Two life tables, two questions
Life expectancy is calculated from a life table. A period life table takes the age-specific mortality rates observed or projected for one calendar year and applies that single-year pattern across a hypothetical lifetime. It answers a snapshot question: how long would people live if the mortality conditions of this period stayed in place?
A cohort life table follows a birth cohort through the sequence of calendar years in which it reaches each age. Its calculation therefore uses a changing schedule of mortality rates. For cohorts that are not yet extinct, some of those future rates must be estimated. The CDC’s technical description of life tables makes the same distinction, while the SSA publishes both series side by side in Tables V.A4 and V.A5.
Period: one year’s mortality pattern applied to a hypothetical lifetime. Cohort: the mortality pattern encountered as one birth cohort moves through time.
Why the chart uses cohort values
Life Days Left compares birth years. A person born in 1955 and a person born in 2004 do not move through the same calendar years at the same ages. Using a cohort series preserves that distinction: each birth year receives the SSA 2026 Intermediate cohort value attached to that year.
The SSA value is not used as the chart’s complete estimate of age at death. Instead, the model uses the difference between a birth year’s cohort value and the 1967 cohort value as a birth-year adjustment to the income-linked baseline. That construction keeps the selected Chetty baseline fixed while allowing the cohort trend to vary by birth year. The full equation and rounding rules are on the methodology page.
Projection is part of the measure
A complete observed cohort life table can exist only after the cohort’s members have lived through every age. For living and future cohorts, the later-age mortality rates are necessarily estimated. SSA states that its 2026 cohort life expectancy at birth for birth years 1940 through 2023 combines actual, estimated, and projected rates; values after 2023 depend on estimated and projected rates.
That does not make cohort life expectancy a personal prediction. It makes it a model-based population summary. New mortality experience and new Trustees Reports can revise the series. Life Days Left therefore labels the model version and treats source updates as methodology changes, not silent refreshes.
How to read the result
Use the chart to compare the scale and direction of estimates across birth-year bands. Do not read the displayed day as a deadline, a survival probability, or medical guidance. The number has no knowledge of a person’s health, sex, current income, location, behavior, family history, or random events.
The most useful follow-up is to inspect the current valuesand their model version, then decide whether this population-level framing is appropriate for the question you are asking.