A baby born in Japan and a baby born in Lesotho do not get the same number. That gap, more than 20 years, is the first thing to understand: life expectancy describes a place and a moment, not a person.
The figure answers a narrow question. If a large group of people experienced today's death rates at every age, how long would they live on average? That mean is the headline. Within any group, some live far longer and some far shorter. The single number sits somewhere in the middle.
Someone who reaches 65 has a higher remaining life expectancy than the at-birth figure minus 65. They have already survived childhood, accidents, and early diseases. The average is recalculated for their smaller, hardier group.
The number describes a population. It is closer to the mean temperature of a country than a forecast for your street.
Period Life Expectancy: The Headline Method
This is the method behind almost every news report and government statistic.
Period life expectancy takes the age-specific death rates from a single year and applies them to a hypothetical group of newborns. It asks: if these death rates stayed exactly the same for the next century, how long would the typical newborn live?
The calculation is fast, simple, and comparable across countries. It does not assume death rates will actually stay frozen. It simply holds today's conditions still to produce a clean number.
The weakness is obvious. Death rates change. Medical advances, pandemics, wars, and lifestyle shifts all alter the real survival of a generation. Period life expectancy is a snapshot, not a film.
Cohort Life Expectancy: Following Real Generations
Cohort life expectancy follows a real birth group through time. It uses the actual death rates the generation experienced at each age, then projects future rates for the years still ahead.
A person born in 1950 has already lived through decades of real death rates. To calculate their remaining life expectancy, actuaries use the actual mortality they have already survived, plus projected rates for what remains. This is more accurate for individuals. It requires forecasting.
The catch: you cannot calculate a full cohort life expectancy until the generation has died. It is a retrospective measure. For a newborn today, a cohort estimate depends entirely on assumptions about medical progress, climate change, and future pandemics.
Period figures dominate headlines because they are available now. Cohort figures are more useful for insurance and pension planning, but they carry more uncertainty.
Life Tables Explained: lx, dx, qx, ex
The life table is the engine behind both methods. It contains four columns.
| Symbol | Meaning |
|---|---|
| lx | Number surviving to exact age x |
| dx | Deaths between age x and x+1 |
| qx | Probability of dying before next birthday |
| ex | Remaining years of life expected at age x |
Statisticians start with a hypothetical group of newborns. At each age, they apply the qx value to remove the expected number of deaths. The survivors move to the next age. By adding up all the years lived and dividing by the starting group, they get the mean.
The table is rebuilt every year using the latest data. If death rates at age 70 improve, the entire column for ages 70 and above shifts.
Why Life Expectancy at Birth Is Lower Than at Age 65
This is the most common point of confusion.
Life expectancy at birth in a high-income country is roughly 80 years. Someone who reaches 65 has a remaining life expectancy that pushes their expected total to about 84.
That is four years more than the at-birth figure minus 65 would suggest. Why?
Infant mortality and childhood accidents pull the at-birth mean down. A baby who dies at age one contributes only one year to the calculation. Someone who reaches 65 has already survived all those early risks. The mean for their group is recalculated from a pool that has already been filtered by survival.
The same logic applies at every age. A 90-year-old has a higher remaining life expectancy than the at-birth figure minus 90, because only the longest-lived people are still in the group.
Which Organisations Calculate Life Expectancy, and How
Several organisations publish life expectancy, and they do not always agree.
The World Health Organization publishes World Health Statistics every year, with life expectancy at birth by country. The United Nations Population Division produces World Population Prospects every two years, offering both period and cohort estimates for every country.
National agencies add their own figures. The UK Office for National Statistics, the US National Center for Health Statistics, and Eurostat all produce life tables using their own data and methods.
Actuaries produce separate tables for insurance and pension use. These diverge from population means because they cover insured or annuitant populations, which tend to be healthier and wealthier than the general population.
How Life Expectancy Is Calculated: Why Different Sources Disagree
There is no single official global life expectancy.
WHO, UN, and national agencies use distinct data sources, distinct methods for handling missing data, and distinct assumptions about future trends. Differences of one to two years between sources are normal and expected.
The UN and WHO both use period methods for their headline figures, but they may use separate age groupings or separate adjustments for countries with incomplete death registration. A country that registers only some deaths requires statistical estimation, and different organisations estimate differently.
If you see a figure of 81 years from one source and 82 from another, neither is wrong. They are answering a slightly different question with slightly different data.
Healthy Life Expectancy
Life expectancy counts all years, including those lived with illness or disability. Healthy life expectancy subtracts the years lived in poor health.
The World Health Organization publishes healthy life expectancy alongside regular life expectancy. In most countries, the gap is 8 to 10 years. People live to 80, but they may spend the last decade with significant health problems.
This measure is useful for planning retirement and understanding quality of life, not just quantity. It answers: how many years of good health can you expect?