2014 Survey of GASB Accounting Measures for Public Pension Plans

Transcription

1 2014 Survey of GASB Accounting Measures for Public Pension Plans By Paul Zorn 1 In June 2012, the Governmental Accounting Standards Board (GASB) made significant changes to the accounting and financial reporting standards for state and local government pension plans, as well as for their sponsoring governments. Prior to the GASB s changes, a public pension plan s measures of the actuarial accrued liability, actuarial value of assets, and the unfunded pension liability were essentially the same for both funding and accounting purposes. However, under the GASB s recent changes, the accounting measures will be different from the funding measures. 2 For public pension plans, the new measures under GASB Statement No. 67 are effective for plan fiscal years beginning after June 15, As a result, the new measures are available in the financial reports of state and local government pension plans for fiscal years ending on June 30, 2014 and thereafter. To obtain a better sense of the new measures and what they mean, this research report surveys the June 30, 2014 financial reports of 44 large public pension plans covering general employees and teachers. The GASB s New Accounting Measures The GASB s new pension accounting measures include the total pension liability, the plan s fiduciary net position, and the net pension liability. The total pension liability (TPL) is the present value of pension benefits attributed to prior service that has accrued to active employees and retirees who are currently covered by the plan. The TPL is determined using the Entry Age Normal actuarial cost method and includes benefits related to projected salary and service, as well as cost-of-living adjustments that are deemed to be automatic. The plan s fiduciary net position (FNP) is the fair (market) value of the pension plan s assets that are available to pay benefits, including accumulated employer and employee contributions to the plan, and investment earnings. The net pension liability (NPL) is the difference between the total pension liability and the plan s fiduciary net position. In many ways, the GASB s total pension liability is similar to the actuarial accrued liability (AAL) that is used to determine the ongoing contributions needed to fund the promised benefits. However, there are several key differences between the accounting and funding measures: For funding purposes, the discount rate used to determine the present value of pension benefits is the long-term expected rate of return on plan assets. For accounting purposes, the discount rate is based on: 1) the long-term expected rate of return on plan assets, to the extent the plan s fiduciary net position is projected to pay benefits for current members; and 2) a tax-exempt, general obligation municipal bond rate, to the extent the projected FNP is not sufficient to pay projected benefits. Consequently, if the projected FNP is not sufficient, the discount rate for accounting purposes would include a portion based on the municipal bond rate. At times when municipal bond rates are lower 1 The author thanks Mark Randall, David Kausch, James Rizzo and Mary Ann Vitale at GRS for their thoughtful comments and assistance with this research report; however, he retains full responsibility for the accuracy of the information. 2 For a detailed summary of the GASB s changes, see The GASB s New Pension Accounting and Financial Reporting Standards, in the October 2012 issue of GRS Insight, available on the GRS website at 2/25/ Gabriel Roeder Smith & Company -1-

2 than the long-term expected return (as is currently the case), the accounting discount rate will be lower than the funding discount rate and the TPL will be higher than the AAL. Even if the discount rates for accounting and funding are the same, the TPL and AAL will differ if the actuarial cost method for funding purposes is not Entry Age or if it is a variation of Entry Age which differs from the GASB s parameters. Another important difference between the accounting and funding measures relates to the way plan assets are valued. For funding purposes, the actuarial value of assets (AVA) often smoothes investment gains and losses into the value of assets over time, typically 5 years. This is done to reduce the volatility in contribution rates. However, for accounting purposes, the fair (market) value of assets is used to determine the net pension liability, making the NPL more volatile than the unfunded actuarial accrued liability (UAAL). Survey Results The survey was developed to examine the GASB s new accounting measures under GASB Statement No. 67 and compare them with the measures used for pension funding. The survey results are based on 44 large statewide plans with fiscal years ending June 30, 2014 and covering general employees and teachers. 3 The surveyed plans represent about one-third of the large public plans in the United States. 4 As shown in Chart 1, the total pension liability for the surveyed plans grew from $1,427.7 billion as of June 30, 2013 to $1,487.5 billion as of June 30, 2014, an increase of $59.8 billion, or 4.2% of the TPL as of June 30, The surveyed plans fiduciary net position grew from $942.7 billion as of June 30, 2013 to $1,070.0 billion as of June 30, 2014, an increase of $127.3 billion, or 13.5% of the FNP as of June 30, As will be discussed later, much of the increase in the FNP was due to strong investment earnings during the fiscal year, as well as employer and employee contributions. The FNP was also reduced, in part, as a result of the retirement benefits paid during the year. The overall net pension liability for the surveyed plans actually fell from $485.0 billion as of June 30, 2013 to $417.5 billion as of June 30, 2014, a decline of $67.5 billion, which reduced the surveyed plans overall 3 Other plans covering public safety, legislatures, and judges have somewhat different characteristics and so are not included in this study. Future studies may include them as a greater number of financial reports become available. 4 The Public Fund Survey, sponsored by the National Association of State Retirement Administrators and the National Council on Teacher Retirement, provides information on 126 large public plans. 5 Note: the numbers may not add due to rounding. 2/25/ Gabriel Roeder Smith & Company -2-

3 NPL by 13.9%. Again, this reduction is largely attributable to strong investment earnings over the fiscal year. Changes in the Total Pension Liability As discussed above, the total pension liability is the accounting measure of the accrued pension liability for active employees and retirees who are currently covered by the pension plan. For the surveyed plans overall, the TPL increased Table 1: Changes in the Surveyed Plans' Total Pension Liability Components of Changes in the TPL $ Billions % Change in TPL Service Cost $ % Interest on TPL % Changes of Benefit Terms % Difference Between Expected and Actual Experience % Changes of Assumptions % Benefit Payments and Refunds (80.7) -5.7% Net Change in the Plans' TPL Over the Fiscal Year % Note: Numbers may not add due to rounding. by $59.8 billion (or 4.2%) over the fiscal year ending June 30, The increase resulted from changes in the components of the TPL, as shown in Table 1. Service Cost: The service cost represents the cost of benefits that have accrued to active employees over the fiscal year. As shown in Table 1, the service cost for the surveyed plans amounted to $28.8 billion, increasing the TPL by 2.0 percentage points over the fiscal year. Interest on TPL: This component reflects the amount of interest charged to the TPL over the year based on the plan s discount rate. 6 For the surveyed plans, interest was the largest component of the change in the TPL, amounting to $105.1 billion and increasing the TPL by 7.4 percentage points over the fiscal year. Changes of Benefits Terms: Under the GASB s new standards, the full cost of changes in benefit terms is immediately recognized in the TPL, rather than being recognized more gradually over time. Interestingly, few of the surveyed plans changed their benefits during this fiscal year. As a result, the overall impact of benefit changes had little impact on the TPL for the surveyed plans. Difference Between Expected and Actual Experience: This component reflects changes in the TPL due to the difference between expected actuarial experience and actual experience. As shown in Table 1, these differences for the surveyed plans amounted to $1.5 billion, increasing the TPL by 0.1 percentage points over the fiscal year. Changes of Assumptions: This component reflects the change in the TPL due to changes in the actuarial assumptions or actuarial methods used to calculate the TPL. For the surveyed plans, these changes increased the TPL by $4.8 billion or 0.3 percentage points over the fiscal year. Benefit Payments and Refunds: The TPL is reduced by paying the annual benefits due to retired members and refunding contributions for members who leave the plan before vesting. Overall, the surveyed plans paid $80.7 billion in benefits and refunds, lowering the TPL by 5.7 percentage points over the fiscal year. When the changes in the TPL are added together, the results show that the TPL of $1,427.7 billion as of June 30, 2013 was increased by $59.8 billion (or 4.2%) to $1,487.5 billion as of June 30, As discussed on page 1, the discount rate used for accounting purposes may be different from the discount rate used to value the plan s liabilities for funding purposes. 2/25/ Gabriel Roeder Smith & Company -3-

4 Changes in the Fiduciary Net Position The plans fiduciary net Table 2: Changes in the Surveyed Plans' Fiduciary Net Position position reflects the fair Components of Changes in the FNP $ Billions % Change in FNP (market) value of plan Employer (and Nonemployer) Contributions $ % assets that are available Member Contributions % to pay benefits. Overall, the FNP for the sur- Net Investment Income % Benefit Payments and Refunds (80.7) -8.6% veyed plans increased Administrative Expenses (0.7) -0.1% by $127.3 billion, or 13.5% over the fiscal Other % year. Table 2 shows Net Change in the Plans' FNP Over the Fiscal Year % that the increase consisted of the following changes in the components of the Note: Numbers may not add due to rounding. FNP: Employer (and Nonemployer) Contributions: Contributions by employers (and nonemployer contributing entities) increased the FNP by $33.2 billion for the surveyed plans, or 3.5 percentage points over the fiscal year. Member Contributions: In addition, member contributions amounted to $14.6 billion over the year, increasing the surveyed plans FNP by 1.5 percentage points. Net Investment Income: The largest increase in the FNP during this fiscal year came from investment income. For the surveyed plans, investment earnings (net of investment fees) increased the FNP by $160.5 billion or 17.0 percentage points, due to strong investment returns. Benefit Payments and Refunds: As is the case with the TPL, benefit payments and refunds also reduce the FNP. Benefits and refunds paid by the surveyed plans amounted to $80.7 billion over the fiscal year, reducing the FNP by 8.6 percentage points. Administrative Expenses: Plan administrative expenses also reduce the FNP, but by a relatively small amount. Administrative expenses for the surveyed plans reduced the FNP by about $700 million, or 0.1 percentage points over the fiscal year. Other: Other changes to the FNP were relatively insignificant. When the changes in the FNP for the surveyed plans are added together, the results show that the FNP of $942.7 billion as of June 30, 2013 was increased by $127.3 billion (or 13.5%) to $1,070.0 billion as of June 30, Changes in the Net Pension Liability As a result of the changes in the TPL and FNP, the overall net pension liability for the surveyed plans decreased by $67.5 billion, or 13.9% over the fiscal year. Since the NPL is the difference between the TPL and the FNP, it essentially contains the same components. 7 Components that increase the TPL also increase the NPL, since the NPL is a measure of the pension liability. Components that increase the FNP decrease the NPL, since the FNP is subtracted from the TPL to determine the NPL. Table 3 on the following page shows the changes in the NPL. Note that the dollar values for the components of the NPL are the same as in Tables 1 and 2, but the percentages are different, since they are based on percentages of the NPL (rather than the TPL or FNP) for the fiscal year. 7 However, the component for benefit payments and refunds is not included. Since it reduces both the TPL and the FNP, it is canceled out. 2/25/ Gabriel Roeder Smith & Company -4-

5 For the surveyed plans as Table 3: Changes in the Surveyed Plans' Net Pension Liability a whole, Table 3 shows Components of Changes in the NPL that the service cost increased the NPL by 5.9 of the TPL Components Service Cost 28.8 percentage points over the year. In addition, interest on the TPL increased the Components NPL by 21.7 percentage of the FNP points. Moreover, the combined changes in benefits, differences between expected and actual Net Change in the Plans' NPL Over the Fiscal Year experience, changes in Note: Numbers may not add due to rounding. assumptions, and plan administrative expenses increased the NPL by another 1.5 percentage points. However, combined employer and employee contributions reduced the NPL by 9.8 percentage points and net investment income reduced it by another 33.1 percentage points, resulting in an overall decline in the NPL of 13.9 percentage points. Chart 2 illustrates the changes in the NPL over the year. As shown in the chart, the change in the NPL primarily resulted from changes in: 1) service cost; 2) interest on the TPL; 3) employer and employee contributions; and 4) investment income While these results indicate an improvement in the financial status of public pension plans based on the accounting measures, they also demonstrate the potential volatility of the NPL due to changes in net investment income. Comparing Funding and Accounting Measures $ Billions % Change in NPL $ 5.9% Interest on TPL % Changes of Benefit Terms % Difference Between Expected and Actual Experience % Changes of Assumptions % Employer (and Nonemployer) Contributions (33.2) -6.8% Member Contributions (14.6) -3.0% Net Investment Income (160.5) -33.1% Administrative Expenses % Other (0.4) -0.1% (67.5) -13.9% Another topic of interest is the extent to which the funding measures and accounting measures are different. Chart 3, on the following page, compares the funding and accounting measures as of June 30, For the surveyed plans, it shows the accounting measure of the total pension liability was $1,427.7 billion while the actuarial accrued liability was $1,370.5 billion, resulting in the accounting measure of accrued liability being about 4.2% higher than the funding measure. During the GASB s deliberations related to the changes in the accounting standards, concerns were raised about including a portion of the municipal bond rate in the discount rate for plans where the projected FNP did not cover projected benefits. However, for the vast majority of the surveyed plans, the projected FNP 8 Although some of the plans have had funding valuations done as of June 30, 2014, not enough have been done to draw meaningful conclusions. However, all of the surveyed plans have done accounting and funding valuations as of June 30, /25/ Gabriel Roeder Smith & Company -5-

6 was sufficient to pay projected benefits, and so the discount rate for funding purposes was the same as the discount rate for accounting purposes. 9 Consequently, the difference between the AAL and the TPL was less than expected. Chart 3 also shows that, for the surveyed plans, the value of assets for accounting purposes (FNP) was larger than the actuarial value of assets for funding purposes (AVA). The overall FNP for the surveyed plans was $942.7 billion while the AVA was $912.3 billion, making the FNP about 3.3% higher than the AVA. Since investment experience was generally above expectations as of June 30, 2013, this was expected, given the smoothing of assets for funding purposes. Note that the FNP in Chart 3 does not include the substantial investment income earned between June 30, 2013 and June 30, As a result of the differences between: i) the TPL and AAL and ii) the FNP and AVA, the accounting measure of the net pension liability (NPL) was somewhat higher than the funding measure of the unfunded actuarial accrued liability (UAAL). For the surveyed plans overall, the NPL amounted to $485.0 billion, while the UAAL amounted to $458.2 billion, making the NPL 5.8% larger than the UAAL as of June 30, Conclusions The GASB s new accounting measures for state and local government pensions include the total pension liability, the fiduciary net position, and the net pension liability, which are different from the funding measures. To understand the new measures, it is important to understand their components and how the components interact. The major components of changes to the TPL include: 1) service cost; 2) interest on the TPL; and 3) benefit payments and refunds. However, changes in benefit terms, plan assumptions, and plan experience can also have significant impacts. The major components of changes to the FNP include: 1) employer and member contributions; 2) net investment earnings; and 3) benefit payments and refunds. Overall, for the 44 surveyed plans over the fiscal year, the TPL increased $59.8 billion (4.2%) from $1,427.7 billion to $1,487.5 billion. However, the FNP increased $127.3 billion (13.5%) from $942.7 billion to $1,070.0 billion. This resulted in a $67.5 billion (13.9%) decline in the NPL from $485.0 billion to $417.5 billion. In addition, while the differences between the accounting and funding measures were less than anticipated, there will likely be substantial future volatility in the accounting measures due to the measure of plan assets (i.e., the FNP) being based on the fair (market) value. 9 This is likely due to plans establishing or updating funding policies to include closing their amortization periods, but further research is needed to confirm. 2/25/ Gabriel Roeder Smith & Company -6-