Financial reporting considerations related to pension and other postretirement benefits
Financial Reporting Alert 17-7
This publication highlights some of the important accounting considerations related to the calculations and disclosures entities provide under U.S. GAAP in connection with their defined benefit pension and other postretirement benefit plans.
Presentation of Net Periodic Benefit Cost
On March 10, 2017, the FASB issued ASU 2017-07, which amends the requirements in ASC 715 related to the income statement presentation of the components of net periodic benefit cost for an entity’s sponsored defined benefit pension and other postretirement plans.
Under current U.S. GAAP, net benefit cost (i.e., defined benefit pension cost and postretirement benefit cost) consists of several components that reflect different aspects of an employer’s financial arrangements as well as the cost of benefits earned by employees. These components are aggregated and reported net in the financial statements.
ASU 2017-07 requires entities to (1) disaggregate the current-service-cost component from the other components of net benefit cost (the “other components”) and present it with other current compensation costs for related employees in the income statement and (2) present the other components elsewhere in the income statement and outside of income from operations if such a subtotal is presented.
The ASU also requires entities to disclose the income statement lines that contain the other components if those components are not presented on appropriately described separate lines.
Connecting the Dots
While the ASU does not require entities to further disaggregate the other components, they may do so if they believe that the information would be helpful to financial statement users. However, entities must disclose which financial statement lines contain the disaggregated components.
In addition, only the service-cost component of net benefit cost is eligible for capitalization (e.g., as part of inventory or property, plant, and equipment). This is a change from current practice, under which entities capitalize the aggregate net benefit cost when applicable.
The ASU’s amendments are effective for public business entities for interim and annual periods beginning after December 15, 2017. For other entities, the amendments are effective for annual periods beginning after December 15, 2018, and interim periods in the subsequent annual period. Early adoption is permitted.
Entities must use (1) a retrospective transition method to adopt the requirement for separate presentation in the income statement of service costs and other components and (2) a prospective transition method to adopt the requirement to limit the capitalization (e.g., as part of inventory) of benefit costs to the service cost component. Further, entities must disclose the nature of and reason for the change in accounting principle in both the first interim and annual reporting periods in which they adopt the amendments.
The ASU also establishes a practical expedient upon transition that permits entities to use their previously disclosed service cost and other costs from the prior years’ pension and other postretirement benefit plan footnotes in the comparative periods as appropriate estimates when retrospectively changing the presentation of these costs in the income statement. Entities that apply the practical expedient need to disclose that they did so.
For more information, see Deloitte’s March 14, 2017, Heads Up.
Over the past few years, we have provided insights into approaches used to support discount rates for defined benefit plans (e.g., hypothetical bond portfolio, yield curve, index-based discount rate), as well as considerations related to how the discount rates should be applied when an entity measures its benefit obligation. Recently, one of the most discussed emerging issues related to discount rates for defined benefit plans has been the use of a more granular approach to measure components of benefit cost. Considerations related to an entity’s discount rate selection method, its use of a yield curve, and its measurement of components of benefit cost are addressed below.
Discount Rate Selection Method
ASC 715-30-35-43 requires the discount rate to reflect rates at which the defined benefit obligation could be effectively settled. In the estimation of those rates, it would be appropriate for an entity to use information about rates implicit in current prices of annuity contracts that could be used to settle the obligation. Alternatively, employers may look to rates of return on high-quality fixed-income investments that are currently available and expected to be available during the benefits’ period to maturity.
One acceptable method of deriving the discount rate would be to use a model that reflects rates of zero-coupon, high-quality corporate bonds with maturity dates and amounts that match the timing and amount of the expected future benefit payments. Since there are a limited number of zero-coupon corporate bonds in the market, models are constructed with coupon-paying bonds whose yields are adjusted to approximate results that would have been obtained through the use of the zero-coupon bonds. Constructing a hypothetical portfolio of high-quality instruments with maturities that mirror the benefit obligation is one method that can be used to achieve this objective. Other methods that can be expected to produce results that are not materially different would also be acceptable — for example, use of a yield curve constructed by a third party such as an actuarial firm. The use of indexes may also be acceptable.
Connecting the Dots
In determining the appropriate discount rate, entities should consider the following SEC staff guidance (codified in ASC 715-20-S99-1):
At each measurement date, the SEC staff expects registrants to use discount rates to measure obligations for pension benefits and postretirement benefits other than pensions that reflect the then current level of interest rates. The staff suggests that fixed-income debt securities that receive one of the two highest ratings given by a recognized ratings agency be considered high quality (for example, a fixed-income security that receives a rating of Aa or higher from Moody’s Investors Service, Inc.).
Entity’s Use of a Yield Curve
To support its discount rate, an entity may elect to use a yield curve constructed by an actuarial firm or other third party. Many yield curves constructed by actuarial firms or other third parties are supported by a white paper or other documentation that discusses how the yield curves are constructed. Management should understand how the yield curve it has used to develop its discount rate was constructed as well as the universe of bonds included in the analysis. If applicable, management should also evaluate and reach conclusions about the reasonableness of the approach the third party applied to adjust the bond universe used to develop the yield curve.
We have been advised by some third parties, particularly those constructing yield curves for non-U.S. markets (e.g., the eurozone and Canada), that because of a lack of sufficient high-quality instruments with longer maturities, they have employed a method in which they adjust yields of bonds that are not rated AA by an estimated credit spread to derive a yield representative of an AA-quality bond. This bond, as adjusted, is included in the bond universe when the third party constructs its yield curve. Management should understand the adjustments made to such bond yields in the construction of those yield curves and why those adjustments are appropriate.
Measurement of Interest Cost Component
In the past year, the most discussed emerging issue related to discount rates has been the alternatives for applying discount rates under a bond-matching approach (sometimes also referred to as a hypothetical bond portfolio or bond-model approach). In light of the SEC staff’s acceptance of the use of a spot rate approach for measuring interest cost by entities that develop their discount rate assumption by using a yield curve approach, entities and actuaries have explored whether other acceptable methods similar to the spot rate approach could be developed for entities that use a bond-matching approach to measure their defined benefit obligation. Specifically, the alternative approach focuses on measuring the interest cost component of net periodic benefit cost by using individual spot rates derived from an acceptable high-quality corporate bond yield curve and matched with separate cash flows for each future year.
During the spring and early summer of 2016, representatives of the Big Four accounting firms and a large actuarial firm engaged in discussions with the SEC staff regarding the viability of a similar granular approach to measure interest cost for registrants that use a bond-matching approach to support the discount rate. In an August 2, 2016, meeting, the SEC staff stated that it objected to the approach presented because of the following factors:
- The staff’s overall concern is that using such derived spot rates to measure interest cost on the defined benefit obligation could not be demonstrated, at each maturity, to be based on the same rates inherent in the measurement of the defined benefit obligation under the bond-matching approach (i.e., the spot rates inherent in the bond portfolio are not observable). Therefore, the proposed approach would fail to comply with ASC 715-30-35-8, which requires entities to use the same interest rates to measure the defined benefit obligation and interest cost.
- The staff also expressed concern that the derived spot rates in the proposed approach would be inconsistent with the reinvestment-rate assumption used in the cash flow matching process that is part of building the cash flow matched hypothetical bond portfolio used to measure the defined benefit obligation under a bond-matching approach.
Connecting the Dots
We believe that in the absence of entity-specific changes in facts and circumstances, it could be challenging to justify or support a change from a bond-matching approach to a yield curve approach. Historically, entities have generally made the switch only from a yield curve approach to a bond-matching approach, which suggests that of the two methods, the bond-matching approach results in a better estimate. This historical practice, along with the SEC staff’s position that the acceptability of the spot rate approach would not by itself be a change in facts and circumstances that justifies a change in approach to selecting discount rates, reduces the likelihood that switching from a bond-matching approach to a yield curve approach would be considered a better estimate in accordance with the bestestimate objective of ASC 715. For further background on a change in approach to determining discount rates, see Deloitte’s August 24, 2016, and December 21, 2015, Financial Reporting Alert newsletters.
Many entities rely on their actuarial firms for advice or recommendations related to demographic assumptions, such as the mortality assumption. Frequently, actuaries recommend published tables that reflect broad-based studies of mortality. Under ASC 715-30 and ASC 715-60, each assumption should represent the “best estimate” for that assumption as of the current measurement date. The mortality tables used and adjustments made (e.g., for longevity improvements) should be appropriate for the employee base covered under the plan.
Mortality Tables Used for IRS Tax-Qualified Plans
On October 4, 2017, the IRS issued final regulations prescribing mortality tables to be used by most defined benefit pension plans. The purpose of these mortality tables is to determine (1) the minimum funding requirements for a defined benefit plan and (2) the minimum required amount of a lump-sum distribution from such a plan. The regulations became effective on October 5, 2017, and apply to plan years beginning on or after January 1, 2018.
For defined benefit pension plans (particularly IRS tax-qualified plans) that permit settlement of the obligation to an employee through payment of a lump sum at retirement, entities generally compute the payment by using IRS-mandated tables in effect on the date of the lump-sum payment. Similarly, for qualified cash balance plans, if an employee elects to convert the lump-sum benefit amount at retirement to an annuity, the entity uses IRS-mandated tables to calculate the annuity. In making assumptions about either the amount of future lump-sum benefits expected to be paid or any annuities expected to be paid that are related to a cash balance plan, entities have questioned whether they should base these assumptions on the IRS’s practice of annually updating the current tables with an additional year of longevity improvement as well as on the IRS’s expected future adoption of new tables that are updated on the basis of the latest available mortality tables published by the SOA.
We believe that there are two acceptable approaches under U.S. GAAP that entities can use to account for the impact of the IRS’s expected adoption of revised mortality tables. Under one view that we believe is supportable, entities would reflect their best estimate of the future IRS tables, taking into consideration both the recent IRS regulations and the IRS’s history of annual updates to its tables. This approach is consistent with the guidance in ASC 715-30-35-31, which indicates that indirect effects on the amount of a benefit, such as future changes in Social Security benefits or benefit limitations required by existing laws, should be taken into account in the measurement of the defined benefit obligation (although amendments to a law should not be anticipated).
Under an alternative view, entities would not anticipate future updates to the IRS-mandated mortality tables in performing measurements related to lump-sum payments because the IRS’s update to its mortality tables is akin to a new law or regulation, which should not be anticipated. This view only pertains to the effects of the IRS’s update to its tables to be used in compliance with the regulatory requirements for measuring lump-sum settlements for tax-qualified plans and is not related to an entity’s determination of its best estimate of the mortality assumption for those plans.
We believe that both approaches are acceptable under U.S. GAAP and that an entity should be consistent in applying the chosen approach. However, if an entity chooses the alternative approach of not incorporating the effects of new mortality data in its estimates of future lump-sum settlements for an IRS tax-qualified plan and the results of applying the two respective approaches are expected to differ materially, the entity should consider consulting with its independent auditors.
View the rest of the Financial Reporting Alert.
Subscribe and Archives
Financial Reporting Alerts highlight up-to-the-minute accounting, regulatory, or other developments that may require immediate action by our professionals and clients or that may affect current financial reporting and disclosure.