In a new study, researchers were able to influence someone’s estimate of their lifespan — by more than nine years — simply by asking them if they expected to “live to” versus “die by” a certain age. In addition, the way the question was posed affected people’s preferences for life annuities, which provide insurance for people who outlive their savings.
“We wanted to know whether beliefs about how long one might live would differ with simple changes in framing, and if so, how big is the effect,” says John Payne, a Duke University business professor. “The answer is yes, it does matter, and the effects are huge.”
Why Semantics Matter
Payne and colleagues from Duke, Columbia University and the University of California at Los Angeles surveyed more than 2,000 respondents about their probable longevity. Using a slider scale, one group was asked to estimate the percentage chance that they would live to a certain age; the other group was asked about their chances of dying by a certain age.
Half the participants received scales that read 0% on the left and 100% on the right; the other half saw a scale that read 100% on the left and 0% on the right. All of the participants provided their estimates on whether they would live or die by the time they are 75, 85 and 95 years old.
If the way the question is posed makes no difference, the two group’s answers should have matched up — i.e., a 60% chance of living would be matched by a 40% chance of dying. But that’s not what happened.
People considering how long they would live were more optimistic, estimating that they had a 55% chance of being alive at age 85. Those thinking about when they would die were more pessimistic, estimating a 68% chance of dying at 85 — or just a 32% chance of being alive. Overall, there was also a 10-year gap in median expected age of death (around 85 years for the live-to group and 75 years for the die-by group).
“This 10-year difference in the median expected age of being dead or alive is not only statistically significant, but also highly meaningful to a number of important life decisions, such as how to finance one’s consumption during retirement,” the researchers write.
Based on judged probabilities, estimated mean life expectations were 9.38 years longer when presented in the live-to frame. “In terms of decisions about how much I need to retire, when I need to claim Social Security benefits and what I want to do in managing my retirement, this is a big number,” Payne says.
The Bottom Line
How do longevity expectations influence financial choices? In a separate study, the researchers asked respondents about life annuities – an investment that requires a large, immediate and nonrefundable payment in exchange for a stream of monthly payments, guaranteed for life. Obviously, the sooner you expect to die, the less attractive an annuity investment becomes.
Participants were presented first with an online brochure titled, “How to Invest for Retirement.” It described a life annuity and a self-managed retirement account. Then participants completed the same life-expectancy task, with half in the “live to” frame and half in the “die-by” frame.
On average, individuals in the live-to frame judged their probability of being alive at age 85 at 52%, while those in the die- by frame judged their probability of being alive at 85 at 30%. As with the previous studies, there was an approximately 10-year gap in the median expected age of death.
Next, half of these participants evaluated a single life annuity and reported the likelihood of a purchase. On average, 39% of those who judged themselves likely to live longer said they were likely to buy, while only 26% for those who expected to die sooner planned to buy.
“Individuals in the live-to frame, in particular, showed the strongest sensitivity to estimated life expectations in their annuity preferences,” the researchers write. That suggests that the way the longevity question is framed — and the way you think about your life expectancy — weighs heavily on decisions about future financial outcomes.
Tagged: annuities, annuity, columbia, Columbia University, ColumbiaUniversity, Duke, duke university, DukeUniversity, life annuities, life expectancy, LifeAnnuities, LifeExpectancy, research, retirement,