Why Don’t People Insure Late-Life Consumption? A Framing Explanation of the Under-Annuitization Puzzle

The researchers conducted a four-arm internet survey. The sample frame was a panel of respondents over the age of 50 that had agreed to participate in surveys for a small incentive. The survey consisted of seven forced-choice questions that described financial products worth $100,000 purchased by two fictitious people for retirement income and asked, “Who has made the better choice”? Five products were described, all actuarially equivalent. The first product was a life annuity paying $650 each month until death, after which any remaining principal was forfeit. The second was a traditional savings account earning 4% APY. The third was a consol bond paying $400 per month. The fourth was a thirty-five year annuity paying $500 per month. The fifth and last was a twenty-five year annuity paying $650 per month. The fictitious individuals were 50 years old, had already provided an inheritance for their children, and would be receiving $1000 per month in Social Security benefits.

2007-12-01

2008-01-01

the percentage of individuals who preferred a life annuity to actuarially equivalent alternative options

In the first survey arm, the decisions were presented from a consumption frame, i.e. how much money each person could spend per month after purchasing that product. The remaining principal or payments could be donated to charity after death, except in the case of the life annuity. In the second arm, the decisions were also framed in terms of consumption, but the remainder after death could be left to descendants instead. In the third survey arm, the decisions were presented from an investment frame, i.e. how much money each product earns per month. After death, the remainder could be donated to charity, except in the case of the life annuity. In the fourth survey arm, the decisions were also presented from an investment frame, but the remainder after death could be left to descendants.

1342 individuals took part in the survey, with roughly 335 individuals in each survey arm. In all arms, the options were referred to in general terms of amounts and durations. Specific terms such as annuity, savings account, or bond were not used. In each arm, the life annuity was compared separately to the other products. Other comparisons not featuring the life annuity were included to avoid spurious effects. The order of the options was varied as well.

randomization performed in office on computer

Individual

No

1342 individuals

1342 individuals

about 335 individuals per arm

Rational models of risk-averse consumers have difficulty explaining limited annuity demand. We posit that consumers evaluate annuity products using a narrow "investment frame" that focuses on risk and return, rather than a "consumption frame" that considers the consequences for lifelong consumption. Under an investment frame, annuities are quite unattractive, exhibiting high risk without high returns. Survey evidence supports this hypothesis: whereas 72 percent of respondents prefer a life annuity over a savings account when the choice is framed in terms of consumption, only 21 percent of respondents prefer it when the choice is framed in terms of investment features.

Brown, Jeffrey R., Jeffrey R. Kling, Sendhil Mullainathan, and Marian V. Wrobel. 2008. "Why Don't People Insure Late-Life Consumption? A Framing Explanation of the Under Annuitization Puzzle." American Economic Review: Papers Proceedings 98(2): 304-9.

https://www.aeaweb.org/articles?id=10.1257/aer.98.2.304