Life Expectancy as a Constructed Belief: Evidence of a Live-To or Die-By Framing Effect

Abstract

Expectations about how long one will live are essential for making informed choices about many important personal decisions. We propose that beliefs (expectations) about longevity are a response constructed at the time of judgment, subject to irrelevant task and context factors, and leading to predictable biases. Specifically, we examine whether life expectancy is affected by the framing of expectations questions, as well as by factors that actually affect longevity such as the age, gender, and self-reported health status of the respondent. One frame asks people to provide probabilities of living to a certain age or older; the other frame asks people to provide probabilities of dying by a certain age or younger. These two answers should be complements, but we find that estimated probabilities differ significantly in the two conditions. People in the live-to frame report that they have a 55% chance of being alive at age 85, whereas people in the die-by frame report that they have a 68% chance of being dead at age 85. Overall, estimated mean life expectancies, across three studies and over 2300 respondents were between 7.29 to 9.17 years longer when solicited in the live-to frame. We compare estimated life expectancies with Social Security Administration (SSA) life tables and find that the judgments of individuals in the live-to frame were closer to actual life expectancies for ages 65 and 75, while in the die-by condition, the respondents were more accurate for older ages, e.g., age 95. On a process level, we show that the framing effect on judgments is partially mediated by the relative number of thoughts in favor of being alive at that age. Finally, we find that individual differences in life expectancies relate to differences in stated preference for life annuities, a product that provides insurance against outliving one’s savings. The implications of “constructed” life expectancies for models of financial decision making, and for improving financial decision making are discussed.