Volume 40 | Issue 04
The IRS has released proposed regulations on applicable mortality tables to be used by defined benefit plans subject to ERISA to determine minimum funding requirements beginning with 2018 plan years. When finalized, the revised tables will also form the basis for the unisex tables used for minimum lump-sum values and maximum benefits under Code Section 415. The proposed rules generally adopt the Society of Actuaries’ RP-2014 tables and MP-2016 projection scale. Tables reflecting static projections of longevity improvement would still be permitted, as would combined annuitant/nonannuitant (small plan) versions of the tables. Updated rules for substitute, plan-specific mortality tables are also proposed, which would permit the use of tables reflecting partially credible experience for plans that are too small to have fully credible experience.
Defined benefit plan funding rules established in the Pension Protection Act of 2006 (PPA) included the requirement to use mortality tables approved by the IRS to determine minimum funding levels for single-employer plans. In addition to funding, the IRS tables are used in determining a plan’s funded position (or AFTAP) under the benefit restriction rules of Code Section 436, and PBGC variable-rate premiums. They are also the basis for unisex mortality tables used to determine minimum lump-sum payments under Code Section 417(e) and maximum benefits under Code Section 415. PPA also required the use of specified mortality tables in determining current liability for multiemployer plans. In lieu of the IRS tables, PPA provided that plan-specific tables could be used for some purposes, if certain conditions were met.
In 2008, the IRS issued final regulations defining the requisite mortality tables based on the RP-2000 Mortality Tables Report released by the Society of Actuaries in 2000, and associated rules for plan-specific tables. PPA mandates that the IRS update the tables at least every 10 years to reflect actual participant mortality experience since the last update, and projected trends in that experience.
In 2014, the Society of Actuaries issued a new mortality study of participants in private pension plans, releasing both the RP-2014 mortality tables and the MP-2014 mortality improvement scales. The MP-2014 improvement scales were updated in October 2015 (MP-2015) and again in October 2016 (MP-2016). See our October 20, 2016, October 8, 2015 and October 27, 2014 issues of FYI Alert for discussion of the report and updates.
For plan years beginning after December 31, 2015, the Bipartisan Budget Act of 2015 changed the rules on the use of substitute mortality tables to allow more plans to reflect adjustments to the generally applicable mortality table based on the actual experience of the pension plan maintained by the sponsor.
IRS Proposes Update for 2018
IRS proposed changes to the 2008 regulation would update the PPA mortality table based on the RP-2014 mortality tables and the improvement rates contained in the Mortality Improvement Scale MP-2016 Report. Recall that the 2016 update lowered anticipated life expectancy improvements, thus moderating the increases in liabilities that would have been produced using earlier improvement scales.
The method for developing substitute mortality tables under the proposed regulations would be simpler than the method that applies under the 2008 regulations, and allows the use of substitute mortality tables by plans with smaller populations that have only partially credible mortality experience. As under current regulations, substitute tables would have to be submitted to the IRS for approval before use.
Generally Applicable Mortality Tables
Like the 2008 mortality tables, the tables in these regulations provide mortality rates for males and females, and separately for annuitants (for the period after assumed benefit commencement) and nonannuitants (for the period before assumed benefit commencement).
Like the current regulation, the proposal would require expected mortality improvements to be reflected in one of two ways – a static projection (one table for all participants updated annually) or a fully generational projection (projected to all future years from the base table for each participant). The regulations would provide static tables for use in 2018 valuations designed to approximate the result determined by the generational tables. Updated static tables will be provided by the IRS annually taking into account the updates the SOA is expected to issue each year to reflect mortality improvement trends for the general population.
Small plans (i.e., those with 500 or fewer active and inactive participants as of the valuation date) are permitted to use a single blended (annuitant and nonannuitant combined) static table for all participants. The proposed regulations would use the same male and female weighting factors as the 2008 regulations to create these blended rates.
Plan-Specific Mortality Tables
The use of plan-specific mortality tables (referred to as substitute tables) instead of the generally applicable mortality tables would continue to be permitted, if certain conditions are met.
Substitute tables must be generational (based on the IRS-prescribed projection scale), gender-specific, and must reflect the actual “credible” experience of the plan. To be fully credible, the experience must be based on more than 1,082 deaths within a gender over the period of the mortality experience study, with the new higher thresholds based on benefit dispersion within the population. New rules, referred to as the partial credibility rules, are proposed for plans with experience that is not sufficient to be fully credible. If the plan has information on at least 100 deaths in a population, the plan would use a weighted average of the standard table and the table that would be developed if fully credible information were available. If credible, or partially credible, experience is only available for one gender for a plan, substitute tables would be used for that gender and the generally applicable tables would be used for the other gender.
Like the current regulation, the proposed regulation contains restrictions and options covering the use of substitute mortality across controlled-group plans, for separate populations such as hourly versus salaried groups, for the aggregation of plans and for handling newly acquired plans. They retain rules from the 2008 regulations on when substitute mortality tables may no longer be used because of changes in credible information, changes in the composition of the controlled group, and similar events.
Anticipated Liability Increases
The updated IRS mortality tables will tend to increase liabilities, but the specific impact will vary depending on the plan design, demographics, and the versions of the tables (static or generational) used before and after the update. Probably the most common scenario will be plans that use static mortality both before and after the change. Based on sample present value factors, we anticipate liabilities will typically increase 3 to 5 percent as a result of changing from the 2017 static mortality to the 2018 updated static mortality.
The incremental liability increase between 2018 static and 2018 generational mortality might be on the order of another 1 percent. However, the structural differences between generational mortality and the IRS methodology for creating static tables make it difficult to predict the impact of such a change.
Going directly from 2017 generational to 2018 generational mortality might increase liabilities 2 to 5 percent, fairly comparable to the static-to-static update. We emphasize that all these figures are estimates and generalizations; individual plan results could differ significantly from these approximations.
Impact on Plan Benefits
Changing how we measure the value of plan benefits doesn’t really change the cost of the plan in most cases if benefits are paid in annuity form. For a plan with a non-hybrid, traditional design, actual costs go up only as a result of increases in actual lifetime payouts over time. If a plan allows lump sums or other distribution options subject to Section 417(e), plan benefits will increase once the new funding table is reflected in the Section 417 applicable mortality table. Similarly, larger amounts will be permitted to be paid when the limits of Section 415 are an issue for a plan.
For hybrid plans such as cash balance plans, much will depend on whether the plan’s annuity options incorporate the Section 417(e) table to determine conversions. But to the extent a plan offers lump sums based on the hypothetical account balance or accumulation and participants lean toward taking distributions in that form, the new tables will not change the cost of the plan over time.
The ability to pay benefits under certain payment options may be affected by the proposed changes. The benefit restrictions of Section 436 are based on the Section 430 funding target liabilities. An increase triggered by the new mortality table could reduce the plan’s AFTAP below one of the thresholds that trigger a benefit payment restriction.
Multiemployer Current Liability, PBGC Variable Rate Premiums and Reporting Requirements
The proposed changes to the Section 430 mortality tables will directly affect certain current liability calculations for multiemployer plans. In addition, PPA liabilities are referenced by various PBGC reporting requirements and are used in determining a plan’s variable-rate premium. Higher liabilities due to the expanded longevity reflected in the updated tables will impact these other determinations accordingly. For example, plan sponsors may encounter higher variable-rate premiums and may be required to report “reportable events” more frequently because they no longer qualify for waivers based on the absence of a variable-rate premium requirement. See our For Your Information issues from March 25, 2016 on changes to 4010 Annual Financial Reporting, and from September 16, 2015 on PBGC’s Reportable Event waivers.
It appears that the audit firms would not expect the proposed regulations to be the basis used for valuing lump-sum distributions for disclosure or expense purposes if the plan has historically used the “current” regulations as its basis. However, plans that have instead used a “best estimate” basis for setting lump-sum assumptions will likely need to consider the proposed regulations in the selection of assumptions for financial statement information.
The regulations are proposed to be effective for plan years beginning in 2018. Although these regulations are not directly applicable to Sections 417(e) and 415, we anticipate that IRS will issue guidance incorporating the regulation as finalized for these other purposes.
A plan sponsor wishing to use substitute mortality tables is generally required to submit a request to IRS at least seven months before the beginning of the plan year in which the substitute tables would be used. If the final regulations are not adopted in time to meet this deadline for 2018 plan years, IRS expects to provide a transition rule allowing for additional time.
More Than Mortality
The proposed mortality tables will increase measured plan liabilities over those that reflected earlier mortality studies. However, appropriate changes in other demographic assumptions may also be considered. For instance, in response to longer life expectancy statistics and the longer period of time retirement savings must stretch, many employees may plan to continue working beyond a plan’s “normal” retirement date instead of retiring in their late 50s or early 60s. Aligning plan retirement assumptions with this new paradigm can potentially offset some of the increase due to the new mortality assumption. This is particularly true for retiree medical plans and pension plans with suspension-of-benefits provisions and generous early retirement subsidies.
Plan sponsors should review their budgets and funding forecasts to determine if they should be updated to reflect the proposed regulations. While the regulations are still only proposed, it is unlikely significant changes will be reflected in the final rules. In addition, plan sponsors should consider whether a plan-specific mortality table would be appropriate. More plans may be able to take advantage of this option now that tables can be based on partially credible data. In particular, plans that primarily cover blue-collar participants may find that the unmodified IRS tables incorporate lower levels of mortality than they realistically expect and thus overstate their liabilities.