14.9%; Modified Senate Approach: 11.8%. Fiscal year: 2018; Modified Current Law Approach: 12.5%; Modified House Approach: 11.8%; Modified Administration Approach: 15.3%; Modified Senate Approach: 12.2%. Fiscal year: 2019; Modified Current Law Approach: 12.8%; Modified House Approach: 12.2%; Modified Administration Approach: 15.6%; Modified Senate Approach: 12.6%. Fiscal year: 2020; Modified Current Law Approach: 13.2%; Modified House Approach: 12.6%; Modified Administration Approach: 15.9%; Modified Senate Approach: 13%. Fiscal year: 2021; Modified Current Law Approach: 13.7%; Modified House Approach: 13%; Modified Administration Approach: 16.3%; Modified Senate Approach: 13.4%. Fiscal year: 2022; Modified Current Law Approach: 14%; Modified House Approach: 13.4%; Modified Administration Approach: 16.6%; Modified Senate Approach: 13.8%. Fiscal year: 2023; Modified Current Law Approach: 14.1%; Modified House Approach: 13.6%; Modified Administration Approach: 16.7%; Modified Senate Approach: 14%. Fiscal year: 2024; Modified Current Law Approach: 14.3%; Modified House Approach: 13.7%; Modified Administration Approach: 16.8%; Modified Senate Approach: 14.2%. Fiscal year: 2025; Modified Current Law Approach: 14.4%; Modified House Approach: 13.9%; Modified Administration Approach: 16.8%; Modified Senate Approach: 14.4%. Fiscal year: 2026; Modified Current Law Approach: 14.6%; Modified House Approach: 14%; Modified Administration Approach: 16.9%; Modified Senate Approach: 14.6%. Fiscal year: 2027; Modified Current Law Approach: 14.7%; Modified House Approach: 14.2%; Modified Administration Approach: 16.9%; Modified Senate Approach: 14.8%. Fiscal year: 2028; Modified Current Law Approach: 14.8%; Modified House Approach: 14.3%; Modified Administration Approach: 16.9%; Modified Senate Approach: 15%. Fiscal year: 2029; Modified Current Law Approach: 14.9%; Modified House Approach: 14.4%; Modified Administration Approach: 17%; Modified Senate Approach: 15.2%. Fiscal year: 2030; Modified Current Law Approach: 14.9%; Modified House Approach: 14.5%; Modified Administration Approach: 17%; Modified Senate Approach: 15.3%. Fiscal year: 2031; Modified Current Law Approach: 15%; Modified House Approach: 14.6%; Modified Administration Approach: 17%; Modified Senate Approach: 15.5%. Fiscal year: 2032; Modified Current Law Approach: 15.1%; Modified House Approach: 14.6%; Modified Administration Approach: 17%; Modified Senate Approach: 15.6%. Fiscal year: 2033; Modified Current Law Approach: 15.1%; Modified House Approach: 14.7%; Modified Administration Approach: 16.9%; Modified Senate Approach: 15.7%. Fiscal year: 2034; Modified Current Law Approach: 15.1%; Modified House Approach: 14.7%; Modified Administration Approach: 16.9%; Modified Senate Approach: 15.8%. Fiscal year: 2035; Modified Current Law Approach: 15.1%; Modified House Approach: 14.7%; Modified Administration Approach: 16.8%; Modified Senate Approach: 15.9%. Fiscal year: 2036; Modified Current Law Approach: 15.1%; Modified House Approach: 14.7%; Modified Administration Approach: 16.7%; Modified Senate Approach: 15.9%. Fiscal year: 2037; Modified Current Law Approach: 15%; Modified House Approach: 14.7%; Modified Administration Approach: 16.6%; Modified Senate Approach: 15.9%. Fiscal year: 2038; Modified Current Law Approach: 14.9%; Modified House Approach: 14.6%; Modified Administration Approach: 16.5%; Modified Senate Approach: 15.9%. Fiscal year: 2039; Modified Current Law Approach: 14.8%; Modified House Approach: 14.5%; Modified Administration Approach: 16.3%; Modified Senate Approach: 15.7%. Fiscal year: 2040; Modified Current Law Approach: 14.6%; Modified House Approach: 14.3%; Modified Administration Approach: 16.1%; Modified Senate Approach: 15.5%. Source: GAO. [End of figure] Figure 6 illustrates the projected unfunded liability, as a percentage of projected USPS annual compensation costs, as of the end of each fiscal year from 2012 through 2040. For each approach modeled, the projection starts at the estimated unfunded liability of 108 percent of compensation as of the end of fiscal year 2012.[Footnote 40] In the short term, the unfunded liability as a percentage of compensation trends down under the Modified House Approach and Modified Current Law Approach because of their relatively high required early payments, while the unfunded liability as a percentage of compensation trends upward under the Modified Administration Approach and, for a somewhat longer period, under the Modified Senate Approach. In the longer term, the unfunded liability as a percentage of compensation trends down under all four approaches. By the end of the projection period in fiscal year 2040, the vast majority of the unfunded liability, measured as a percentage of compensation, is projected to be eliminated under the Modified House and Modified Current Law Approaches, while a smaller majority of it is projected to be eliminated under the Modified Administration Approach. A larger unfunded liability as a percentage of compensation is projected to be retained under the Modified Senate Approach because of its 80 percent funding target. Figure 6: Estimated Unfunded Liability (as a Percentage of USPS's Annual Compensation Costs) for Different Approaches from Fiscal Year- end 2012 to Fiscal Year-end 2040 under Current Law Assumption Basis: [Refer to PDF for image: multiple line graph] Fiscal year: 2012; Modified Current Law Approach: 107.9%; Modified House Approach: 108%; Modified Administration Approach: 107.9%%; Modified Senate Approach: 107.9%. Fiscal year: 2013; Modified Current Law Approach: 113.6; Modified House Approach: 113.6%; Modified Administration Approach: 131.2%; Modified Senate Approach: 123.6%. Fiscal year: 2014; Modified Current Law Approach: 108%; Modified House Approach: 108%; Modified Administration Approach: 135.4%; Modified Senate Approach: 129.7%. Fiscal year: 2015; Modified Current Law Approach: 100.2%; Modified House Approach: 94.7%; Modified Administration Approach: 138.1%; Modified Senate Approach: 134.6%. Fiscal year: 2016; Modified Current Law Approach: 90.3%; Modified House Approach: 79.1%; Modified Administration Approach: 139.3%; Modified Senate Approach: 138.4%. Fiscal year: 2017; Modified Current Law Approach: 91.1%; Modified House Approach: 80.1%; Modified Administration Approach: 139.4%; Modified Senate Approach: 141.6%. Fiscal year: 2018; Modified Current Law Approach: 91.6%; Modified House Approach: 80.8%; Modified Administration Approach: 139.1%; Modified Senate Approach: 144.4%. Fiscal year: 2019; Modified Current Law Approach: 92%; Modified House Approach: 81.3%; Modified Administration Approach: 138.6%; Modified Senate Approach: 147.2%. Fiscal year: 2020; Modified Current Law Approach: 91.6%; Modified House Approach: 81.2%; Modified Administration Approach: 137.2%; Modified Senate Approach: 149.1%. Fiscal year: 2021; Modified Current Law Approach: 91.1%; Modified House Approach: 80.9%; Modified Administration Approach: 135.7%; Modified Senate Approach: 150.9%. Fiscal year: 2022; Modified Current Law Approach: 90.1%; Modified House Approach: 80.2%; Modified Administration Approach: 133.4%; Modified Senate Approach: 151.8%. Fiscal year: 2023; Modified Current Law Approach: 87.6%; Modified House Approach: 78.1%; Modified Administration Approach: 128.9%; Modified Senate Approach: 150.3%. Fiscal year: 2024; Modified Current Law Approach: 84.8%; Modified House Approach: 75.8%; Modified Administration Approach: 124.3%; Modified Senate Approach: 148.5%. Fiscal year: 2025; Modified Current Law Approach: 81.8%; Modified House Approach: 73.2%; Modified Administration Approach: 119.5%; Modified Senate Approach: 146.4%. Fiscal year: 2026; Modified Current Law Approach: 78.6%; Modified House Approach: 70.5%; Modified Administration Approach: 114.5%; Modified Senate Approach: 144.1%. Fiscal year: 2027; Modified Current Law Approach: 75.3%; Modified House Approach: 67.5%; Modified Administration Approach: 109.4%; Modified Senate Approach: 141.5%. Fiscal year: 2028; Modified Current Law Approach: 71.7%; Modified House Approach: 64.2%; Modified Administration Approach: 104.2%; Modified Senate Approach: 138.7%. Fiscal year: 2029; Modified Current Law Approach: 67.8%; Modified House Approach: 60.8%; Modified Administration Approach: 98.7%; Modified Senate Approach: 135.5%. Fiscal year: 2030; Modified Current Law Approach: 63.8%; Modified House Approach: 57.1%; Modified Administration Approach: 93.1%; Modified Senate Approach: 132.1%. Fiscal year: 2031; Modified Current Law Approach: 59.5%; Modified House Approach: 53.2%; Modified Administration Approach: 87.3%; Modified Senate Approach: 128.4%. Fiscal year: 2032; Modified Current Law Approach: 55%; Modified House Approach: 49%; Modified Administration Approach: 81.3%; Modified Senate Approach: 124.3%. Fiscal year: 2033; Modified Current Law Approach: 50.3%; Modified House Approach: 44.6%; Modified Administration Approach: 75.2%; Modified Senate Approach: 120%. Fiscal year: 2034; Modified Current Law Approach: 45.4%; Modified House Approach: 40%; Modified Administration Approach: 68.8%; Modified Senate Approach: 115.4%. Fiscal year: 2035; Modified Current Law Approach: 40.2%; Modified House Approach: 35.2%; Modified Administration Approach: 62.3%; Modified Senate Approach: 110.5%. Fiscal year: 2036; Modified Current Law Approach: 34.8%; Modified House Approach: 30.1%; Modified Administration Approach: 55.6%; Modified Senate Approach: 105.3%. Fiscal year: 2037; Modified Current Law Approach: 29.3%; Modified House Approach: 24.9%; Modified Administration Approach: 48.8%; Modified Senate Approach: 99.9%. Fiscal year: 2038; Modified Current Law Approach: 23.6%; Modified House Approach: 19.4%; Modified Administration Approach: 41.8%; Modified Senate Approach: 94.2%. Fiscal year: 2039; Modified Current Law Approach: 17.7%; Modified House Approach: 13.8%; Modified Administration Approach: 34.7%; Modified Senate Approach: 88.5%. Fiscal year: 2040; Modified Current Law Approach: 11.7%; Modified House Approach: 8.1%; Modified Administration Approach: 27.5%; Modified Senate Approach: 82.6%. Source: GAO. Note: Under all four modified approaches to prefunding, the unfunded liability starts at the same point at September 30, 2012, and begins to diverge in fiscal year 2013 with the different approaches to prefunding. [End of figure] Figure 7 illustrates the funding gap by another measure, the funded ratio--that is, the percentage of the liability that is covered by PSRHBF assets. This figure illustrates how a divergence emerges from fiscal year 2013 to fiscal year 2016 between the Modified House and Current Law Approaches on the one hand, and the Modified Administration and Senate Approaches on the other hand. By the end of the projection period, the funded ratio is projected to be just short of the 100 percent target under the Modified House and Modified Current Law Approaches, slightly further away from 100 percent under the modified Administration Approach, and, under the Modified Senate Approach, approaching its lower 80 percent funded ratio target. Figure 7: Estimated Funded Ratios for Different Approaches from Fiscal Year-end 2012 to Fiscal Year-end 2040 under Current Law Assumption Basis: [Refer to PDF for image: multiple line graph] Fiscal year: 2012; Modified Current Law Approach: 48%; Modified House Approach: 48%; Modified Administration Approach: 48%; Modified Senate Approach: 48%. Fiscal year: 2013; Modified Current Law Approach: 53%; Modified House Approach: 53%; Modified Administration Approach: 46%; Modified Senate Approach: 49%. Fiscal year: 2014; Modified Current Law Approach: 58%; Modified House Approach: 58%; Modified Administration Approach: 48%; Modified Senate Approach: 50%. Fiscal year: 2015; Modified Current Law Approach: 63%; Modified House Approach: 65%; Modified Administration Approach: 49%; Modified Senate Approach: 51%. Fiscal year: 2016; Modified Current Law Approach: 68%; Modified House Approach: 72%; Modified Administration Approach: 51%; Modified Senate Approach: 52%. Fiscal year: 2017; Modified Current Law Approach: 69%; Modified House Approach: 73%; Modified Administration Approach: 53%; Modified Senate Approach: 52%. Fiscal year: 2018; Modified Current Law Approach: 70%; Modified House Approach: 74%; Modified Administration Approach: 55%; Modified Senate Approach: 53%. Fiscal year: 2019; Modified Current Law Approach: 71%; Modified House Approach: 74%; Modified Administration Approach: 56%; Modified Senate Approach: 54%. Fiscal year: 2020; Modified Current Law Approach: 72%; Modified House Approach: 75%; Modified Administration Approach: 58%; Modified Senate Approach: 54%. Fiscal year: 2021; Modified Current Law Approach: 73%; Modified House Approach: 76%; Modified Administration Approach: 59%; Modified Senate Approach: 55%. Fiscal year: 2022; Modified Current Law Approach: 74%; Modified House Approach: 76%; Modified Administration Approach: 61%; Modified Senate Approach: 55%. Fiscal year: 2023; Modified Current Law Approach: 74%; Modified House Approach: 77%; Modified Administration Approach: 62%; Modified Senate Approach: 56%. Fiscal year: 2024; Modified Current Law Approach: 75%; Modified House Approach: 78%; Modified Administration Approach: 64%; Modified Senate Approach: 57%. Fiscal year: 2025; Modified Current Law Approach: 76%; Modified House Approach: 79%; Modified Administration Approach: 65%; Modified Senate Approach: 57%. Fiscal year: 2026; Modified Current Law Approach: 77%; Modified House Approach: 79%; Modified Administration Approach: 67%; Modified Senate Approach: 58%. Fiscal year: 2027; Modified Current Law Approach: 78%; Modified House Approach: 80%; Modified Administration Approach: 68%; Modified Senate Approach: 59%. Fiscal year: 2028; Modified Current Law Approach: 79%; Modified House Approach: 81%; Modified Administration Approach: 69%; Modified Senate Approach: 59%. Fiscal year: 2029; Modified Current Law Approach: 80%; Modified House Approach: 82%; Modified Administration Approach: 71%; Modified Senate Approach: 60%. Fiscal year: 2030; Modified Current Law Approach: 81%; Modified House Approach: 83%; Modified Administration Approach: 72%; Modified Senate Approach: 61%. Fiscal year: 2031; Modified Current Law Approach: 82%; Modified House Approach: 84%; Modified Administration Approach: 74%; Modified Senate Approach: 62%. Fiscal year: 2032; Modified Current Law Approach: 84%; Modified House Approach: 85%; Modified Administration Approach: 76%; Modified Senate Approach: 63%. Fiscal year: 2033; Modified Current Law Approach: 85%; Modified House Approach: 87%; Modified Administration Approach: 77%; Modified Senate Approach: 64%. Fiscal year: 2034; Modified Current Law Approach: 86%; Modified House Approach: 88%; Modified Administration Approach: 79%; Modified Senate Approach: 65%. Fiscal year: 2035; Modified Current Law Approach: 88%; Modified House Approach: 89%; Modified Administration Approach: 81%; Modified Senate Approach: 66%. Fiscal year: 2036; Modified Current Law Approach: 89%; Modified House Approach: 91%; Modified Administration Approach: 83%; Modified Senate Approach: 68%. Fiscal year: 2037; Modified Current Law Approach: 91%; Modified House Approach: 92%; Modified Administration Approach: 85%; Modified Senate Approach: 69%. Fiscal year: 2038; Modified Current Law Approach: 93%; Modified House Approach: 94%; Modified Administration Approach: 87%; Modified Senate Approach: 70%. Fiscal year: 2039; Modified Current Law Approach: 94%; Modified House Approach: 96%; Modified Administration Approach: 89%; Modified Senate Approach: 72%. Fiscal year: 2040; Modified Current Law Approach: 96%; Modified House Approach: 97%; Modified Administration Approach: 91%; Modified Senate Approach: 73%. Source: GAO. Note: Under all four modified approaches to prefunding, the funded ratio starts at the same point at September 30, 2012, and begins to diverge in fiscal year 2013 with the different approaches to prefunding. [End of figure] The annual prefunding payments that have been made since prefunding commenced in 2007--and that would continue to be made under any of the four prefunding approaches examined here--can be broken down into two components: a portion to pay for the cost of future benefits attributable to the current year of employees' service (the "normal cost"), and the remainder, which pays down part of the unfunded liability. One of the rationales for prefunding is to pay for benefits as they are earned--during the working years--rather than later after the workers have retired and are no longer generating revenue for the enterprise. Further, this serves the purpose of assigning full costs of current employee compensation to current ratepayers, rather than to future ratepayers. A complicating factor is what might be called the "legacy" unfunded liability, i.e., the existing unfunded liability that conceptually should have been paid by ratepayers in prior years but was not. There is no obvious answer as to who should be responsible for the legacy unfunded liability, which ultimately comes down to a policy decision. The approach in PAEA spreads the cost of USPS's legacy unfunded liability over 50-plus years of then-future postal ratepayers. To illustrate the portion of prefunding requirements that are attributable to legacy costs, we found that across the four different prefunding approaches that we examined, legacy costs would account for anywhere from 39 percent to 53 percent of the prefunding requirement in fiscal year 2017, tapering down to anywhere from 8 percent to 18 percent by fiscal year 2040.[Footnote 41] Sensitivity to Assumptions: Measurements of actuarial costs and liabilities, as well as projections of such measures into the future, are subject to inherent uncertainty, and depend on a combination of economic and demographic assumptions as to future experience. Current law requires OPM to determine the value of USPS's retiree health benefit liability based on actuarial assumptions that are consistent with those used by OPM for its financial reporting of liabilities for federal employee benefits. These assumptions are to be used to determine USPS's funding requirements beginning in fiscal year 2017, when current law switches from a fixed-payment prefunding requirement to actuarially determined prefunding requirements. When the current law was enacted, this approach to selecting actuarial assumptions was consistent with the approach used by OPM for determining funding requirements for USPS's participation in the CSRS and FERS pension programs. In 2008, the Federal Accounting Standards Advisory Board (FASAB), which promulgates financial reporting standards for the federal government, issued Statement of Federal Financial Accounting Standards No. 33 (SFFAS 33), which, beginning in 2010, specified particular, new methodologies for the selection of economic assumptions for valuing various post-employment benefits for financial reporting purposes. [Footnote 42] As a result, the assumptions used by OPM for financial reporting for federal employee benefits--and by extension under the current law, for determining USPS's future prefunding requirements for retiree health care benefits--became different from the assumptions used by OPM to determine USPS's funding requirements for CSRS and FERS. The House and Administration Approaches retain the current law assumption basis for determining USPS's prefunding requirements; the Senate Approach would switch the determination of USPS's prefunding requirements to assumptions consistent with those now used for USPS's funding requirements for CSRS and FERS. The particular assumptions that differ are with respect to the interest rate (also known as the discount rate), the general inflation assumption, and the medical inflation (also known as the "trend") assumption. Table 5 shows the differences in these assumptions for the September 30, 2011, actuarial valuations performed by OPM, which served as the basis for our projections.[Footnote 43] Table 5: Key Differences between the Two sets of PSRHBF Funding Assumptions, as of September 30, 2011: Basis: Current law assumption basis (also used in House and Administration Approaches): Consistent with OPM's financial reporting of actuarial value of federal employee benefits; Senate bill assumption basis: Consistent with USPS's funding requirements for CSRS and FERS. Interest rate (also known as discount rate): Current law assumption basis (also used in House and Administration Approaches): 4.90; Senate bill assumption basis: 5.75. General inflation: Current law assumption basis (also used in House and Administration Approaches): 2.40; Senate bill assumption basis: 3.00. Medical inflation (also known as the "trend" assumption)[A]: Initial rate; Current law assumption basis (also used in House and Administration Approaches): 5.50; Senate bill assumption basis: 5.50; Medical inflation (also known as the "trend" assumption)[A]: Peak rate; Current law assumption basis (also used in House and Administration Approaches): 6.39; Senate bill assumption basis: 4.96. Medical inflation (also known as the "trend" assumption)[A]: Ultimate rate; Current law assumption basis (also used in House and Administration Approaches): 4.35; Senate bill assumption basis: 7.01; Source: GAO. [A] The trend assumption rises to a peak rate for 2015 and very gradually falls to an ultimate rate for 2084 and later. [End of table] Under SFFAS 33, the discount rate assumption should reflect average historical interest rates, over the prior 5 years or longer, on marketable Treasury securities with maturities consistent with the cash flows being discounted. The number of historical rates used in the calculation of this historical average should be consistent from year to year. OPM uses a 10-year historical averaging period.[Footnote 44] Further, the discount rate, the inflation assumption, and other economic assumptions should be consistent with one another.[Footnote 45] OPM determines the general inflation assumption under SFFAS 33 using the same 10-year historical averaging period that it uses in determining the discount rate. The selection of assumptions is also guided by relevant Actuarial Standards of Practice, which are promulgated by the Actuarial Standards Board. In contrast to the current law assumption basis that is now tied to a historical averaging period, the assumptions for determining USPS's funding requirements for CSRS and FERS represent OPM's estimate of future, long-term experience, informed by advice from an independent Board of Actuaries, and similarly guided by relevant Actuarial Standards of Practice. These standards too require that economic assumptions be consistent with one another. The relationship between the two assumption bases illustrated in table 5 is not static, so that the gap between the two assumption bases, and even which assumption base has higher rates, could change over time. It is important to note that assumptions that are tied to historical averages--as is the case under the current law assumption basis since the promulgation of SFFAS 33--can potentially diverge significantly from either current economic circumstances or from the current long- term economic outlook. The assumption criteria in SFFAS 33 were designed to accomplish financial reporting objectives rather than funding objectives. In selecting the medical inflation assumption, OPM relies on a model developed by the Society of Actuaries.[Footnote 46] This model ties medical inflation to the general inflation assumption (among other factors), so that a higher expected general inflation rate implies higher expected medical inflation. As mentioned earlier, we modeled all four modified prefunding approaches in two ways: first, as if they all used the current law assumption basis, and second, as if they all used the Senate bill assumption basis. Our findings and conclusions are not materially different under the two different assumption bases. Figure 8 compares the dollar amount of estimated USPS payments, in each fiscal year from 2013 through 2020, under the Modified Senate Approach to prefunding, using both the current law assumption basis and the Senate bill assumption basis. These dollar payment amounts differ by just 2 percent in aggregate over the period, and by not more than 4 percent in any particular year. Figure 8: Estimated Annual Payments for Modified Senate Prefunding Approach from Fiscal Year 2013 to Fiscal Year 2020 under Current Law Assumption Basis and Senate Bill Assumption Basis: [Refer to PDF for image: vertical bar graph] Fiscal year: 2013; Current law assumption basis: $4.432 billion; Senate bill assumption basis: $4.275 billion. Fiscal year: 2014; Current law assumption basis: $4.459 billion; Senate bill assumption basis: $4.319 billion. Fiscal year: 2015; Current law assumption basis: $4.545 billion; Senate bill assumption basis: $4.430 billion. Fiscal year: 2016; Current law assumption basis: $4.711 billion; Senate bill assumption basis: $4.612 billion. Fiscal year: 2017; Current law assumption basis: $4.902 billion; Senate bill assumption basis: $4.817 billion. Fiscal year: 2018; Current law assumption basis: $5.097 billion; Senate bill assumption basis: $5.024 billion. Fiscal year: 2019; Current law assumption basis: $5.305 billion; Senate bill assumption basis: $5.243 billion. Fiscal year: 2020; Current law assumption basis: $5.549 billion; Senate bill assumption basis: $5.495 billion. Source: GAO. [End of figure] Figure 9 compares estimated USPS payments under the Modified Senate Prefunding Approach, over the entire projection period from fiscal years 2013 through 2040, as a percentage of projected USPS annual compensation costs, again under both the current law assumption basis and the Senate bill assumption basis. The difference in the average payment percentage is just 0.7 percentage point, with the difference never exceeding 1 percentage point in any year. Figure 9: Estimated Annual Payments (as a Percentage of Compensation) for Modified Senate Prefunding Approach from Fiscal Year 2013 through Fiscal Year 2040 under Current Law Assumption Basis and Senate Bill Assumption Basis: [Refer to PDF for image: multiple line graph] Fiscal year: 2013; Current law assumption basis: $10.6%; Senate bill assumption basis: $10.1%. Fiscal year: 2014; Current law assumption basis: $10.8%; Senate bill assumption basis: $10%. Fiscal year: 2015; Current law assumption basis: $11%; Senate bill assumption basis: $10.6%. Fiscal year: 2016; Current law assumption basis: $11.4%; Senate bill assumption basis: $10.9%. Fiscal year: 2017; Current law assumption basis: $11.8%; Senate bill assumption basis: $11.2%. Fiscal year: 2018; Current law assumption basis: $12.2%; Senate bill assumption basis: $11.6%. Fiscal year: 2019; Current law assumption basis: $12.6%; Senate bill assumption basis: $11.9%. Fiscal year: 2020; Current law assumption basis: $13%; Senate bill assumption basis: $12.2%. Fiscal year: 2021; Current law assumption basis: $13.4%; Senate bill assumption basis: $12.6%. Fiscal year: 2022; Current law assumption basis: $13.8%; Senate bill assumption basis: $12.9%. Fiscal year: 2023; Current law assumption basis: $14%; Senate bill assumption basis: $13.2%. Fiscal year: 2024; Current law assumption basis: $14.2%; Senate bill assumption basis: $13.5%. Fiscal year: 2025; Current law assumption basis: $14.4%; Senate bill assumption basis: $13.8%. Fiscal year: 2026; Current law assumption basis: $14.6%; Senate bill assumption basis: $14%. Fiscal year: 2027; Current law assumption basis: $14.8%; Senate bill assumption basis: $14.2%. Fiscal year: 2028; Current law assumption basis: $15%; Senate bill assumption basis: $14.4%. Fiscal year: 2029; Current law assumption basis: $15.2%; Senate bill assumption basis: $14.6%. Fiscal year: 2030; Current law assumption basis: $15.3%; Senate bill assumption basis: $14.8%. Fiscal year: 2031; Current law assumption basis: $15.5%; Senate bill assumption basis: $14.9%. Fiscal year: 2032; Current law assumption basis: $15.6%; Senate bill assumption basis: $15%. Fiscal year: 2033; Current law assumption basis: $15.7%; Senate bill assumption basis: $15.1%. Fiscal year: 2034; Current law assumption basis: $15.8%; Senate bill assumption basis: $15.1%. Fiscal year: 2035; Current law assumption basis: $15.9%; Senate bill assumption basis: $15.2%. Fiscal year: 2036; Current law assumption basis: $15.9%; Senate bill assumption basis: $15.2%. Fiscal year: 2037; Current law assumption basis: $15.9%; Senate bill assumption basis: $15.2%. Fiscal year: 2038; Current law assumption basis: $15.9%; Senate bill assumption basis: $15.1%. Fiscal year: 2039; Current law assumption basis: $15.7%; Senate bill assumption basis: $14.8%. Fiscal year: 2040; Current law assumption basis: $15.5%; Senate bill assumption basis: $14.5%. Source: GAO. [End of figure] Because of the closeness of the results using the two assumption bases, we have chosen to present numerical results across all four prefunding approaches using the current law assumption basis in the main body of this report. Appendix II contains comparable numerical results using the Senate bill assumption basis. The primary reason for similar results under the two assumption bases is that the effects of differences in particular assumptions are offsetting to a certain extent. For example, the discount rate of 5.75 percent under the Senate bill assumption basis is more optimistic than the discount rate of 4.90 percent under the current law assumption basis. However, a higher discount rate suggests higher inflation and medical inflation; the higher medical inflation offsets much of the benefit of the higher discount rate. The two different inflation assumptions were also incorporated into our projection of USPS's annual compensation costs, which we extended based on a 10-year forecast of workforce and compensation provided to us by USPS. More information on these data and projections is provided in appendix I. We did not otherwise analyze variations in the workforce and compensation assumptions, as a more extensive analysis of assumption variations was beyond the scope of our study. As USPS notes correctly in its fiscal year 2011 Form 10-K report, "Because calculation of this [retiree health benefits] liability involves several areas of judgment, estimates of the liability could vary significantly depending on the assumptions used."[Footnote 47] In comparing the effects of the current law assumption basis versus the Senate bill assumption basis, we noted that differences in the discount rate and medical inflation assumptions have offsetting effects, so that the aggregate difference between the two assumption bases is not large. If, however, one of the assumptions were to change without an offsetting change in another assumption, the impact would be larger. OPM provided information on the sensitivity of the liability to variation in the medical trend alone, holding other assumptions constant. OPM's most recent measure of USPS's liability for retiree health benefits would have been 16 percent higher if the medical trend assumption had been one percentage point higher in all years (i.e., in table 5 above, 6.5 percent instead of 5.5 percent in the first year, etc.), and would have been 13 percent lower if the medical trend assumption had been one percentage point lower. Moreover, because the unfunded liability is equal to the difference between the liability itself and the amount of assets, a given percentage change in the liability can produce a larger percentage change in the unfunded liability. If the 2011 liability had been 16 percent higher, the unfunded liability would have been 31 percent higher; if the liability had been 13 percent lower, the unfunded liability would have been 26 percent lower. Thus, the $46 billion unfunded liability as of September 30, 2011, varies from $34 billion to $60 billion over this range of alternative assumptions. See table 6 below. Table 6: Sensitivity of Fiscal Year-end 2011 Unfunded Liability to One- Percentage-Point Change in Medical Inflation Assumption: Liability; Actual amount: $90.3; If medical inflation one percentage point higher: New amount: $104.4; Percent change: +16%; If medical inflation one percentage point lower: New amount: $78.5; Percent change: -13%. Assets; Actual amount: $44.1; If medical inflation one percentage point higher: New amount: $44.1; Percent change: 0; If medical inflation one percentage point lower: New amount: $44.1; Percent change: 0. Unfunded Liability = Liability - Assets; Actual amount: $46.2; If medical inflation one percentage point higher: New amount: $60.3; Percent change: +31%; If medical inflation one percentage point lower: New amount: $34.4; Percent change: -26%. Source: GAO. [End of table] Pay-as-You-Go Funding: Arguments have been made that requiring USPS to prefund its retiree health care benefits is unnecessary, unfair, or inadvisable, so we also examined the effects of a Pay-as-You-Go Approach. Under pay-as- you-go funding, each year USPS would only pay its share of premium payments for then-existing retirees and beneficiaries--there would be no prefunding. Given that money has already been prefunded in the PSRHBF, we first modeled a pay-as-you-go funding approach in which the fund would be drawn upon to pay USPS's share of premium payments for as long as possible. Under this approach, no additional contributions would be made to the fund, the fund would grow with interest, and USPS's share of premium payments for retirees and beneficiaries would be paid out of the fund until the fund was exhausted. Once the fund was exhausted, USPS would pay these premiums directly as they became due. Our projections show that, under either of the two sets of assumptions [Footnote 48]--the current law and Senate bill assumption bases--the PSRHBF would become exhausted in 14 years, in 2026. USPS would have zero reported costs for retiree health benefits until then. Beginning in 2026, USPS would begin paying its share of premium payments. By 2040, under the current law assumption basis, this annual cost is projected to be about $13 billion, not much different than the annual prefunding cost in fiscal year 2040 under the four different prefunding approaches. The big difference would be in the unfunded liability. Under this Pay-as-You-Go Approach, the unfunded liability in fiscal year 2040 would be about $250 billion, which would be about $130 billion in 2012 dollars, and about 310 percent of USPS's projected annual compensation cost. By comparison, under the modified Senate prefunding approach, which produces the largest unfunded liability of the four prefunding approaches, the unfunded liability in fiscal year 2040 would be about 85 percent of projected annual compensation cost. In summary, once the trust fund became exhausted, annual pay-as-you-go payments would not become significantly more onerous than annual prefunding payments, at least through the end of our projection period in fiscal year 2040. However, the Pay-as-You-Go Approach would produce a vastly bigger unfunded liability--which could eventually require an escalation of postal rates or reduction in costs. We examined a second variation of pay-as-you-go funding, an approach that the USPS OIG analyzed and reported on in February 2012.[Footnote 49] Under this approach, USPS would stop making prefunding payments and would pay its share of premium payments for retirees and beneficiaries as they become due. The existing fund would be left to grow with interest, with no other cash inflow or outflow. The intention would be for this to continue only until USPS's liability was fully funded.[Footnote 50] The USPS OIG has informally referred to this approach as the "Seal and Grow" Approach. The USPS OIG estimated that the fund would grow from $44 billion (its September 30, 2011, level) to $90 billion in 21 years. The USPS OIG did not estimate the liability or unfunded liability in 21 years, but noted that while the liability is not a static amount, and has risen over time historically, it had not changed significantly over the prior 3 years, going from $87 billion at fiscal year-end 2009 to $91 billion at fiscal year-end 2010 to $90 billion at fiscal year-end 2011.[Footnote 51] Some have concluded from this analysis that USPS's unfunded liability of $46 billion would be eliminated in 21 years by adopting this approach. However, our projections of the unfunded liability, which incorporate OPM's projections of the liability itself, show that the liability, in fact, would increase, resulting in a significant increase in the unfunded liability rather than its elimination. Specifically, we project that the unfunded liability would grow from $46 billion at fiscal year-end 2011 to $86 billion at fiscal year-end 2032 under this approach. The $86 billion estimate is equal to $53 billion in 2012 dollars and 139 percent of fiscal year 2032 compensation (up from 96 percent for fiscal year 2011).[Footnote 52] The USPS OIG's projection of assets - from $44 billion to $90 billion over 21 years - represents a 3.5 percent annual return over this period. Under our projection, using the current law assumption basis, assets grow from $44 billion to $120 billion over this period, at the assumed return of 4.9 percent, but the liability grows from $90 billion to $206 billion, or at an average rate of 4.0 percent per year. This projected liability growth reflects the net effect of accretions for interest, accretions for normal cost (with a reduced workforce), and reductions as premium payments are made, thereby discharging a portion of the liability. The projected liability would have to be 42 percent lower than projected for the unfunded liability to disappear by 2032. For this to occur (in the absence of cuts to benefits), future experience would have to be much more favorable than predicted by the assumptions. Nonetheless, under this Seal and Grow Approach the funded percentage is projected to improve over time. Because premium payments are projected to exceed normal cost for most of the projection period, the liability is projected to grow at a slower rate than assets (as noted in the preceding paragraph). As a result, the liability is projected to be 70 percent funded by 2040, close to the 73 percent projected funded percentage under the Modified Senate Approach.[Footnote 53] Thus, like the Modified Senate Approach, the Seal and Grow Approach is projected to result in a significant improvement in the funded percentage over the projection period, while still leaving a substantially larger unfunded liability relative to the Modified Current Law, House, and Administration Approaches. Moreover, USPS's payments under the Seal and Grow Approach would be more backloaded than under the Modified Senate Approach--with lower payments in the short term and higher payments later--making it more affordable in the short term but resulting in higher estimated unfunded liabilities in the short term as well. Key Considerations for Assessing Funding Approaches: To assist Congress in considering the various funding approaches, we identified some factors to consider in assessing what would constitute reasonable short-term and long-term funding requirements. We also examined the prefunding requirements of other organizations that offer retiree health benefits to their employees. Given that USPS is intended to be a self-sustaining entity funded almost entirely by postal revenue, we have previously stated that USPS should prefund its retiree health benefit liability to the maximum extent that its finances permit. Funding Approach Considerations: The following considerations should be taken into account when assessing the various funding approaches for USPS. Rationale for Prefunding: Consideration of whether to prefund retiree health benefits includes the associated consequences of the potential inability to fund the remaining unfunded liability or keep up with annual premium payments. In general, rationales for prefunding post-retirement benefits for any enterprise, whether for pension benefits or retiree medical benefits, can include the following: * Achieving an equitable allocation of cost over time by paying for retirement benefits during the employees' working years, when such benefits are earned. For USPS, the relevant cost allocation is between current and future postal ratepayers. The rationale is to have current ratepayers pay for the full cost of compensation for current employees, including the portion of such current compensation that is not paid until these current employees are retired. However, as noted earlier, an additional consideration is the "legacy" unfunded liability that was not paid by ratepayers in prior years. The conceptual rationale for prefunding does not answer the question of who should be responsible for a legacy unfunded liability. * Protecting the future viability of an enterprise by not saddling it with bills later after employees have already retired. In the case of USPS, this consideration is complicated by the organization's financial condition. * Providing greater benefit security to employees, retired employees, and their beneficiaries. Funded benefits protect against an inability to make payments later on, and can also make the promised benefits less vulnerable to cuts. In the private sector, failure to prefund retiree health benefits may have contributed to private employers terminating or reducing such benefits. In the state and local government sector, large unfunded liabilities for both retiree health and pension benefits have led to pressure and actions to trim the levels of these benefits. Others have contended that the mere requirement to account for the cost of these benefits in employers' financial reporting has led to benefits being cut. While an analysis of the cause of retiree health benefit cuts in other sectors is beyond the scope of our research, failure to prefund these benefits is a potential benefit security concern. * Providing security to any other party that might become responsible for part of the liability in the event of an enterprise's inability to pay for the remainder of the unfunded liability. For example, the Pension Benefit Guaranty Corporation is responsible for backing up private sector pension benefits when companies are unable to do so. According to the OPM OIG, the consequences if USPS could not pay for its retiree health benefits are unclear.[Footnote 54] Trade-offs: The effect of trade-offs among the different approaches on a number of issues would need to be considered, including trade-offs affecting: * USPS's financial condition. Protecting the future viability of USPS by not overwhelming it with bills and unfunded liabilities for the cost of employee benefits after these employees have already retired is complicated by the organization's immediate cash flow challenges including having reached the maximum of its borrowing authority. Prefunding payments under current law have contributed about $21 billion toward USPS's $25 billion of net losses over the past 5 years. If USPS continues to experience operational losses even before factoring in prefunding requirements, prefunding would add to such losses. As such, USPS would need to find larger cuts in operational costs now in order to have the cash to make its short-term prefunding payments. On the other hand, to the extent short-term prefunding payments are postponed, greater payments would be required later, supported by a smaller base of mail volume, with price caps further limiting revenue. Such a scenario would produce even greater pressure for cuts in operational costs later as well as raise concerns about USPS's ability to make prefunding payments, when unfunded liabilities would be greater because of the deferral of prefunding payments. USPS's OIG has stated that as an alternative to additional prefunding, USPS's extensive real estate holdings could provide collateral for the remaining unfunded liability.[Footnote 55] However, USPS has stated that it does not believe that USPS-occupied real estate would be a suitable asset within the PSRHBF because employer-occupied real estate cannot be readily sold to provide cash when needed to pay benefits.[Footnote 56] In addition, we would note that in the event of USPS's being unable to fund its liabilities, USPS might have other debts and obligations in addition to unfunded retiree health care liabilities for which any available real estate would be needed. Some comprehensive proposals to address USPS's financial condition have included provisions to transfer USPS's FERS pension surplus from the Civil Service Retirement and Disability Fund (CSRDF) to USPS; such a transfer could be viewed as a short-term source for some of the required PSRHBF prefunding payments. However, the most recent estimate of this surplus is significantly lower than the two prior estimates. We have previously reported on options and considerations with regard to this surplus.[Footnote 57] Use of any FERS surplus would not be a long-term solution to address USPS's financial outlook and operational imbalances. * Size of the annual payment and the unfunded liability. More near- term funding reduces payments and the amount of the unfunded liability later, while less near-term funding produces larger unfunded liabilities and requires higher funding payments later. The unfunded liability can also be viewed in a larger context. From fiscal years 2007 through 2010, USPS contributed a total of $17.9 billion to the PSRHBF. Over this same period, USPS increased its debt to the U.S. Treasury from $2.1 billion at fiscal year-end 2006 to $12.0 billion at fiscal year-end 2010, an increase of $9.9 billion.[Footnote 58] Thus, from fiscal year-end 2006 to 2010, USPS made payments to the PSRHBF of $17.9 billion while borrowing an additional $9.9 billion from U.S. Treasury. * Allocation of costs between current and future postal ratepayers. More near-term funding assigns more cost to current postal ratepayers that is reflected in rates, while less near-term funding assigns more cost to future ratepayers. As noted above, a complicating factor is the existing unfunded liability, which conceptually should have been paid by prior ratepayers but was not. Instead, this legacy cost is being spread among current and future ratepayers since fiscal year 2007. * Allocation of risks. Less prefunding now increases the risk that later some party(ies) could be called upon to pick up a greater share of the costs if USPS could not make its payments or pay off its unfunded liability. Another risk is that the level of employee pay and benefits may not be sustainable and could be reduced. As stated earlier, OPM's OIG reported that the exact consequences of these risks are unclear.[Footnote 59] Achieving reasonable postal rates from an equitable and consistent allocation of costs for pay and benefits earned during employees' work years could provide greater benefit security to employees, retirees, and beneficiaries. Fixed versus Actuarially Determined Payments: Another consideration with regard to the timing of prefunding payments is whether Congress wishes to continue requiring fixed prefunding contributions that are significantly in excess, through 2016, of actuarially-determined amounts. The House Bill largely retains this Current Law Approach, while the Senate Bill and the Administration Approach would produce a more consistent funding pattern. Targeted Funding Level: The Senate Bill targets an 80 percent funding level while the other approaches target a 100 percent funding level. The Senate committee report accompanying the Senate Bill stated that the committee set an 80 percent target-funding level on the presumption that USPS, if necessary, had additional assets it could draw upon to meet its liabilities.[Footnote 60] As previously stated, USPS's OIG report stated that USPS's extensive real estate holdings could provide collateral for the remaining unfunded liability, but we would note that in the event of USPS's being unable to fund its liabilities, USPS might have other debts and obligations in addition to an unfunded retiree health benefit liability for which any available real estate would be needed. If an 80 percent funding target level were selected because of concerns about USPS's ability to achieve a 100 percent target level within a particular time frame, an additional option could be to build in a schedule to achieve 100 percent funding in a subsequent time period after the 80 percent level is achieved. [Footnote 61] Basis for Assumptions: As discussed earlier, the issuance of SFFAS 33 had the effect of creating a divergence between the actuarial assumptions used in determining USPS's funding requirements for PSRHBF and those used in determining its funding requirements for CSRS and FERS. Another consideration is whether Congress desires more uniform funding assumptions across these programs. As noted, the funding assumptions for PSRHBF under current law, which are retained in the House Bill and Administration Approach, are now, post-SFFAS 33, based on 10-year historical averages. Assumptions that are based on historical averages can potentially diverge significantly from either current economic circumstances or from the current long-term economic outlook. The assumption criteria in SFFAS 33 were designed to accomplish financial reporting objectives rather than funding objectives. Comparison with Other Entities: We also reviewed the prefunding requirements for other organizations that offer retiree health benefits to their employees: private sector entities, state and local governments, and other federal entities. Although other federal, state and local, and private sector entities generally are not required to prefund retiree health care benefits, a few do prefund at limited percentages of their total liability. However, most are required to recognize the future costs of these benefits in their financial reporting if they follow generally accepted accounting principles. Although recognizing the cost of retiree health benefits for financial reporting purposes is a separate issue from the question of whether to prefund these benefits, such reporting does enhance the transparency of the cost of these benefits. USPS accounts for these benefits using private-sector multiemployer accounting rules, under which USPS does not recognize the unfunded liability for these benefits on its balance sheet.[Footnote 62] In 2002, GAO suggested that USPS reconsider its method of accounting for these benefits.[Footnote 63] In addition, although prefunding is not required, a number of private, state, local, and federal entities have elected to prefund some percentage of their retiree health benefits. For example, Standard & Poor's (S&P) reported that 126 of the 296 companies in the S&P 500 that offered "other post-employment benefits" (OPEB)[Footnote 64] prefunded some percentage of the associated liabilities, while the USPS OIG has reported that 38 percent of Fortune 1000 companies that offer retiree health care benefits prefund them, at a median funding level of 37 percent.[Footnote 65] Further, in November 2009, we found that 18 states and 13 of the 39 largest local governments had set aside at least a combined $25 billion in assets to cover their OPEB liabilities.[Footnote 66] Although the majority of federal civilian agencies do not prefund these benefits, a few small, civilian, federal agencies do so. In addition, the Department of Defense (DOD) prefunds its retiree health benefits for Medicare-eligible retirees and beneficiaries, with a 100 percent target funded percentage. This fund was started in 2002 in reaction to rapidly rising health care costs. The fund had assets of $166 billion as of September 30, 2010, which represented a funding level of 38 percent.[Footnote 67] DOD does not prefund its pre- Medicare-eligible retiree health benefits, although its independent Board of Actuaries has recommended that it consider prefunding these costs as well, in order to reflect the full costs of these future benefits and promote a better understanding of the program's value. [Footnote 68] While private sector, state and local government, and other federal entities generally are not required to prefund these benefits, most are required to recognize the future costs of these benefits on an accrual basis as they are earned, rather than when they are paid, in their financial reporting. Standards governing financial reporting (i.e., accounting) are separate and apart, and under different jurisdiction, from any laws, regulations, or rules governing prefunding.[Footnote 69] In contrast to most other federal entities, USPS reports under private sector (FASB) accounting standards, and follows FASB's multiemployer accounting rules, rather than FASB's single-employer accounting rules, in reporting its participation in FEHBP. These multi-employer standards exempt employers from reporting the cost of these retirement benefits on an accrual basis. Instead, expense for a year is set equal to required cash payments--which currently for USPS means the sum of its required prefunding payment and its share of premium payments that it pays directly--while no liability is shown on the USPS balance sheet except for any required payments that have been missed (such as the missed fiscal year 2011 and 2012 prefunding payments). In contrast, if USPS were following FASB's single-employer accounting standards, USPS would show a liability on its balance sheet for the entire unfunded liability, and expense for a year would be an actuarially determined accrual cost independent of whether USPS had to make a small or large prefunding payment for that year. In 2002, the Comptroller General wrote to the Postmaster General and, based on a reassessment of the applicability of multiemployer versus single- employer accounting standards to USPS, suggested that USPS reassess its accounting treatment of retiree health benefits, and consider accounting for its retiree health benefits on an accrual basis, meaning, to consider adopting the single-employer accounting procedures. A basic premise behind the exemption from accrual accounting for multiemployer plans was that the liability for an individual employer would be difficult to determine and would be of limited value, a premise that is not the case for USPS.[Footnote 70] Recognizing the cost of retiree health benefits on an accrual basis for financial reporting enhances the transparency of the cost of these benefits even in the absence of prefunding. As a result, in situations where prefunding requirements do not exist or are significantly relaxed or eliminated, accrual accounting provides an important function by recognizing the costs of these future benefits even in the absence of prefunding. It can also be the case that a year's accrual cost can be lower than the amount funded. For example, the fixed payments required under current law may well be higher than the annual accrual cost that USPS would recognize under single employer accounting, although, again, the full unfunded liability would be recognized on the balance sheet. Note that there is one other significant program, federal workers' compensation under the Federal Employees' Compensation Act, for which USPS's financial reporting is based on actuarial projections of future benefits rather than on its annual required cash payments. USPS pays the Department of Labor each year for the cash benefits to current beneficiaries, but USPS records a liability on its balance sheet for the entire actuarial present value of future benefits for those who have already been injured, and recognizes the growth in this liability as an expense each year. This unfunded FECA liability on USPS's balance sheet was $17.6 billion as of September 30, 2012. Concluding Observations: Timely action is essential in addressing the funding of USPS's retiree health benefits. We have suggested that Congress must take action to address the uncertainty related to: 1) USPS's inability to meet the current retiree health prefunding requirements, 2) reducing the unfunded retiree health benefit liability over time, 3) determining the proper allocation of costs between current and future ratepayers, and: 4) enacting comprehensive postal reform legislation that would improve prospects for USPS's long-term financial viability. USPS's recent defaults on its retiree-health-prefunding payments and its inability to borrow now that it has reached its $15 billion borrowing limit create an even more urgent need for congressional action. The continued uncertainty around resolution of USPS's financial problems and the funding of these payments creates uncertainty for mailers in developing their business plans, an uncertainty that could negatively affect mailers' willingness to use USPS's services. As noted earlier, USPS has also proposed withdrawing from FEHBP and administering its own health care plan for both workers and retirees, a proposal that is the subject of our ongoing work in another study. Congress should also consider how quickly and to what level prefunding of retiree health benefits should occur. As previously cited, deferrals and lower payments in the short-term will reduce USPS's reported financial losses in the short-term, but would increase its unfunded retiree health benefit liability and require larger annual payments in the future; yet at the same time, currently required short- term payments are higher than what would be required under the actuarial approach that begins in 2017. Both of these points raise issues regarding fairness to future and current ratepayers. Furthermore, postal ratepayers provide USPS with funding, but as mail volumes decline, there may be fewer ratepayers in the future to pay for deferred costs. In addition, the less USPS reduces its retiree health unfunded liability, the greater the potential consequences, with unclear impact, if USPS is ultimately unable to pay this unfunded liability. In considering the options for USPS to address its retiree health benefit liability, Congress should keep in mind that stopping or deferring prefunding of these benefits would serve as short-term relief, but would also increase the risk that USPS may not be able to make future payments if its core business continues to decline. Therefore, we continue to believe it is important that USPS prefund its retiree health benefit liability to the maximum extent that its finances permit. None of the funding approaches will be viable unless USPS has the ability to make the required payments. Without congressional or further USPS actions to cut postal costs, USPS will not have the finances needed to make annual payments in the short term and reduce its retiree health unfunded liability over the long term. USPS has stated that it will be unable to make any prefunding payment toward reducing its retiree health unfunded liability if it continues to experience cash flow difficulties. While USPS may have limited control of its revenue stream because of advances in technological communication, it is important that USPS reduce its expenses to avoid even greater financial losses, repay its outstanding debt, and increase capital for investment. Consequently, as we have repeatedly stated, Congress and USPS need to reach agreement on a comprehensive package of actions to improve USPS's financial viability. In previous reports, we have provided strategies and options, to both reduce costs and enhance revenues, that Congress could consider to better align USPS costs with revenues and address constraints and legal restrictions that limit USPS's ability to reduce costs and improve efficiency. Implementing strategies and options to better align costs with revenues may better enable USPS to be in a financial position to prefund its retiree health benefit liability for its over one million active and retired postal employees and their beneficiaries. Agency Comments and Our Evaluation: We provided a draft of this report to USPS, the USPS OIG, and OPM for review and comment. USPS and the USPS OIG provided comments, which are reprinted in appendixes III and IV, respectively. USPS and the USPS OIG did not disagree with the report's conclusions and analysis about the trade-offs involved with the alternative funding approaches, but both commented that USPS cannot afford to make prefunding payments and provided additional context. OPM had no comments but provided technical clarifications, which we incorporated into the report as appropriate. USPS agreed that comprehensive reform is necessary to achieve financial sustainability. It also recognized its obligation to provide effective, affordable health benefits to its employees and retirees, but said that it does not have the financial resources to make prefunding payments required by current law. Further, USPS said that releasing this report is inappropriate because, in its view, the solution to managing its health care costs is to reduce the cost of future health care coverage by allowing USPS to sponsor its own medical plan. In response to USPS's comment, we noted in the report that adopting any of the prefunding approaches analyzed in this report would not preclude USPS from continuing to pursue its proposal to administer its own plan, and that any resulting expected cost savings would be reflected in a lower unfunded liability and lower actuarially determined prefunding payments than otherwise. As USPS noted, we are currently reviewing USPS's proposal to administer its own plan. The USPS OIG concurred with our analysis of the trade-offs among the alternative funding approaches that would result in paying more now or in the future, but stated its concern that the report needed additional context in four areas: 1) historical, 2) financial, 3) use of other assets to satisfy the retiree health benefit obligation, and (4) the problems with prefunding. First, the USPS OIG stated that USPS started prefunding its retiree health benefits as a result of the discovery that, because of external fund management misjudgments, it was on track to seriously overfund its pension obligations by $78 billion. The USPS OIG also said that a decision to turn a mistake into a second prefunding obligation created its own problems, including a 10-year schedule of prefunding payments that was structured toward a 100 percent funding goal, and that the aggressive payment schedule appears to have been set based on byzantine "budget scoring" considerations rather than actuarial assumptions or an evaluation of USPS's ability to make the payments. In our report, we noted USPS's reduction in pension contributions to the Civil Service Retirement System occurred as a result of the Postal Civil Service Retirement System Funding Reform Act of 2003, which switched the actuarial basis for future contributions to "dynamic" assumptions from the "static" assumptions that OPM projected would result in overfunding. Further, we pointed out that the 10-year schedule of prefunding payments for fiscal years 2007 through 2016 was not based on an actuarial assessment, and that the remaining required payments through fiscal year 2016 are significantly in excess of what would be calculated under the actuarial approach that begins in fiscal year 2017. We also noted that USPS proposed prefunding to Congress in 2003. Second, the USPS OIG discussed several points in a financial context. It said that USPS has never been able to afford a single payment--that it has either borrowed from the U.S. Treasury to make prefunding payments to date or that it has defaulted on them. However, we noted in our report that from fiscal year-end 2006 to 2010, USPS made total prefunding payments of $17.9 billion while borrowing an additional $9.9 billion from the U.S. Treasury. The USPS OIG also stated that now that the USPS has reached the limit of the amount it can borrow, it can no longer make the payments. We noted in our report that none of the funding approaches will be viable unless USPS has the ability to make the required payments, and that a comprehensive package of actions is needed to improve USPS's financial viability. The USPS OIG also said that its "seal and grow" proposal was made in the context of USPS's urgent financial situation and was meant as a temporary--not permanent--measure, and that we mistakenly represented it as a permanent payment plan. Our report actually noted that the Seal and Grow Approach was intended to continue until USPS's liability was fully funded--meaning, not thereafter; we added additional wording to clarify this point. The USPS OIG also pointed out that USPS has substantially funded its retiree benefit programs, with its pensions fully funded and its retiree health benefits half funded, with enough to cover current retirees. We did note in our report that the retiree health benefit liability is 49 percent funded and that approximately half of the liability is for current retirees. As for pensions, USPS reported in its most recent annual financial report (10-K) for fiscal year 2012 that it had an unfunded pension liability of almost $16 billion, which represented a 95 percent funded percentage (i.e., close to fully funded), based on a projected year-end fund balance of $285 billion and a liability of $300 billion; the prior year's estimate had indeed been a pension surplus. Third, the USPS OIG stated that our report did not adequately explore the use of other assets USPS holds as a means of satisfying its retiree health benefit obligation. The USPS OIG noted that it has reported on two sources of assets worth billions that could be used to cover any unfunded obligation, including 1) an estimated $85 billion in real estate holdings and 2) surpluses in USPS's pension funds. As we noted in our report, USPS has stated that it does not believe that USPS-occupied real estate would be a suitable asset within the PSRHBF because employer-occupied real estate cannot be readily sold to provide cash when needed to pay benefits.[Footnote 71] We noted that in the event of USPS's being unable to fund its liabilities, USPS might have other debts and obligations in addition to unfunded retiree health benefit liabilities for which any available real estate proceeds would be needed. We noted that we reported on options and considerations with regard to any USPS pension surplus (in particular regarding FERS) in a prior report. Finally, the USPS OIG commented that our report should examine the problems of prefunding and examine why no business or government entity has taken advantage of prefunding, and that making prefunding payments at the current levels will bankrupt USPS. Our report did discuss these issues, beginning with the section entitled, Comparison with Other Entities. While our report did not examine comprehensively the reasons for other entities' prefunding decisions, we noted that although prefunding is not required, a number of private, state, local, and federal entities have elected to prefund some percentage of their retiree health benefits, as follows: * Standard & Poor's (S&P) reported that 126 of the 296 companies in the S&P 500 that offered "other post-employment benefits" (OPEB) prefunded some percentage of the associated liabilities; * the USPS OIG reported that 38 percent of Fortune 1000 companies that offer retiree health benefits prefund them, at a median funding level of 37 percent; * 18 states and 13 of the 39 largest local governments had set aside at least a combined $25 billion in assets to cover their OPEB liabilities; and: * the Department of Defense prefunds its retiree health benefits for Medicare-eligible retirees and beneficiaries, with a 100 percent target funding percentage, and that this fund, which was started in 2002 in reaction to rapidly rising health care costs, had assets of $166 billion as of fiscal year-end 2010. We also recognized USPS's inability to meet the current retiree health prefunding requirements along with the need for comprehensive legislative action. Specifically, we said, "None of the funding approaches will be viable unless USPS has the ability to make the required payments. Without congressional or further USPS actions to cut postal costs, USPS will not have the finances needed to make annual payments in the short term and reduce its retiree health benefit liabilities over the long term." As arranged with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days after its date. At that time, we will send copies of this report to the appropriate congressional committees, the Postmaster General, OPM, the USPS Inspector General, and other interested parties. In addition, the report will be available at no charge on GAO's Web site at [hyperlink, http://www.gao.gov]. If you or your staffs have any questions on this report, please contact Frank Todisco at todiscof@gao.gov; Lorelei St. James at stjamesl@gao.gov: or call (202) 512-2834. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Contact information and key contributors to the report are listed in appendix V. Sincerely yours, Signed by: Frank Todisco: Chief Actuary: Applied Research and Methods: Signed by: Lorelei St. James: Director: Physical Infrastructure Issues: The undersigned meets the qualification standards of the American Academy of Actuaries to render the actuarial findings contained in this report. Signed by: Frank Todisco, FSA, MAAA, EA: Chief Actuary: [End of section] Appendix I: Objectives, Scope, and Methodology: This report (1) describes the status and financial outlook of the Postal Service Retiree Health Benefits Fund (PSRHBF), (2) analyzes how alternative proposals for funding retiree health benefits could affect future USPS payments and unfunded liabilities, and (3) determines key considerations for policymakers assessing the alternative proposals or other approaches. To describe the status and financial outlook of the PSRHBF, we reviewed and summarized USPS financial data regarding payments made to the fund, interest earned from such contributions, overall fund balance, and retiree health benefit liability. We also reviewed our prior work and reviewed and summarized reports and data from USPS and others on how USPS's financial condition has changed since 2006. We reviewed relevant statutes, proposed legislation, and sections of the President's budget request for fiscal year 2012 pertaining to USPS's health and pension benefit programs. We also interviewed USPS and OPM officials on the status and financial outlook of the PSRHBF. To determine the impact on USPS payments and unfunded liabilities under alternative approaches to fund retiree health benefits, we analyzed and compared current funding requirements and five alternatives. We interviewed USPS officials on the USPS's ability to meet future mandated payments and to obtain information on current and projected employee (FTE) levels, compensation, and revenue. In addition, we met with OPM officials to discuss projection methodology, and assumption selection, for using the data provided by USPS to project future premium payments, normal costs, and liabilities. OPM provided us projections of these amounts, which we further analyzed to project future prefunding contributions and unfunded liabilities under the different approaches to prefunding that we analyzed. Additional information on data, assumptions, and methods is provided below. To determine key factors for policymakers to consider when assessing alternative approaches, we used our own actuarial judgment and expertise. We also examined prefunding requirements for retiree health benefits, and prefunding behavior, of other entities (federal, state, and local governments and private sector). In addition, we examined financial reporting requirements applicable to other entities for these benefits, reviewing relevant accounting standards promulgated by the Financial Accounting Standards Board (FASB), Governmental Accounting Standards Board (GASB), and Federal Accounting Standards Advisory Board (FASAB); we compared these standards to USPS's financial reporting for these benefits. Additional Information on Data, Assumptions, and Methods Underlying the Actuarial Projections: We relied on OPM's actuarial projections of normal cost, accrued liability (referred to in the report, and in the remainder of this appendix, simply as "liability") and premium payments. We obtained data on workforce projections from USPS, as described further below, which we projected further and supplied to OPM for use in the projections. OPM's valuation of the cost of USPS's retiree health benefit obligations entails the collection and analysis of participant data and claims cost data, the setting of demographic and economic assumptions, and the application of these data and assumptions to the provisions of the benefit program. We had extensive discussions with OPM regarding its valuation methodology and were satisfied with the reasonableness of the approach with regard to the issues discussed. However, we did not otherwise audit or evaluate OPM's actuarial assumptions, methodology, calculations, or underlying data. Such an evaluation would have required a substantial amount of additional work beyond the scope of our assignment, and would also have required engaging additional actuarial resources with particular expertise in the valuation of health care benefits. For projecting the most recent valuation results into the future, we selected the methodology and projection assumptions in consultation with OPM. Additional detail on OPM's methods and assumptions is available from OPM. It should be understood that projections of this type contain a significant degree of uncertainty, as discussed further in the section of the report on Sensitivity to Assumptions. Nonetheless, given the magnitude of the liabilities and the importance of being able to pay for these benefits, reasonable projections of these costs and liabilities provide essential information for enabling responsible stewardship of resources. OPM provided us with projected normal cost and premium payments for each year through 2040. OPM calculated and provided us with projected liability as of three points: the end of 2010 (the measurement date of the most recent data collection at the time of our request), the end of 2021, and the end of 2040. We used a linear interpolation to estimate the liability for each of the intervening years. For each future year, we calculated the prefunding contribution, based on the normal cost and unfunded liability, when an actuarial approach applied; rolled the assets forward by adding the prefunding contribution and investment income and subtracting premium payments, as applicable; calculated the next year's unfunded liability based on these projected assets and the projected liability for that year; calculated the next year's prefunding contribution based on this new unfunded liability; and so on to the end of the projection period. The calculation of the prefunding contribution--as well as the applicability of fixed versus actuarially determined contributions and whether premium payments came out of the fund--was based on the provisions of the prefunding approaches we modeled, as described in the main body of this report (table 2 and preceding text). Where an actuarially determined prefunding contribution was used, it was the sum of the normal cost and an amortization payment (mortgage-style amortization calculation) calculated to pay off the unfunded liability in equal annual installments. Note that under the terms of the Senate Bill, which uses an 80 percent funded percentage target instead of 100 percent, the amortization is based on [(0.80*Liability) - Assets], rather than [0.80*(Liability - Assets)], and 100 percent of the normal cost is added to the amortization payment, rather than 80 percent of the normal cost. OPM's projections of liabilities are based on the current level of plan health benefits and do not reflect any proposals to reduce the actuarial value of benefits. USPS has proposed withdrawing from the Federal Employees Health Benefits Program (FEHBP) and administering its own health care plan for its employees and retirees. This report looks at retiree health benefit funding options assuming that USPS continues to participate in FEHBP under current provisions. We will be issuing a separate report on USPS's proposal to administer its own health care plan. OPM's projections also reflect the projected changes over time in the U.S. Treasury's share of USPS's retiree health benefit costs. U.S. Treasury is responsible for the portion of USPS's share of retiree health benefit premiums attributable to service prior to 1971, when the Post Office Department was transformed into the USPS. The U.S. Treasury's share of costs is diminishing over time as the proportion of retirees who had pre-1971 service decreases. One of the factors affecting future changes in USPS's liability for retiree health benefits is the size of its future workforce. The liability grows with future accruals of employee service and is also affected by when employees retire. USPS provided us with projected counts of career employees from 2011 through 2020. USPS noted that its intermediate-term planning horizon was through 2016 and that because of the rapidly changing nature of the mailing environment and the overall economy, projections beyond that point are likely to have a higher margin of error. USPS's projection had its career-employee complement dropping, from 561,000 in 2011 and 534,000 in 2012 (representing approximate averages over the fiscal year) to approximately 416,000 by 2016 and to 392,000 by 2020. USPS told us that it would be reasonable to assume that the complement would stabilize at that level thereafter. We assumed a constant career workforce of 392,000 for the remainder of the projection from 2020 through 2040. USPS viewed this workforce projection as its optimal, target workforce path, assuming USPS would be able to achieve certain objectives regarding its network and other operational issues. It noted that its ability to achieve these reductions remains to be determined, and would be affected by negotiations with unions and any congressional actions. USPS also noted that its workforce projections were based on long-term projections of mail volume. There is, of course, uncertainty regarding future levels of mail volume. OPM found that using its standard valuation assumptions for such factors as employee retention and retirement, and adding in an amount of new hires necessary to stay on target, its projection model reasonably approximated USPS's projected workforce path. Based on this workforce path and the number of projected retirements and other workforce reductions, OPM projected some new hiring to begin in 2014, and to continue as necessary to keep the workforce constant after 2020. OPM based new hire demographic profiles on the government-wide distribution of recent hires, since USPS has not been hiring enough recently to have adequate data for that purpose. So that we could also calculate USPS payments and unfunded liabilities as a percentage of employee compensation, USPS provided us with projections of compensation (salary and wages and benefits) to accompany the workforce projections, through 2020. The data provided by USPS encompassed salary plus a portion of employee benefits; it did not include retiree health benefits, worker's compensation, or any forecasted contract negotiations savings. For simplicity, we refer to these amounts as "compensation." We projected these compensation amounts beyond 2020 to 2040. Since we assumed the USPS workforce to be constant over that period, we projected total compensation to increase by inflation plus one percent. USPS had provided us with two sets of compensation projections through 2020: one based on USPS's own internal inflation assumption ranging from 1.7 to 2.2 percent annually over that period; and a second, at our request, assuming 3.0 percent inflation. We estimated an additional compensation projection based on 2.4 percent inflation from these data. We used the two sets of compensation projections--one based on 2.4 percent inflation and one based on 3.0 percent inflation--for our projections under the current law assumption basis and the Senate bill assumption basis, respectively. Liabilities and normal costs are based on the "Aggregate Entry Age Normal" actuarial cost method. A per-participant normal cost rate is determined based on an aggregate ratio of present value of future benefits at entry age to present value of future service at entry age, with service weighted to increase with medical inflation and with the accrual period from entry age to assumed retirement. The normal cost rate is computed based on the demographics and claims' costs of the entire FEHBP population, not just the USPS population, to reflect how the plan actually works. OPM would need additional USPS-specific data to determine a USPS-specific normal cost. The accrued liability is equal to the present value of future benefits (PVB) minus the present value of future normal costs. The PVB is just for the USPS population, but based on demographic assumptions for the entire FEHBP population, and without USPS-specific utilization, as this is how FEHBP premiums are determined. The actuarial cost method is the same one used by OPM in its financial reporting of the cost of these benefits (as required under FASAB accounting standards) and the same one used by OPM for determining funding requirements for the CSRS and FERS federal employee pension programs. Other actuarial cost methods could reasonably be adopted[Footnote 72] for determining USPS prefunding requirements, such as the projected unit credit method (which is also the method used for single employer accounting under FASB). The actuarial cost method determines the portion of future retiree costs that are attributable to each year of employee service, and different methods build up the accrued liability more or less quickly over the working years. As discussed in the body of the report, OPM provided current and projected liabilities, normal costs, and premium payments on two different assumption bases: (1) the current law basis, which ties funding assumptions to those used by OPM for its financial reporting, which in turn is guided by the FASAB accounting standards and (2) the Senate bill basis, which ties funding assumptions to those assumptions used by OPM to determine USPS's funding requirements for CSRS and FERS. The assumptions differ with respect to discount rate, general inflation, and medical inflation (trend). These assumptions are disclosed in table 5 in the report. Demographic assumptions that are common to both the current law and Senate bill assumption bases can be found in OPM's most recent funding valuation report for CSRS and FERS, [Footnote 73] though these are applied on a per-participant basis in the retiree health valuation and on a dollar-amount basis in the pension valuations. OPM also assumes that present retiree participation rates in FEHBP, calculated by age and gender, continue into the future. The discount rate of 4.90 percent used for the current law assumption basis, which is the discount rate used by OPM in its reporting at September 30, 2011, represents the single rate equivalent to a 10-year average of Treasury yield curves, with yield curve maturities matched to the timing of projected payments, a methodology that satisfies SFFAS 33. We assumed that the discount rate would remain at 4.90 percent in future years. In fact, in each future year, a new 10-year average discount rate will be developed, and if interest rates were to remain unchanged from present levels, this would result in a lower future discount rate, as higher interest rates at the beginning of the 10-year averaging period are replaced by lower interest rates at the end of the averaging period. OPM indicated that modeling such changes would present significant computational difficulties.[Footnote 74] Accordingly, we used a steady 4.90 percent discount rate, which implies that interest rates would rise from current low levels. In making this assumption, we noted that a steady 4.90 percent discount rate is still significantly lower than the 5.75 percent discount rate assumed for the Senate bill assumption basis, and so still provides useful information regarding potential effects of variations in assumptions. We also note that the medical inflation assumption used in the projection was developed to be consistent with the discount rate and general inflation assumption (the latter is also based on a 10-year average), and that OPM's model would produce a lower medical trend assumption if discount rates and inflation assumptions were to decrease, offsetting much of the effect of the lower discount rate. Also, in projecting future premium payments, which went into projecting future liabilities, OPM did not "restart" the trend assumption vector each year. A final methodological decision that had to be made was whether the projection assumptions should differ from the valuation assumptions. In actuarial projections, there is a distinction between "valuation assumptions" and "projection (or experience) assumptions." Valuation assumptions are those used to compute the liability and normal cost at any point in time. Projection assumptions model what actually happens as you move the projection forward, which might differ from the expectations embedded in the valuation assumptions. The Senate Bill specifies a different assumption basis than current law, the House Bill, or the Administration's Approach, but these specifications are referring to valuation assumptions. While different valuation assumptions might be used, only one scenario can actually unfold in the real world. One way to reflect this situation in a projection would be to retain the different valuation assumptions for the different prefunding approaches, but then to project all the approaches under a uniform set of projection assumptions. However, this approach would create false precision, because at some point the valuation assumptions would change to reflect emerging experience, and the projection would then need to incorporate additional assumptions as to when that would happen. Accordingly, as a reasonable approach to compare the four prefunding approaches on an apples-to-apples basis, we modeled them under uniform assumptions--first using the current law assumption basis, with results presented in the main body of the report, and then using the Senate bill assumption basis, with results presented in appendix II of this report. As discussed further in the section of the report on "Sensitivity to Assumptions," it turns out that these two assumption bases do not produce significant differences in basic findings because of the offsetting effects of different discount rates and medical inflation assumptions. We conducted this performance audit from May 2012 through December 2012 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. [End of section] Appendix II: Projected Results Using Senate Bill Assumption Basis: As discussed in the report, we projected USPS's annual payments and unfunded liability under four prefunding approaches (Modified Current Law, Modified House, Modified Administration, and Modified Senate) and under two different assumption bases: the assumption basis specified in current law, and the assumption basis specified in the Senate bill. The differences between these two assumption bases are described in the report. The report presents projection results based on the current law assumption basis. This appendix presents the corresponding results based on the Senate bill assumption basis. As discussed in the report, our findings and conclusions are not materially different under the two different assumption bases. Short-term Effects under Senate Bill Assumption Basis: Table 7: Estimated Annual Payments and Unfunded Liabilities in the Short Term (Fiscal Years 2013 to 2020) under Senate Bill Assumption Basis: Fiscal year: 2013; Modified Current Law Approach: Annual Payments: $8.6 billion; Unfunded Liability (at Year-end): $44.9 billion; Modified House Approach: Annual Payments: $8.6 billion; Unfunded Liability(at Year-end): $44.9 billion; Modified Administration Approach: Annual Payments: $1.3 billion; Unfunded liability (at Year-end): $52.3 billion; Modified Senate Approach: Annual Payments: $4.3 billion; Unfunded liability (at Year-end): $49.3 billion. Fiscal year: 2014; Modified Current Law Approach: Annual Payments: $8.9 billion; Unfunded Liability (at Year-end): $42.0 billion; Modified House Approach: Annual Payments: $8.9 billion; Unfunded Liability(at Year-end): $42.0 billion; Modified Administration Approach: Annual Payments: $5.4 billion; Unfunded liability (at Year-end): $53.4 billion; Modified Senate Approach: Annual Payments: $4.3 billion; Unfunded liability (at Year-end): $51.4 billion. Fiscal year: 2015; Modified Current Law Approach: Annual Payments: $9.2 billion; Unfunded Liability (at Year-end): $38.5 billion; Modified House Approach: Annual Payments: $11.5 billion; Unfunded Liability(at Year-end): $36.3 billion; Modified Administration Approach: Annual Payments: $5.6 billion; Unfunded liability (at Year-end): $54.4 billion; Modified Senate Approach: Annual Payments: $4.4 billion; Unfunded liability (at Year-end): $53.4 billion. Fiscal year: 2016; Modified Current Law Approach: Annual Payments: $9.7 billion; Unfunded Liability (at Year-end): $34.5 billion; Modified House Approach: Annual Payments: $11.9 billion; Unfunded Liability(at Year-end): $29.8 billion; Modified Administration Approach: Annual Payments: $5.8 billion; Unfunded liability (at Year-end): $55.2 billion; Modified Senate Approach: Annual Payments: $4.6 billion; Unfunded liability (at Year-end): $55.3 billion. Fiscal year: 2017; Modified Current Law Approach: Annual Payments: $4.9 billion; Unfunded Liability (at Year-end): $35.0 billion; Modified House Approach: Annual Payments: $4.6 billion; Unfunded Liability(at Year-end): $30.4 billion; Modified Administration Approach: Annual Payments: $6.3 billion; Unfunded liability (at Year-end): $55.6 billion; Modified Senate Approach: Annual Payments: $4.8 billion; Unfunded liability (at Year-end): $57.2 billion. Fiscal year: 2018; Modified Current Law Approach: Annual Payments: $5.1 billion; Unfunded Liability (at Year-end): $35.5 billion; Modified House Approach: Annual Payments: $4.8 billion; Unfunded Liability(at Year-end): $30.9 billion; Modified Administration Approach: Annual Payments: $6.5 billion; Unfunded liability (at Year-end): $55.9 billion; Modified Senate Approach: Annual Payments: $5.0 billion; Unfunded liability (at Year-end): $59.0 billion. Fiscal year: 2019; Modified Current Law Approach: Annual Payments: $5.3 billion; Unfunded Liability (at Year-end): $35.8 billion; Modified House Approach: Annual Payments: $5.0 billion; Unfunded Liability(at Year-end): $31.2 billion; Modified Administration Approach: Annual Payments: $6.7 billion; Unfunded liability (at Year-end): $56.1 billion; Modified Senate Approach: Annual Payments: $5.2 billion; Unfunded liability (at Year-end): $60.7 billion. Fiscal year: 2020; Modified Current Law Approach: Annual Payments: $5.6 billion; Unfunded Liability (at Year-end): $35.9 billion; Modified House Approach: Annual Payments: $5.3 billion; Unfunded Liability(at Year-end): $31.5 billion; Modified Administration Approach: Annual Payments: $6.9 billion; Unfunded liability (at Year-end): $56.1 billion; Modified Senate Approach: Annual Payments: $5.5 billion; Unfunded liability (at Year-end): $62.4 billion. Total Payments; Modified Current Law Approach: Annual Payments: $$57.3 billion; Unfunded Liability (at Year-end): NA; Modified House Approach: Annual Payments: $$60.7 billion; Unfunded Liability(at Year-end): NA; Modified Administration Approach: Annual Payments: $$44.4 billion; Unfunded liability (at Year-end): NA; Modified Senate Approach: Annual Payments: $$38.2 billion; Unfunded liability (at Year-end): NA. Average Annual Payment; Modified Current Law Approach: Annual Payments: $$7.2 billion; Unfunded Liability (at Year-end): NA; Modified House Approach: Annual Payments: $$7.6 billion; Unfunded Liability(at Year-end): NA; Modified Administration Approach: Annual Payments: $$5.5 billion; Unfunded liability (at Year-end): NA; Modified Senate Approach: Annual Payments: $4.8 billion; Unfunded liability (at Year-end): NA. Source: GAO. Notes: Totals may not add because of rounding. [End of table] Annual payments consist of any prefunding payments to the PSRHBF, and, where applicable, USPS's share of premium payments for existing retirees and beneficiaries (see descriptions of the different approaches). Unfunded liabilities are the amounts as of the end of the fiscal year, and start at an estimated value of $47.3 billion as of September 30, 2012, under all four approaches. Subsequent to the development of these projections, USPS reported the actual unfunded liability as of September 30, 2012, to be $47.9 billion. Figure 10: Estimated Annual Payments for Different Approaches from Fiscal Year 2013 to Fiscal Year 2020 under Senate Bill Assumption Basis: [Refer to PDF for image: vertical bar graph] Annual payments: Fiscal year: 2013; Modified Current Law Approach: $8.569 billion; Modified House Approach: $8.569 billion; Modified Administration Approach: $1.276 billion; Modified Senate Approach: $4.275 billion. Fiscal year: 2014; Modified Current Law Approach: $8.938 billion; Modified House Approach: $8.938 billion; Modified Administration Approach: $5.423 billion; Modified Senate Approach: $4.319 billion. Fiscal year: 2015; Modified Current Law Approach: $9.248 billion; Modified House Approach: $11.498 billion; Modified Administration Approach: $5.573 billion; Modified Senate Approach: $4.430 billion. Fiscal year: 2016; Modified Current Law Approach: $9.676 billion; Modified House Approach: $11.926 billion; Modified Administration Approach: $5.829 billion; Modified Senate Approach: $4.612 billion. Fiscal year: 2017; Modified Current Law Approach: $4.925 billion; Modified House Approach: $4.627 billion; Modified Administration Approach: $6.261 billion; Modified Senate Approach: $4.817 billion. Fiscal year: 2018; Modified Current Law Approach: $5.119 billion; Modified House Approach: $4.821 billion; Modified Administration Approach: $6.455 billion; Modified Senate Approach: $5.024 billion. Fiscal year: 2019; Modified Current Law Approach: $5.320 billion; Modified House Approach: $5.022 billion; Modified Administration Approach: $6.656 billion; Modified Senate Approach: $5.243 billion. Fiscal year: 2020; Modified Current Law Approach: $5.550 billion; Modified House Approach: $5.252 billion; Modified Administration Approach: $6.886 billion; Modified Senate Approach: $5.495 billion. Source: GAO. [End of figure] Long-term Effects under Senate Bill Assumption Basis: Table 8: Annual Payment and Unfunded Liability in the Long Term (in Fiscal Year 2040, the Last Year of the Projection) under Senate Bill Assumption Basis: Dollars in billions. Fiscal year 2040: Nominal $: Modified Current law Approach: Annual Payment: $12.7 billion; Unfunded Liability: $10.4 billion; Modified House Approach: Annual Payment: $12.4 billion; Unfunded Liability: $7.3 billion; Modified Administration Approach: Annual Payment: $14.1 billion; Unfunded Liability: $24.1 billion; Modified Senate Approach: Annual Payment: $13.8 billion; Unfunded Liability: $79.4 billion. Constant $; Modified Current law Approach: Annual Payment: $5.6 billion; Unfunded Liability: $4.5 billion; Modified House Approach: Annual Payment: $5.4 billion; Unfunded Liability: $3.2 billion; Modified Administration Approach: Annual Payment: $6.2 billion; Unfunded Liability: $10.6 billion; Modified Senate Approach: Annual Payment: $6.0 billion; Unfunded Liability: $34.7 billion. Percentage of Compensation; Modified Current law Approach: Annual Payment: 13%; Unfunded Liability: 11%; Modified House Approach: Annual Payment: 13%; Unfunded Liability: 8%; Modified Administration Approach: Annual Payment: 15%; Unfunded Liability: 25%; Modified Senate Approach: Annual Payment: 15%; Unfunded Liability: 84%. Funded percentage; Modified Current law Approach: Annual Payment: NA; Unfunded Liability: 96%; Modified House Approach: Annual Payment: NA; Unfunded Liability: 97%; Modified Administration Approach: Annual Payment: NA; Unfunded Liability: 92%; Modified Senate Approach: Annual Payment: NA; Unfunded Liability: 73%. Source: GAO. Notes: Annual payment is prefunding payment to the PSRHBF. Unfunded liability is as of the end of the fiscal year. Constant dollar amounts were derived by converting the projected 2040 nominal dollar amounts into 2012 dollars, using an inflation assumption based on the Senate bill assumption basis. Compensation was projected by GAO based on information provided by USPS. Additional information on methodology and assumptions is presented in appendix I. "Compensation" is used as a shorthand expression to represent employee salary and wage costs plus employee benefit costs exclusive of retiree health benefits, workers' compensation, or any forecasted contract negotiations savings. [End of table] Figure 11: Estimated Annual Payments (as a Percentage of Compensation) for Different Approaches from Fiscal Year 2013 to Fiscal Year 2040 under Senate Bill Assumption Basis: [Refer to PDF for image: multiple line graph] Fiscal year: 2013; Modified Current Law Approach: 20.3%; Modified House Approach: 20.3%; Modified Administration Approach: 3%; Modified Senate Approach: 10.1%. Fiscal year: 2014; Modified Current Law Approach: 21.4%; Modified House Approach: 21.4%; Modified Administration Approach: 13%; Modified Senate Approach: 10.3%. Fiscal year: 2015; Modified Current Law Approach: 22.1%; Modified House Approach: 27.4%; Modified Administration Approach: 13.3%; Modified Senate Approach: 10.6%. Fiscal year: 2016; Modified Current Law Approach: 22.9%; Modified House Approach: 28.2%; Modified Administration Approach: 13.8%; Modified Senate Approach: 10.9%. Fiscal year: 2017; Modified Current Law Approach: 11.5%; Modified House Approach: 10.8%; Modified Administration Approach: 14.6%; Modified Senate Approach: 11.2%. Fiscal year: 2018; Modified Current Law Approach: 11.8%; Modified House Approach: 11.1%; Modified Administration Approach: 14.8%; Modified Senate Approach: 11.6%. Fiscal year: 2019; Modified Current Law Approach: 12.1%; Modified House Approach: 11.4%; Modified Administration Approach: 15.1%; Modified Senate Approach: 11.9%. Fiscal year: 2020; Modified Current Law Approach: 12.4%; Modified House Approach: 11.7%; Modified Administration Approach: 15.3%; Modified Senate Approach: 12.2%. Fiscal year: 2021; Modified Current Law Approach: 12.7%; Modified House Approach: 12%; Modified Administration Approach: 15.6%; Modified Senate Approach: 12.6%. Fiscal year: 2022; Modified Current Law Approach: 12.9%; Modified House Approach: 12.3%; Modified Administration Approach: 15.8%; Modified Senate Approach: 12.9%. Fiscal year: 2023; Modified Current Law Approach: 13.2%; Modified House Approach: 12.6%; Modified Administration Approach: 15.9%; Modified Senate Approach: 13.2%. Fiscal year: 2024; Modified Current Law Approach: 13.4%; Modified House Approach: 12.9%; Modified Administration Approach: 16.1%; Modified Senate Approach: 13.5%. Fiscal year: 2025; Modified Current Law Approach: 13.6%; Modified House Approach: 13.1%; Modified Administration Approach: 16.2%; Modified Senate Approach: 13.8%. Fiscal year: 2026; Modified Current Law Approach: 13.8%; Modified House Approach: 13.3%; Modified Administration Approach: 16.3%; Modified Senate Approach: 14%. Fiscal year: 2027; Modified Current Law Approach: 14%; Modified House Approach: 13.4%; Modified Administration Approach: 16.3%; Modified Senate Approach: 14.2%. Fiscal year: 2028; Modified Current Law Approach: 14.1%; Modified House Approach: 13.6%; Modified Administration Approach: 16.3%; Modified Senate Approach: 14.4%. Fiscal year: 2029; Modified Current Law Approach: 14.2%; Modified House Approach: 13.7%; Modified Administration Approach: 16.4%; Modified Senate Approach: 14.6%. Fiscal year: 2030; Modified Current Law Approach: 14.3%; Modified House Approach: 13.8%; Modified Administration Approach: 16.3%; Modified Senate Approach: 14.8%. Fiscal year: 2031; Modified Current Law Approach: 14.3%; Modified House Approach: 13.8%; Modified Administration Approach: 16.3%; Modified Senate Approach: 14.9%. Fiscal year: 2032; Modified Current Law Approach: 14.3%; Modified House Approach: 13.9%; Modified Administration Approach: 16.2%; Modified Senate Approach: 15%. Fiscal year: 2033; Modified Current Law Approach: 14.3%; Modified House Approach: 13.9%; Modified Administration Approach: 16.2%; Modified Senate Approach: 15.1%. Fiscal year: 2034; Modified Current Law Approach: 14.3%; Modified House Approach: 13.9%; Modified Administration Approach: 16.1%; Modified Senate Approach: 15.1%. Fiscal year: 2035; Modified Current Law Approach: 14.2%; Modified House Approach: 13.8%; Modified Administration Approach: 15.9%; Modified Senate Approach: 15.2%. Fiscal year: 2036; Modified Current Law Approach: 14.1%; Modified House Approach: 13.8%; Modified Administration Approach: 15.8%; Modified Senate Approach: 15.2%. Fiscal year: 2037; Modified Current Law Approach: 14%; Modified House Approach: 13.7%; Modified Administration Approach: 15.6%; Modified Senate Approach: 15.2%. Fiscal year: 2038; Modified Current Law Approach: 13.9%; Modified House Approach: 13.5%; Modified Administration Approach: 15.4%; Modified Senate Approach: 15.1%. Fiscal year: 2039; Modified Current Law Approach: 13.7%; Modified House Approach: 13.3%; Modified Administration Approach: 15.1%; Modified Senate Approach: 14.8%. Fiscal year: 2040; Modified Current Law Approach: 13.4%; Modified House Approach: 13.1%; Modified Administration Approach: 14.8%; Modified Senate Approach: 14.5%. Source: GAO. [End of figure] Figure 12: Estimated Unfunded Liability (as a Percentage of USPS's Annual Compensation Costs) for Different Approaches from Fiscal Year- end 2012 to Fiscal Year-end 2040 under Senate Bill Assumption Basis: [Refer to PDF for image: multiple line graph] Fiscal year: 2012; Modified Current Law Approach: 101.8%; Modified House Approach: 101.8%; Modified Administration Approach: 101.8%; Modified Senate Approach: 101.8%. Fiscal year: 2013; Modified Current Law Approach: 106.5%; Modified House Approach: 106.5%; Modified Administration Approach: 124%; Modified Senate Approach: 116.9%. Fiscal year: 2014; Modified Current Law Approach: 100.4%; Modified House Approach: 100.4%; Modified Administration Approach: 127.7%; Modified Senate Approach: 122.7%. Fiscal year: 2015; Modified Current Law Approach: 92%; Modified House Approach: 86.6%; Modified Administration Approach: 129.8%; Modified Senate Approach: 127.3%. Fiscal year: 2016; Modified Current Law Approach: 81.4%; Modified House Approach: 70.5%; Modified Administration Approach: 130.5%; Modified Senate Approach: 130.7%. Fiscal year: 2017; Modified Current Law Approach: 81.7%; Modified House Approach: 71%; Modified Administration Approach: 129.8%; Modified Senate Approach: 133.4%. Fiscal year: 2018; Modified Current Law Approach: 81.6%; Modified House Approach: 71.1%; Modified Administration Approach: 128.6%; Modified Senate Approach: 135.7%. Fiscal year: 2019; Modified Current Law Approach: 81.2%; Modified House Approach: 70.9%; Modified Administration Approach: 127.2%; Modified Senate Approach: 137.8%. Fiscal year: 2020; Modified Current Law Approach: 80.1%; Modified House Approach: 70.1%; Modified Administration Approach: 124.9%; Modified Senate Approach: 139%. Fiscal year: 2021; Modified Current Law Approach: 78.8%; Modified House Approach: 69%; Modified Administration Approach: 122.4%; Modified Senate Approach: 140.1%. Fiscal year: 2022; Modified Current Law Approach: 80.2%; Modified House Approach: 70.8%; Modified Administration Approach: 122.4%; Modified Senate Approach: 143.4%. Fiscal year: 2023; Modified Current Law Approach: 79.8%; Modified House Approach: 70.8%; Modified Administration Approach: 119.9%; Modified Senate Approach: 144%. Fiscal year: 2024; Modified Current Law Approach: 78.8%; Modified House Approach: 70.3%; Modified Administration Approach: 117%; Modified Senate Approach: 144.1%. Fiscal year: 2025; Modified Current Law Approach: 77.3%; Modified House Approach: 69.2%; Modified Administration Approach: 113.6%; Modified Senate Approach: 143.5%. Fiscal year: 2026; Modified Current Law Approach: 75.3%; Modified House Approach: 67.6%; Modified Administration Approach: 109.8%; Modified Senate Approach: 142.4%. Fiscal year: 2027; Modified Current Law Approach: 72.8%; Modified House Approach: 65.5%; Modified Administration Approach: 105.5%; Modified Senate Approach: 140.8%. Fiscal year: 2028; Modified Current Law Approach: 69.9%; Modified House Approach: 63%; Modified Administration Approach: 100.9%; Modified Senate Approach: 138.7%. Fiscal year: 2029; Modified Current Law Approach: 66.6%; Modified House Approach: 60.1%; Modified Administration Approach: 96%; Modified Senate Approach: 136.1%. Fiscal year: 2030; Modified Current Law Approach: 63%; Modified House Approach: 56.8%; Modified Administration Approach: 90.7%; Modified Senate Approach: 133.1%. Fiscal year: 2031; Modified Current Law Approach: 58.9%; Modified House Approach: 53.1%; Modified Administration Approach: 85.2%; Modified Senate Approach: 129.7%. Fiscal year: 2032; Modified Current Law Approach: 54.6%; Modified House Approach: 49.1%; Modified Administration Approach: 79.3%; Modified Senate Approach: 125.8%. Fiscal year: 2033; Modified Current Law Approach: 49.9%; Modified House Approach: 44.7%; Modified Administration Approach: 73.2%; Modified Senate Approach: 121.6%. Fiscal year: 2034; Modified Current Law Approach: 45%; Modified House Approach: 40.1%; Modified Administration Approach: 66.9%; Modified Senate Approach: 117%. Fiscal year: 2035; Modified Current Law Approach: 39.8%; Modified House Approach: 35.2%; Modified Administration Approach: 60.4%; Modified Senate Approach: 112%. Fiscal year: 2036; Modified Current Law Approach: 34.4%; Modified House Approach: 30.1%; Modified Administration Approach: 53.7%; Modified Senate Approach: 106.8%. Fiscal year: 2037; Modified Current Law Approach: 28.8%; Modified House Approach: 24.8%; Modified Administration Approach: 46.8%; Modified Senate Approach: 101.3%. Fiscal year: 2038; Modified Current Law Approach: 23%; Modified House Approach: 19.2%; Modified Administration Approach: 39.8%; Modified Senate Approach: 95.5%. Fiscal year: 2039; Modified Current Law Approach: 17%; Modified House Approach: 13.5%; Modified Administration Approach: 32.6%; Modified Senate Approach: 89.6%. Fiscal year: 2040; Modified Current Law Approach: 11%; Modified House Approach: 7.7%; Modified Administration Approach: 25.4%; Modified Senate Approach: 83.6%. Source: GAO. Note: Under all four modified approaches to prefunding, the unfunded liability starts at the same point at September 30, 2012, and begins to diverge in fiscal year 2013 with the different approaches to prefunding. [End of figure] Figure 13: Estimated Funded Ratios for Different Approaches from Fiscal Year-end 2012 to Fiscal Year-end 2040 under Senate Bill Assumption Basis: [Refer to PDF for image: multiple line graph] Fiscal year: 2012; Modified Current Law Approach: 49.7%; Modified House Approach: 49.7%; Modified Administration Approach: 49.7%; Modified Senate Approach: 49.7%. Fiscal year: 2013; Modified Current Law Approach: 55%; Modified House Approach: 55%; Modified Administration Approach: 47.6%; Modified Senate Approach: 50.6%. Fiscal year: 2014; Modified Current Law Approach: 60.3%; Modified House Approach: 60.3%; Modified Administration Approach: 49.5%; Modified Senate Approach: 51.5%. Fiscal year: 2015; Modified Current Law Approach: 65.5%; Modified House Approach: 67.5%; Modified Administration Approach: 51.3%; Modified Senate Approach: 52.2%. Fiscal year: 2016; Modified Current Law Approach: 70.7%; Modified House Approach: 74.6%; Modified Administration Approach: 53.1%; Modified Senate Approach: 53%. Fiscal year: 2017; Modified Current Law Approach: 71.7%; Modified House Approach: 75.4%; Modified Administration Approach: 55%; Modified Senate Approach: 53.7%. Fiscal year: 2018; Modified Current Law Approach: 72.6%; Modified House Approach: 76.1%; Modified Administration Approach: 56.8%; Modified Senate Approach: 54.4%. Fiscal year: 2019; Modified Current Law Approach: 73.6%; Modified House Approach: 76.9%; Modified Administration Approach: 58.6%; Modified Senate Approach: 55.1%. Fiscal year: 2020; Modified Current Law Approach: 74.6%; Modified House Approach: 77.7%; Modified Administration Approach: 60.3%; Modified Senate Approach: 55.8%. Fiscal year: 2021; Modified Current Law Approach: 75.6%; Modified House Approach: 78.6%; Modified Administration Approach: 62%; Modified Senate Approach: 56.5%. Fiscal year: 2022; Modified Current Law Approach: 75.7%; Modified House Approach: 78.6%; Modified Administration Approach: 63%; Modified Senate Approach: 56.6%. Fiscal year: 2023; Modified Current Law Approach: 76.1%; Modified House Approach: 78.8%; Modified Administration Approach: 64%; Modified Senate Approach: 56.8%. Fiscal year: 2024; Modified Current Law Approach: 76.5%; Modified House Approach: 79.1%; Modified Administration Approach: 65.1%; Modified Senate Approach: 57.1%. Fiscal year: 2025; Modified Current Law Approach: 77.1%; Modified House Approach: 79.5%; Modified Administration Approach: 66.3%; Modified Senate Approach: 57.4%. Fiscal year: 2026; Modified Current Law Approach: 77.7%; Modified House Approach: 80%; Modified Administration Approach: 67.5%; Modified Senate Approach: 57.9%. Fiscal year: 2027; Modified Current Law Approach: 78.5%; Modified House Approach: 80.6%; Modified Administration Approach: 68.8%; Modified Senate Approach: 58.4%. Fiscal year: 2028; Modified Current Law Approach: 79.3%; Modified House Approach: 81.4%; Modified Administration Approach: 70.2%; Modified Senate Approach: 59%. Fiscal year: 2029; Modified Current Law Approach: 80.3%; Modified House Approach: 82.2%; Modified Administration Approach: 71.6%; Modified Senate Approach: 59.7%. Fiscal year: 2030; Modified Current Law Approach: 81.3%; Modified House Approach: 83.1%; Modified Administration Approach: 73%; Modified Senate Approach: 60.4%. Fiscal year: 2031; Modified Current Law Approach: 82.4%; Modified House Approach: 84.2%; Modified Administration Approach: 74.6%; Modified Senate Approach: 61.3%. Fiscal year: 2032; Modified Current Law Approach: 83.6%; Modified House Approach: 85.3%; Modified Administration Approach: 76.2%; Modified Senate Approach: 62.2%. Fiscal year: 2033; Modified Current Law Approach: 84.9%; Modified House Approach: 86.5%; Modified Administration Approach: 77.9%; Modified Senate Approach: 63.3%. Fiscal year: 2034; Modified Current Law Approach: 86.3%; Modified House Approach: 87.8%; Modified Administration Approach: 79.6%; Modified Senate Approach: 64.4%. Fiscal year: 2035; Modified Current Law Approach: 87.8%; Modified House Approach: 89.2%; Modified Administration Approach: 81.4%; Modified Senate Approach: 65.6%. Fiscal year: 2036; Modified Current Law Approach: 89.3%; Modified House Approach: 90.7%; Modified Administration Approach: 83.3%; Modified Senate Approach: 66.9%. Fiscal year: 2037; Modified Current Law Approach: 91%; Modified House Approach: 92.2%; Modified Administration Approach: 85.3%; Modified Senate Approach: 68.2%. Fiscal year: 2038; Modified Current Law Approach: 92.7%; Modified House Approach: 93.9%; Modified Administration Approach: 87.4%; Modified Senate Approach: 69.7%. Fiscal year: 2039; Modified Current Law Approach: 94.5%; Modified House Approach: 95.7%; Modified Administration Approach: 89.5%; Modified Senate Approach: 71.2%. Fiscal year: 2040; Modified Current Law Approach: 96.4%; Modified House Approach: 97.5%; Modified Administration Approach: 91.7%; Modified Senate Approach: 72.8%. Source: GAO. Note: Under all four modified approaches to prefunding, the funded ratio starts at the same point at September 30, 2012, and begins to diverge in fiscal year 2013 with the different approaches to prefunding. [End of figure] [End of section] Appendix III: Comments from the U.S. Postal Service: Joseph Corbett: Chief Financial Officer: Executive Vice President: United States Postal Service: 475 L'Enfant Plaza SW: Washington, DC 20200-5000: 202-268-5272 Fax: 202-268-4364: [hyperlink, http://www.usps.com] November 19, 2012: Ms. Lorelei St. James: Director, Physical Infrastructure Issues: U.S. Government Accountability Office: 441 G Street, NW: Washington, DC 20548-0001: Dear Ms. St. James: Thank you for the opportunity to provide comments to the Government Accountability Office (GAO) report titled, Status, Financial Outlook and Alternative Approaches to Fund Retiree Health Benefits. We concur with the overarching conclusion of the report that comprehensive reform is necessary to achieve a financially sustainable Postal Service. The Postal Service recognizes and accepts its obligation to provide effective, affordable health benefits to its employees and retirees. However, it does not have the financial resources to make prefunding payments required by current law. We believe that releasing this report is inappropriate given the solution to this situation is to allow the Postal Service to sponsor its own medical plan. The GAO is scheduled to report on our plans to sponsor our own medical programs soon. The report makes clear that the outlook for the Postal Service Retiree Health Benefits Fund (PSRHBF) is primarily determined by the costs of future health care coverage. These costs must be reduced and this can best be clone by adopting our proposal for a Postal Service-sponsored health plan for both current employees and retirees, separate from the Federal Employees Health Benefits Program. Allowing the Postal Service to gain control of its own health care program would save money, reduce or eliminate the current unfunded liability, and allow for better management of health care costs going forward. Any discussion of the PSRHBF's outlook is compromised without discussion of the benefits provided by the postal health plan. Providing a stable and dependable health care plan for all employees is a crucial part of our strategic plan. Having an affordable arrangement, utilizing best practices found in the private sector, will serve Postal Service employees and retirees well. Congressional decision-makers would be better served by issuance of a report that fully considers all alternatives to providing health benefits to postal employees and retirees, including the proposed postal health plan. Sincerely, Signed by: Joseph Corbett: [End of section] Appendix IV: Comments from the U.S. Postal Service, Office of Inspector General: Office of Inspector General: United States Postal Service: November 21, 2012: Ms. Lorelei St. James: Director, Physical Infrastructure: U.S. Government Accountability Office: 441 G Street, N.W. Washington, D.C. 20548: Dear Ms. St. James: Thank you for the opportunity to comment on the Government Accountability Office's report titled "Status, Financial Outlook, and Alternative Approaches to Fund Retiree Health Benefits." The report compares several approaches for funding the Postal Service's retiree health benefits and finds that there is a tradeoff between paying now and paying later. Paying more now reduces the amount that must be paid in the future, but it has created a severe fiscal challenge. Postponing payments will result in large payments in the future or longer amortization. We do not disagree with this basic premise of the report. We are concerned, however, that the report ought to provide needed context regarding the Postal Service's historic prefunding efforts. The context is important for the proper evaluation of the different proposals examined in the report. First, there is the historical context. The Postal Service started prefunding its retiree health benefits as a result of the discovery that, due to external fund management misjudgments, it was on track to seriously overfund its pension obligations by $78 billion. This discovery was one of several fund management issues identified about the same time. The decision to turn a mistake into a second prefunding obligation created its own problems. A 10-year schedule of prefunding payments was structured toward a 100-percent funding goal. The aggressive payment schedule appears to have been set based on byzantine "budget scoring" considerations rather than actuarial assumptions or an evaluation of the Postal Service's ability to make the payments. Second, there is the financial context. The Postal Service has never been able to afford a single payment. In the initial years the payments required borrowing. The long term liability has simply been shifted to a growing short term loan from the Treasury. Then later, the Postal Service began defaulting on the payments. Its ability to pay has even decreased as a result of a decline in the sale of many of its traditional business products. Now, the Postal Service has reached the limit of the amount it can borrow. Clearly, the Postal Service cannot make the payments, particularly during a period of major downsizing of its large operational and retail components. The Inspector General's "seal and grow" proposal identified in your report was made in the context of this urgent situation. Our unusual proposal is reasonable in the context of these highly unusual circumstances. The report mistakenly represents the seal and grow proposal as a permanent payment plan rather than a temporary measure. We examined a 4-year scenario as a possible time horizon in response to a Senate request. We also projected that the sealed fund would reach $90 billion in 21 years as an illustration of the growth rate of the fund from interest earnings only. We did not attempt to estimate the future liability, increase or decrease, but did note that it was likely to change. In recent years, the liability has grown more slowly than first estimated. Actually, when the Postal Service proved unable to contribute to the fund, the seal and grow plan came into de facto existence. We are now entering the third year of this 4-year example of allowing the fund to accrue interest while making no expenditures. As additional background, it could be noted that the Postal Service has substantially funded its retiree benefit programs. The Postal Service's pensions are fully funded including overpayments, and the Postal Service's retiree health liabilities are already nearly half funded. The amount that has been taken and set aside from postal revenues now totals $330 billion. The present health prefunding covers the estimated lifetime obligations for current retirees. It will likely remain challenging for the Postal Service to divert cash it desperately needs now to make prefunding payments for employees who have not yet retired and may not receive benefits in many cases for 30 years, possibly more. Our other major concern with the report is the fact that it does not adequately explore the use of the other assets the Postal Service holds as a means of satisfying this retiree health care obligation. My office has identified two sources of assets in the billions that can cover any unfunded liabilities should there be no Postal Service. One is the Postal Service's substantial real estate holdings. We have estimated their current value to be $85 billion that could be used to satisfy the current $48 billion liability. The properties can be sold in the event the Postal Service ceases operations. The other source comes from surpluses and overpayments in the Postal Service's pension funds, which could also be used to offset retiree health deficits. Together, these options provide a great insurance that assets meet or exceed the Postal Service's benefits funds future liabilities, unless additional prefunding requirements are added to the pension and health obligations. Lastly, the Government Accountability Office (GAO) report provides a useful discussion of the virtues of prefunding. However, we suggest that the report also examine the problems of prefunding and examine why no business or government entity, including the GAO or the Office of Personnel Management, has attempted to take advantage of such a seemingly appealing opportunity. It would certainly appear that despite the many virtues identified, the prefunding scheme is destroying the only entity that has taken the cure. It is important to recognize that making prefunding payments at the current levels will bankrupt the Postal Service regardless of whether mail volumes return or further decline. Before the recent drastic revenue declines, the Postal Service was still borrowing to make the prefunding payments. We believe a number of options exist to assure world-class levels and standards for prefunding. Today the Postal Service prefunding levels exceed those of any business, federal or state government entity that has been identified in the GAO report or by my office. Ironically, the principal risk in the current prefunding plan is the threat of creating conditions preventing the Postal Service from remaining self- funded, since that is the very purpose of pre-funding in the first place. Sincerely, Signed by: David C. Williams: Inspector General: [End of section] Appendix V: GAO Contact and Staff Acknowledgments: GAO Contacts: Frank Todisco, (202) 512-2834 or todiscof@gao.gov: Lorelei St. James, (202) 512-2834 or stjamesl@gao.gov: Staff Acknowledgments: In addition to the individuals named above, Samer Abbas, Teresa Anderson, Beryl Davis, John Dicken, Kim Granger, Jacquelyn Hamilton, Hannah Laufe, Jennifer Leone, W. Stephen Lowrey, Kim McGatlin, Jonathan McMurray, Kristi Peterson, Steve Robblee, Amy Rosewarne, Aron Szapiro, and Crystal Wesco made key contributions to this report. [End of section] Footnotes: [1] Originally due at the end of fiscal year 2011, USPS's $5.5 billion retiree health benefit payment requirement was delayed until August 1, 2012. Pub. L. No. 112-74, § 632, 125 Stat. 786, 928 (Dec. 23, 2011). USPS missed that payment as well as the $5.6 billion that was due by September 30, 2012. [2] Pub. L. 109-435, § 803, 120 Stat. 3198 (Dec. 20, 2006). [3] For the purposes of this report, payments made by USPS refer to payments from the Postal Service Fund. The Postal Service Fund is a revolving fund established in the U.S. Treasury for the deposit of all revenues, interest, appropriations, proceeds from borrowing, or any other receipts from USPS's operations. The Fund is available to USPS to carry out the purposes, functions, and powers of USPS. 39 U.S.C. § 2003. [4] Pursuant to PAEA, no later than June 30, 2017, OPM is required to compute, and by June 30 of each succeeding year, to recompute, a series of annual installments which provide for the liquidation of any liability or surplus by September 30, 2056, or within 15 years, whichever is later. 5 U.S.C. § 8909a(d)(2)(B). OPM told us it calculates these payments by determining the amount that, if contributed every year, would be projected to fully fund the remaining unfunded liability over an amortization period ending in the later of fiscal year 2056, or 15 years subsequent to the then-current fiscal year. [5] The $41 billion dollars in net losses for fiscal years 2007 through 2012 are those reported by USPS in its financial reporting, which is on an accrual basis and not a cash basis. However, USPS reports its retiree health benefits cost in a manner that its reported expense for these benefits is equal to its required cash payments, as discussed later in this report. [6] The retiree health benefit liability represents the actuarial present value of the cost of the portion of future retiree health premiums for which USPS is responsible and that are attributable to past service; this liability reflects all such projected future costs for current retirees and beneficiaries and a portion of such projected future costs for current workers. Actuarial present values of this type, whether for retiree health benefits or pension benefits, are variously referred to as a "liabilities," "accrued liabilities," or "obligations," often depending on the user of the term (e.g., actuaries, accountants, or lawyers) and the context (e.g., financial reporting, statutes, or media). Throughout this report, we use the terms "liability" to refer to this actuarial present value of future costs attributable to past service and "unfunded liability" to refer to the excess of this liability over the amount of funds in the PSRHBF. [7] GAO, U.S. Postal Service: Transformation Challenges Present Significant Risks, [hyperlink, http://www.gao.gov/products/GAO-01-598T] (Washington, D.C.: Apr. 4, 2001). [8] GAO, U.S. Postal Service: Moving Forward on Financial and Transformation Challenges, [hyperlink, http://www.gao.gov/products/GAO-02-694T] (Washington, D.C.: May 13, 2002). [9] GAO, U.S. Postal Service: Key Postal Transformational Issues, [hyperlink, http://www.gao.gov/products/GAO-03-812T] (Washington, D.C.: May 29, 2003). [10] Pub. L. No. 108-18, 117 Stat. 624 (April 23, 2003). [11] CSRS is a defined benefit, contributory retirement system for certain federal employees. It was replaced by the Federal Employees Retirement System (FERS) for federal employees who first entered covered service on and after January 1, 1987. [12] The 2003 Act required USPS to contribute the employer's share of the "dynamic normal cost" (which is a normal cost computed using "dynamic assumptions"). The normal cost is the annual growth in pension liabilities resulting from an additional year of service by plan participants. "Dynamic assumptions" are defined as economic assumptions that are used in determining actuarial costs and liabilities in a retirement system and in anticipating the effects of long-term future investment yields, future increases in rates of basic pay, and future rates of price inflation. Pub. L. No. 108-18, § 2(a)(3) (Apr. 23, 2003). The prior-funding methodology had used "static assumptions," which did not project future pay or cost-of- living increases and used a fixed interest-rate assumption. [13] Pub. L. No. 109-435, 120 Stat. 3198 (Dec. 20, 2006). [14] This shift created a postal actuarial funding surplus in CSRS of $17 billion, which was transferred to the PSRHBF. In addition, PAEA suspended USPS employer contributions to CSRS. Beginning in fiscal year 2017, if OPM determines a postal supplemental liability exists in CSRS, USPS must commence making annual employer contributions to CSRS in the following fiscal year. [15] OPM told us it calculates these payments by determining the amount that, if contributed every year, would be projected to fully fund the remaining unfunded liability over an amortization period ending in the later of fiscal year 2056, or 15 years subsequent to the then-current fiscal year. Therefore, for years subsequent to 2041, OPM would calculate an amortization amount that would be sufficient to liquidate the unfunded liability (or surplus) over the next 15 years. For example, in 2050, OPM would calculate an amortization amount that would be sufficient to liquidate the unfunded liability (or surplus) by 2065. This "rolling" amortization approach would never fully liquidate the unfunded liability (if experience matched the actuarial assumptions), but could be expected to approach full funding sufficiently closely. [16] USPS, United States Postal Service Annual Report 2007, (Washington, D.C.: Nov. 15, 2007). [17] GAO, U.S. Postal Service: Postal Reform Law Provides Opportunities to Address Postal Challenges, [hyperlink, http://www.gao.gov/products/GAO-07-684T] (Washington, D.C.: Apr.17, 2007). [18] 39 U.S.C. § 2005(a)(2). [19] Even though USPS did not make these payments, it recorded a loss for the obligation to make these payments, which remains outstanding. [20] We have ongoing work reviewing USPS's efforts in these areas. [21] USPS's annual appropriations acts have required USPS to provide 6- day delivery of mail at not less than 1983 levels. See e.g., Pub. L. No. 112-74, 125 Stat. 786, 923 (Dec. 23, 2011). [22] PRC noted that USPS improperly deflated mailers' reported volume- reduction projections and that the reported declines should not have been reduced, and determined, based on USPS's survey data, that it is likely to lose almost $600 million in net revenue because of mailer response to the proposal. See Postal Regulatory Commission, Advisory Opinion on Elimination of Saturday Delivery, Docket No. N2010-1 (Washington, D.C.: Mar. 24, 2011). [23] As discussed later in this report, under some of the prefunding approaches, contributions would be lower than otherwise only in fiscal years 2017 and later, when contributions are determined on an actuarial basis. [24] OPM Office of Inspector General, A Study of the Risks and Consequences of the USPS OIG's Approaches to Change USPS's Funding of Retiree Benefits (Washington, D.C.: February 28, 2011). [25] OPM officials told us that not all USPS career employees are in FEHBP, such as those employees covered through another plan, (e.g., a spouse's employee health plan) or those who simply do not participate in FEHBP. Officials added that USPS's participation rate of eligible employees is generally about 90 percent. [26] Pub. L. No. 112-74, § 632. [27] H. R. 2309, 112th Cong. (2011). [28] S. 1789, 112th Cong. (2011). [29] We obtained data and projections from USPS and OPM, and built on this information by performing additional calculations and projections. Methodology and assumptions are presented in more detail in appendix I. [30] We refer to these payments as "fixed" because they are set amounts that do not vary with an actuarial assessment of future retiree health care costs. [31] The increase in premium payments is driven in part by a significant number of expected retirements. As a result, this increase in payments for retirees would be at least partially offset by otherwise lower payments for employees, since there would be fewer of them. [32] Reductions would occur in each year except 2016, when the prefunding payment would be slightly higher than under current law. [33] As calculated by OPM, the amortization period would run to the later of 2052 or 15 years from the then-current fiscal year, in contrast to the later of 2056 or 15 years from the then-current fiscal year under current law. USPS's share of premium payments would be paid from the PSRHBF, instead of payments made by USPS, beginning right away (which we modeled to begin in fiscal year 2013), instead of in fiscal year 2017 under current law. [34] As discussed later in this report, we modified assumptions to analyze the four prefunding approaches on a comparative basis, and also to demonstrate that the modification does not have a material effect on our findings. [35] All approaches discussed in the following sections, except the section on sensitivity to assumptions, are modeled under the current law assumption basis. Comparative results under the Senate bill assumption basis are discussed in the section on sensitivity to assumptions and presented more fully in appendix II. [36] There is no single actuarial approach, but an actuarially determined contribution will typically consist of a normal cost--which reflects future costs attributable to the current year of employee service--plus an amortization payment to pay down the unfunded liability. The size of the amortization payment will vary with the length of the amortization period selected. A longer amortization period requires lower annual payments but over a longer period. A shorter amortization period requires larger amortization payments but over a shorter period. [37] Compensation costs consist of projected employee salary and wage costs plus employee benefit costs exclusive of retiree health benefits, worker's compensation, or any forecasted contract negotiations savings. [38] We assumed compensation growth would vary across the two assumption bases because of the different underlying inflation assumptions. Further details of the projection are contained in appendix I. [39] Although the Modified Senate Approach bases amortization payments on the difference between PSRHBF assets and 80 percent of the liability, the normal cost portion of required payments is based on 100 percent of the normal cost. Normal cost constitutes a higher portion of total payments towards the end of the projection period. [40] These projections were performed prior to the release of actual 2012 results. [41] These percentages are from the projections under the current law assumption basis. [42] The changes were designed to reduce the year-to-year volatility in the value of the liabilities reported by some agencies, done via the use of an historical averaging period, as discussed in the text. [43] In addition to the three economic assumptions discussed in this section, the valuation and projection of retiree health benefit payments and liabilities depend on other assumptions as well, such as assumed mortality rates. These other assumptions would not vary between the current law assumption basis and the Senate Bill assumption basis. We have not audited the actuarial assumptions-- meaning the set of economic and demographic assumptions--used by OPM. [44] At least two other federal agencies that value large benefit programs with liabilities for future pension, medical, or disability benefits--the Department of Defense and the Veterans' Administration-- also use a 10-year historical averaging period, for consistency across federal agencies in federal financial reporting. [45] For example, the level of interest rates will generally be higher or lower depending upon expectations regarding future inflation. [46] This model is known as the Getzen model. [47] A 10-K Form is an annual report that many for-profit corporations must file with the SEC within 90 days of the close of their fiscal year. It is a publicly available document that contains audited financial statements and other information on a corporation's financial condition. [48] Results presented in this section are on the current law assumption basis. Comparative results, as constant dollars and percentage of compensation, are approximately the same under the current law and Senate bill assumption bases. [49] Letter from USPS OIG Inspector General David C. Williams to Senator Bernie Sanders, February 6, 2012. [50] If the liability were to become fully funded under this approach, the fund would then become "unsealed." USPS's share of premium payments would then start to be paid out of the PSRHBF, and USPS would start making prefunding payments into the PSRHBF to cover normal costs. As discussed earlier, the achievement of full funding does not mean that there are no further prefunding payments. [51] The liability rose to $93.6 billion at fiscal year-end 2012, a figure not known at the time of the USPS OIG's analysis in February 2012. [52] This estimate uses the current law assumption basis. Under the Senate bill assumption basis, the corresponding figure is a 2032 unfunded liability of $88 billion, which is equal to $49 billion in 2012 dollars and 127 percent of 2032 compensation. [53] These estimates are under the current law assumption basis. Under the Senate bill assumption basis, the corresponding projected funded percentages in 2040 are 76 percent under the Seal and Grow Approach and 73 percent under the Modified Senate Approach. [54] OPM Office of Inspector General, A Study of the Risks and Consequences of the USPS OIG's Approaches to Change USPS's Funding of Retiree Benefits (Washington, D.C.: Feb. 28, 2011). [55] U.S. Postal Service Office of Inspector General, "Pension and Retiree Health Care Funding Levels: Management Advisory Report," FT-MA- 12-002 (June 18, 2012). [56] FT-MA-12-002. [57] GAO, U.S. Postal Service: Allocation of Responsibility for Pension Benefits between the Postal Service and the Federal Government, [hyperlink, http://www.gao.gov/products/GAO-12-146] (Washington, D.C.: Oct. 13, 2011). The amount of USPS's FERS pension surplus that we cited in this report was $6.9 billion, which was according to the then-most-recent actuarial analysis as of September 30, 2009. More recent actuarial analyses put this surplus at $10.9 billion as of September 30, 2010, and $2.6 billion as of September 30, 2011. Over this same period, estimates of USPS's CSRS pension surplus went from a deficit (unfunded liability) of $7.3 billion as of September 30, 2009, to a surplus of $1.6 billion as of September 30, 2010, to a deficit of $17.8 billion as of September 30, 2011. [58] USPS has since then reached its borrowing limit of $15.0 billion. [59] OPM Office of Inspector General, A Study of the Risks and Consequences of the USPS OIG's Approaches to Change USPS's Funding of Retiree Benefits (Washington, D.C.: Feb. 28, 2011). [60] S. Rep. No. 112-143, at 5 (2012). [61] As noted earlier, as long as USPS continues in operation with the retiree health plan intact, additional prefunding contributions would still be needed, even after full funding was achieved, for the annual growth in the liability for the cost of benefits attributable to ongoing employee service. [62] USPS does, however, disclose the value of its liability and unfunded liability (termed "obligation" rather than "liability," as discussed earlier), and the changes in such values from the prior year, in the footnotes to its financial statements and in its Management's Discussion and Analysis that accompanies the statements. [63] GAO, U.S. Postal Service: Accounting for Postretirement Benefits, [hyperlink, http://www.gao.gov/products/GAO-02-916R] (Washington, D.C.: Sept.12, 2002). [64] Retiree health benefits generally make up the largest part of OPEB. The "other" in OPEB was meant to refer to post-employment benefits other than pensions. [65] FT-MA-12-002. [66] GAO, State and Local Government Retiree Health Benefits, Liabilities are Largely Unfunded, but Some Governments Are Taking Action, [hyperlink, http://www.gao.gov/products/GAO-10-61] (Washington, D.C.: Nov. 30, 2009). [67] Department of Defense, Office of the Actuary, Valuation of the Medicare-Eligible Retiree Health Care Fund (Alexandria, VA: Sept. 30, 2010) [68] Department of Defense, Board of Actuaries, Report to the President and the Congress on the DOD Medicare-Eligible Retiree Health Care Fund (Arlington, VA: Dec. 30, 2009). [69] The financial reporting rules vary by sector, and are generally governed by the FASB in the private sector, the GASB for state and local governments, and the FASAB for federal agencies and the federal government as a whole. The particular accrual rules within each sector vary in a number of ways, regarding such factors as actuarial methods and assumptions, amortization periods or lack thereof, and recognition on the employer's balance sheet. [70] [hyperlink, http://www.gao.gov/products/GAO-02-916R]. [71] FT-MA-12-002 [72] If authorized by statute. [73] OPM, Civil Service Retirement and Disability Fund Annual Report, Fiscal Year Ended September 30, 2011 (issued January 2012). This report is available on request from OPM. [74] OPM does not normally project future liabilities. It needs to calculate current liabilities for future payments each year, but fulfilling its mission does not require any calculation of future liabilities. As such, OPM did not have previously developed software to do such projections, and had to do special programming specifically for this request. [End of section] GAO’s Mission: The Government Accountability Office, the audit, evaluation, and 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 GAO’s website [hyperlink, http://www.gao.gov]. Each weekday afternoon, GAO posts on its website newly released reports, testimony, and correspondence. To have GAO e-mail you a list of newly posted products, go to [hyperlink, http://www.gao.gov] and select “E-mail Updates.” Order by Phone: The price of each GAO publication reflects GAO’s actual cost of production and distribution and depends on the number of pages in the publication and whether the publication is printed in color or black and white. Pricing and ordering information is posted on GAO’s website, [hyperlink, http://www.gao.gov/ordering.htm]. Place orders by calling (202) 512-6000, toll free (866) 801-7077, or TDD (202) 512-2537. Orders may be paid for using American Express, Discover Card, MasterCard, Visa, check, or money order. Call for additional information. Connect with GAO: Connect with GAO on facebook, flickr, twitter, and YouTube. Subscribe to our RSS Feeds or E mail Updates. Listen to our Podcasts. Visit GAO on the web at [hyperlink, http://www.gao.gov]. To Report Fraud, Waste, and Abuse in Federal Programs: Contact: Website: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]; E-mail: fraudnet@gao.gov; Automated answering system: (800) 424-5454 or (202) 512-7470. Congressional Relations: Katherine Siggerud, Managing Director, siggerudk@gao.gov: (202) 512-4400: U.S. Government Accountability Office: 441 G Street NW, Room 7125: Washington, DC 20548. Public Affairs: Chuck Young, Managing Director, youngc1@gao.gov: (202) 512-4800: U.S. Government Accountability Office: 441 G Street NW, Room 7149: Washington, DC 20548. [End of document]