Mortality Tables Impact Depends on DB Plan Demographics

September 5, 2014 ( – Goldman Sachs Asset Management’s conversations with actuaries and clients indicate many defined benefit (DB) plans could expect an increase in liabilities of between 5% and 10% if proposed mortality tables are put into effect.

In a research article, “Challenging First Half of the Year for Corporate DB Plans,” Michael Moran, senior pension strategist at Goldman Sachs Asset Management in New York City, says any actual increase to an employer’s pension plan obligations as a result of mortality table changes will be heavily dependent upon the specific demographics of its plan participants. “The increases in life spans are not evenly distributed,” Moran tells PLANSPONSOR. “That is, we’re not all living longer at the same rate.”

He explains that certain cohorts will see a greater increase in expected life span under the new mortality tables. For example, females will see a greater increase in life expectancy than men; older people will see a greater increase than younger people. “The breakdown of a plan’s participants—older versus younger, male versus female, white collar versus blue collar—will have a significant influence on any change in liabilities.” Moran notes that companies with a higher female or older population may see liability changes higher than that of companies with more males or younger DB plan participants.

Goldman Sachs is seeing DB plan sponsors working with actuaries and looking at plan demographics to see what their expected change in liabilities will be. “At this point, most companies haven’t been publicly disclosing their findings, but Northrup Grumman, for example, said it expects its liabilities to increase somewhere between 4% and 6%,” Moran notes.

According to Moran, one reason plan sponsors are looking at this now is that if—and he says it’s still a big if—the mortality table proposal is finalized this year as the Society of Actuaries is urging, actuaries and auditors will expect plan sponsors to use the new tables in calculating impacts on financial statements this year. “The effect on funding status would immediately show on their balance sheets this year.”

Another reason is many plans’ liability-driven investment strategies use a glide path in which investment allocations change depending on plan funded status, so the effect of new mortality tables on plan liabilities—and therefore, plan funded status—may call for an adjustment of investment allocation (see “Planning for Mortality Tables’ Effect on DB Liabilities”).

However, Moran notes in his article that the mortality proposal is proving to be controversial, which may slow down the process. The proposal was issued in February and the comment period ended in May. During that time, Moran says, a number of employer groups and actuarial firms submitted letters indicating concern about the data the society used to draw conclusions. “The society started out looking at a universe of 120 plans, then eliminated some for various reasons, until there were ultimately only 38 in its analysis,” he explains. “They eliminated about 70% of plans, and when they looked at retiree experience, two-thirds of the data comes from just five plans. Many actuarial firms and plan providers are concerned that the society seems to have used a limited supply of data.”

In its comment letter, human resources consulting firm Mercer said it was concerned that the treatment of the data used in development of the tables may have skewed the results to show lower mortality than would otherwise be the case. It also expressed concern that the analysis does not reflect data annually obtained in mortality studies by the Pension Benefit Guaranty Corporation (PBGC), “which encompasses a broader range of data than that used by the [society’s retirement plan experience committee] (especially after data exclusions).” Mercer said use of the PBGC data “would substantially enhance confidence in the resulting mortality rates, especially for current benefit recipients.”

According to Moran, at the end of July, the society said it received some questions about the data it used, but reiterated its goal is to issue responses to comments and final tables by October 31 of this year. “While they didn’t extend their timeline, they reserved the right to change it,” Moran notes. The society could decide to go back and expand their data.

No matter when the tables are finalized, DB plans’ funded status will take a hit. At the end of 2013, plans were, on average, about 90% funded, “which was a nice increase from 2012,” Moran says. Given the declining interest rate environment, Goldman Sachs recently estimated current funded levels have dropped to 86%, or down to 80% if mortality tables are finalized as currently contemplated. “If we think about where we were during the depths of the financial crisis, we’re not much better off when we consider the new mortality tables. The new deficits will be just as large as those from several years ago,” he notes.

Moran also says it is a good time for DB plan sponsors to consider whether they want to re-risk or de-risk their plans (see “New Mortality Tables Will Impact Pension Plan Management”).

The proposed mortality tables can be viewed here.