With Americans Living Longer, Pension Changes May Be Needed

Faced with compelling new evidence that Americans are living longer, companies may have to reflect potentially big cost increases associated with pension and retiree medical benefits in their financial statements as soon as this calendar year-end.

The Society of Actuaries has released a series of new tables and other data that show people who have retiree pension and medical benefits live longer, and that longevity will continue to increase into the future. Among men, overall longevity for those who reach the age of 65 rose 2 years to 86.6 years. For women, longevity rose 2.4 years to 88.8 years.

SOA mortality data has long been regarded by the Internal Revenue Service and other authorities as key evidence that can and should be considered in projecting costs for pension and other post-employment benefits, such as retiree medical insurance. Based on the latest data, that means roughly two more years of benefits to be paid to retirees than most companies are likely banking on today.

“Depending on the tables you were using previously, we’re seeing 3- to 10-percent increases to some plans. It can be more than 10 percent if you were using older tables.”
Jack Abraham, Principal, PwC

“This will have a very big financial impact,” says Jack Abraham, a principal in PwC’s human resource practice and leader of the U.S. retirement practice. “Depending on the tables you were using previously, we’re seeing 3- to 10-percent increases to some plans. It can be more than 10 percent if you were using older tables.” Companies won’t recognize such cost increases through earnings, but through other comprehensive income, he says. “It will be a direct hit to the balance sheet.”

The recent update, undertaken in a five-year study, is the first since 2000. “We conducted a lot of research and solicited data from a lot of large pension plans in the United States,” says Donald Fuerst, senior pension fellow at the American Academy of Actuaries and a fellow with the SOA. “We learned that mortality over the last 14 years was improving. Actually, we knew that, but it’s even faster than we anticipated. We found an even more rapid improvement than expected, so that’s why these tables are a little bit of a surprise.”

For purposes of calculating the funding obligation for pension and other post-employment benefits, companies follow guidance from the IRS that specifies exactly what mortality evidence must be considered, says Ray Berry, an actuary with Grant Thornton. “The IRS almost certainly will use these, but that will take time,” he says. The IRS is not likely to adopt the new tables for at least a year or two, he says. “Funding in the long term will be affected, but not immediately.”

POTENTIAL IMPLICATIONS

Below, KPMG explains the possible implications of the SOA’s new mortality rates and offers some guidance to companies on what steps to take next.
The effect of plan sponsors assuming improved mortality for plan participants is that their defined benefit obligations will increase. Some companies may experience a significant increase of 5 to 10 percent. Under U.S. GAAP, this increase in the obligation will be reflected in other comprehensive income unless companies have adopted an immediate recognition in earnings approach in which case the effect will be recorded in earnings. If the amount added to accumulated other comprehensive income is greater than the corridor, then the excess should be amortized in 2015, which may be material for some companies. Under IFRS, the increase is reflected as a remeasurement adjustment that is recorded in other comprehensive income. It is not amortized in future periods.
Plan sponsors that previously relied on older data as their best estimate—e.g., RP-2000 mortality tables with Scale AA projection—may see a greater increase in their defined benefit obligation than those using newer data—e.g., RP-2000 mortality tables with Scale BB projection. Increasing the number of years used in a projection scale will also increase the obligation.
Changing mortality assumptions also may have other effects on a defined benefit plan.

As participants live longer, the duration of the benefit obligation increases, which makes the obligation more sensitive to changes in interest rates.

Increased duration and interest rate sensitivity of the benefit obligation may cause plan sponsors to adjust their investment strategy and asset allocation.

If the IRS adopts the new SOA mortality tables, sponsors could see an increase in funding liabilities, contributions, and PBGC premiums.
Next Steps
Plan sponsors should familiarize themselves with the SOA exposure drafts of RP-2014 and MP-2014 and begin the process of developing their best estimate of mortality. Changes in mortality assumptions from those used previously may have a significant effect on the measurement of the benefit obligation.
Some companies may wish to have their actuaries conduct a plan-specific study of mortality experience to obtain credible data to support potential deviations from the new SOA mortality data.This is more likely to produce useful information for plans with large numbers of participants. Because conducting a mortality study may be complex and time-consuming, plan sponsors may wish to begin this work as soon as possible.
Source: KPMG.

That may enable companies to dodge the cash flow bullet in the short term, but the financial statement effect is different. For financial reporting purposes, companies are required to follow U.S. Generally Accepted Accounting Principles, which tell management to show in their financial statements their best estimate of their projected pension and post-employment benefit obligations based on available evidence.

The SOA tables provide pretty credible evidence, but there’s nothing prescriptive in accounting standards that tells plan administrators or sponsors that they must be followed, says James Verlautz, a principal with consulting firm Mercer. “The tables at this point are a resource,” he says. “The requirement in professional standards is to consider a group and come up with a mortality table that you think is appropriate for that group.”

Considering the Evidence

Larger companies sometimes have large enough populations and reliable enough evidence from administering their own plans that they may have convincing mortality evidence of their own, says Verlautz. Mercer is in the midst of developing tables that would address mortality for specific industry sectors or occupations, he says, although it’s not clear whether that would be informative to companies facing year-end reporting deadlines. “Assumption setting is always based on facts and circumstances,” he says. “Management is going to have to look at its own facts and circumstances and decide. We’re going to see lots of cases, particularly smaller companies, that simply use the new tables.”

Karen Wiltsie, an audit partner at Deloitte & Touche, says auditors will be looking at whether companies have considered the new evidence in arriving at their assumptions. “This is significant new information that could definitely impact the mortality assumption that companies use in their measurements,” she says. But it’s not the only information, she says. SOA’s 2000 tables, for example, combined with appropriate longevity improvement adjustments, might also be valid, she says. “So it might still be possible in some circumstances to use an older base table and project it forward if there is a reasonable basis for doing so.”

Abraham says it’s “almost a rebuttable presumption” that companies will be expected to follow the new tables or show good reason why they shouldn’t. And there are two ways to show that, he says—quantitative data on their own experience or qualitative information to show that SOA tables do not match their own population. As an extreme example, a company providing benefits to coal miners or those who work in certain heavy industries might be able to slow lower mortality rates, he says.

Another consideration, says Verlautz, is the importance of consistency in assumptions and projections. “It’s difficult for a company to say I’m going to use one set of mortality information for corporate reporting and a different set for plan reporting,” he says. “You need one perspective for plan mortality.” Some companies may determine the new tables are too new and merit further consideration before adopting at year-end. “It’s unfortunate that these tables came out when they did because there’s not much time to figure out if they make sense for a given company, but there will be a lot of pressure to use these or tell us why not.”

Wiltsie says companies should be discussing the new tables with their actuaries and their auditors, and they should be reviewing their internal controls for arriving at assumptions. “Auditors will be asking questions about the company’s process for determining their best estimates for mortality, so companies should be thinking about that,” she says. “Even if the process is to look at obtaining the assistance of an outside actuary, ultimately the company is responsible for those assumptions.”