Are guaranteed living benefits in a variable annuity a guarantee of a longer life? According to a recent study by Ruark Consulting, LLC, it appears so.
Ruark found that people who purchase variable annuities (VAs) with guaranteed living benefits‑ GMIBs, GLWBs and GMWBs‑live longer than those who don’t buy those riders. In fact, mortality was about 12 percent lower on VA contracts with those guaranteed living benefit clauses than those without such riders, confirmed Peter Gourley, below right, vice president at Simsbury, Conn.-based Ruark, an actuarial consulting firm.
Why that is so is due to a consumer’s perception of his or her lifespan. Say a person has an idea he or she may live a long life, then spending the extra money for that guaranteed lifetime income would make sense, Gourley explained. Conversely, “you are probably not going to buy one if you have some prior knowledge that your longevity might be impaired somehow,” he said.
For the insurance company, lower mortality on VA policies with guaranteed living benefits is good news/bad news, Gourley said. Good news in the sense that a contract is in force longer and therefore, the insurer can collect fees on it for an extended time period. The bad news? Making good on payouts longer than the insurer may have projected.
Yet insurers can mitigate that risk by pricing the annuity contract properly with the right expectations about longevity. “Insurers should probably assume that longevity will be better on these people [who buy the guaranteed living benefits]. So they build that into their price and they won’t have an unpleasant surprise,” Gourley said.
Another unpleasant surprise could come from insurers using the standard industry mortality tables, such as 94MGDB or Annuity2000, to discern mortality rates in VA contracts. The mortality experience for policyholders with guaranteed living benefits embedded in a VA contract is much different than the general population, Gourley noted.
For one, someone who buys a variable annuity tends to be in reasonably good health, “otherwise they wouldn’t purchase the contract,” Gourley said. “And they are probably going to be a bit more affluent than the average person since it’s not necessarily a cheap product. So there are probably some socio-economic factors that show up in the mortality levels.”
Ruark based its findings on data from 17 major insurance companies that contributed more than 30 million policy years of exposure and 340,000 deaths between January 2007 and December 2011.
Ruark did a previous study of mortality rates in variable annuity contracts back in 2007. Between then and now, mortality rates on policies with guaranteed living benefits appear to have decreased. However, Gourley cautioned that comparing the results of the two studies is not analogous since there are far more contracts in force now with those riders. “We think it is due to simply more of these living benefit contracts around that exhibit lower mortality,” he said. For that reason, Ruark maintains that “there is little statistical support for mortality levels declining from 2007 through 2011.”
If there was one anomaly in the study it was that mortality levels generally increase as the policy size increases. Given that wealthier people typically have better access to health care, that would seem to be “counterintuitive to what we might have expected,” Gourley said.
Yet the reason could be that wealthier people who have access to other funds don’t need to tap into their annuity contract when they become ill. Consequently, they can hold onto the policy longer. However, Gourley conceded that theory was “tenuous.”