Using subjective expectations to forecast longevity: do survey respondents know something we don't know?

Area of Science:

• Demography
• Gerontology
• Health Economics

Background:

• Predicting old-age mortality is complex, requiring understanding of senescence influenced by genetic, environmental, and behavioral factors.
• These factors evolve, making long-term mortality prediction challenging for demographers and policymakers.

Purpose of the Study:

• To investigate if individuals' private information on genetic, environmental, and behavioral factors uniquely qualifies them to predict their own mortality.
• To assess if individual survival expectations offer additional predictive power for mortality trends.

Main Methods:

• Utilized expectation data from the 1992 Health and Retirement Study (HRS).
• Constructed subjective, cohort life tables based on individual mortality predictions.
• Compared these subjective life tables with official U.S. life expectancy revisions from 1992 to 2004.

Main Results:

• Subjective life tables accurately predicted the direction of U.S. life expectancy revisions for both genders between 1992 and 2004.
• Observed a 25% narrowing of the gender gap in longevity during this period, as predicted by subjective data.
• Individual male life expectancies aligned with official estimates, while female expectations suggested official estimates might be high.

Conclusions:

• Individuals possess unique insights into their mortality risks, making their subjective expectations a valuable data source.
• Incorporating individual survival expectations can enhance demographic and policy-level mortality predictions.
• Official life expectancy estimates, particularly for women, may require re-evaluation based on subjective data.