Actuarial Life Table

An actuarial life table is a table that summarizes the mortality of a population by age. For each age it gives the probability of dying before the next birthday, the number of people still alive out of a starting group (conventionally 100,000 births), and the average number of years of life remaining for those who reach that age. Demographers and government statisticians usually call it a life table; insurers and pension actuaries often call it a mortality table. The Social Security Administration's Office of the Chief Actuary publishes its version under the title "Actuarial Life Table", and the CDC's National Center for Health Statistics publishes the official United States Life Tables each year.

What the columns mean. The CDC's annual report lays out the standard columns. The probability of dying (written q with the age) is the chance that someone who has reached an exact age dies within the next year. The number surviving (l) starts at 100,000 at birth and falls as those probabilities are applied year after year. Further columns count deaths and person-years lived, and the last, expectation of life (e), is the total person-years still to be lived by the survivors divided by the number of survivors. In the CDC's illustration from its 2024 table, males who reach age 20 have an average of 57.4 years remaining.

Period and cohort tables. A cohort table follows a real group of people born in the same year until the last of them dies, which takes a century of data or projections to fill in. A period table instead takes the death rates observed across all ages in one year and asks what would happen to a hypothetical group that lived out its lifetime under those rates. The CDC calls a period table a "snapshot" of current mortality, and it is the kind most people see quoted. Its limitation follows from its design: if death rates keep falling, people born today will on average live longer than a period table built from today's rates indicates, and if rates rise, shorter.

Why life expectancy grows as you age. A newborn's life expectancy averages in every death at every age, including those in infancy and middle age. A 65-year-old has already survived those, so their remaining expectancy is measured only over people who got that far. In the CDC's 2024 table, life expectancy at birth was 79.0 years, while a person aged 65 could expect another 19.7 years, to 84.7, and a person aged 85 another 6.7 years, to 91.7.

SSA's table. SSA's Office of the Chief Actuary publishes a period life table it titles the "Actuarial Life Table", with separate figures for men and women, covering a wider population than the CDC's: alongside residents of the 50 states and the District of Columbia, it counts groups such as residents of the U.S. territories and citizens living abroad. Each edition is identified by the Trustees Report that used it: the 2023 period table, for example, is the one used in the 2026 report. Because the population and the data year differ from the CDC's, its numbers differ too, and the two should not be mixed in one calculation.

Other tables and the one people confuse with them. Insurers price life insurance and annuities on mortality tables built from insured lives, and pension plans compute lump sums with mortality tables prescribed for that purpose, so a pension buyout figure rests on a table as much as on an interest rate. The IRS tables used for required minimum distributions, such as the Uniform Lifetime Table and the Single Life Table, are life-expectancy divisor tables set by regulation for distribution rules; they are not general-purpose life tables and are not an estimate of how long a particular person will live.

What a table cannot tell you. A national table describes a population, not a person. Sex, health, income, smoking and family history all move individual odds, and a married couple's chance that at least one spouse lives to a very old age is higher than either one's alone. That is the case for testing a retirement plan against living well past the table's average rather than building it to the average.

Reading a life table for a planning question comes down to two calculations.

1. Remaining life expectancy is read directly from the expectation-of-life column at the person's current age.
2. The chance of reaching a later age is the number surviving at the later age divided by the number surviving at the current age.

For example, using the CDC's 2024 United States life table for the total population: of 100,000 people born, 84,133 are alive at 65, 43,776 at 85 and 25,811 at 90. So for someone who has reached 65, the chance of reaching 85 is 43,776 / 84,133, or about 52%, and the chance of reaching 90 is 25,811 / 84,133, or about 31%. Their remaining life expectancy at 65 is 19.7 years, to about 84.7. A retirement plan built to end at 85 would therefore run out of money for about half of the people it was designed for, under 2024 death rates and before any adjustment for the person's own health or for future improvement in mortality.

Pros

• Turns abstract death rates into usable numbers: survival odds and expected years remaining at any age.
• Published free by government statisticians, by age and sex, and updated regularly.
• Makes clear that remaining life expectancy rises with age, which corrects a common planning error.

Cons

• A period table is a snapshot of one year's death rates, not a forecast, and it will be off if mortality keeps changing.
• Population averages say little about one person's health, habits or income.
• Different publishers' tables give different numbers, so mixing them produces inconsistent results.
• An average invites planning to the average, when much of the population lives longer.