Two weeks ago, the Society of Actuaries (SOA), a professional association of 24,000 actuaries, updated the mortality tables it released last fall.
The latest tables show an “ever so slightly lower” rate of improved mortality than what was calculated in 2014, according to Dale Hall, managing director of research at SOA.
The longevity projections the SOA released in 2014—the tables were last updated in 2000—showed the average 65-year old male would live to be 86.6, and the average 65-year old woman could expect to live to be 88.8.
Life span is increasing, the tables showed: 2.4 more years for women and 2.0 more years for men, relative to the projections made in 2000. Those estimates had massive implications for sponsors of defined-benefit pension plans.
As reported in the Wall Street Journal, the 2014 longevity projections created a new US$2.2 billion shortfall in General Motors’ pension funding. Other large plan sponsors, like Verizon and AT&T, also noted substantial write-downs in their earnings reports resulting from the new mortality tables. Towers Watson estimated the largest 400 pension plans in the United States would take a collective $72 billion hit.
When the SOA released its findings last October, it did so against the strains of sponsor advocates. In a letter to Treasury officials after the SOA released a draft of proposed new tables in the spring of 2014, Lynn Dudley, senior vice president at American Benefits Council (ABC), a trade group that represents sponsors of the 100 largest pension plans, argued that the SOA was overstating just how much longer people were living in its new tables.
“The overstatement of life expectancy in the SOA tables, if reflected in regulations issued by Treasury and the Service, would cause defined benefit pension plan liabilities (and lump sums and other optional forms of payments) to be overvalued by tens of billions of dollars nationally, triggering several adverse effects,” wrote Dudley.
“Many plan sponsors would be forced to overfund their plans and to cut or freeze benefits, and would put additional pressure on sponsors to de-risk more quickly to avoid inappropriate liability increases. And, of course, the overfunding required would divert scarce assets to inflated pension liabilities and away from business investment and jobs,” added Dudley.
Basically, sponsors were arguing that the SOA used incomplete data in estimating the new tables. That led to an excessive estimation of just how long retirees will be expected to live, and effectively rendered the SOA’s mortality improvement scale “speculative,” according to the ABC.
Sponsors were not alone. A growing chorus of actuaries also voiced complaints about the SOA’s process. Amid last year’s draft proposal and comment period, the American Academy of Actuaries, another professional association with 18,000 actuary members, wrote the SOA to voice concerns similar to those raised by sponsors and the ABC.
Ultimately, the SOA made some changes to its proposed new tables, but many stakeholders’ core concern remained after the tables were released: that the SOA had used incomplete data sets, and that resulted in an overstatement of improved mortality rates, an error that would cost sponsors billions in funding liabilities.
The back and forth between the SOA, which continued to defend its work, and sponsor advocates and Treasury continued into 2015. When Treasury announced it was delaying the implementation of the new tables this past summer, this indicated to some that it was reviewing the accuracy of the projections. Ultimately, the SOA would make adjustments to its table—the ones released last week.
Americans are living longer, just not as “longer” as the 2014 tables projected: the average 65 year-old male is now expected to live to be 86.6, and the average female 88.2.
Dale Hall says that means a reduction in sponsors’ liabilities of up to 2 percent compared with the 2014 projections. The recently updated tables are indicative of a new priority at the SOA to provide more timely longevity research, he said.
“Longevity is clearly one of the top risks the world faces,” said Hall. “This is a huge priority for us. We expect to come out with new research annually that fills in the most current set of data,” he added.
He said the experience in spending years to finally update the new tables, and then to soon after adjust them, involved improved communications with regulators, stakeholders in the private sector, and throughout the actuarial community.
Hall also said the SOA is building on that momentum, and currently mining more sources for mortality data from the Center for Disease Control and the Centers for Medicare and Medicaid to build the best modeling going forward.
As for plan sponsors, Lynn Dudley said the recently adjusted tables are still being analyzed. “I have nothing but respect for the Society of Actuaries,” she said. “We have the best actuaries in the world, and sponsors have faith in them too.
“But they all have to be on the same page,” she added. “When the actuaries are all on board together, I think industry will support whatever data is produced.”
The stakes for sponsors and participants are simply too high to have dissent in the actuarial community, she said.