A country can register more births than it did the year before while its fertility rate falls. Another can report a rising fertility rate and still have fewer babies. The apparent contradiction disappears once we ask what each measure counts.
The annual birth count is the number of live births registered or estimated during a year. It is a headcount, not a rate. It changes with the childbearing rates at each age and with how many women are living at those ages.
The crude birth rate is the number of live births in a year for every 1,000 people in the population. It is useful for comparing births with deaths and following population change, but it is strongly affected by age structure. A retirement community and a college town can have different birth rates even if women of the same ages have children at identical rates.
The total fertility rate, or TFR, removes much of that age-structure problem. Demographers calculate birth rates for women at each age and combine them into the number of children a hypothetical woman would have if one year’s rates continued throughout her reproductive life. A TFR of 1.5 does not mean the average woman alive has 1.5 children, and it does not count the babies born that year. It summarizes a period schedule.
A period rate, not a completed family South Korea’s total fertility rate from 1925 to 2020. The final value describes that year’s age-specific rates, not the family size of every woman then alive. via Wikimedia Commons
_in_Korea.svg){kind=link}
Completed cohort fertility looks backward instead. It accumulates the births experienced by a generation of women as they pass through their reproductive years. It tells us more about eventual lifetime family size but arrives too late to provide an early reading of change.
The distance between the period and cohort measures is partly about timing. When many women postpone births from their twenties into their thirties, the current TFR falls because fewer births occur at the earlier ages. It can later rise when some postponed births take place. Demographers call this a tempo effect. Delay can depress a period measure without reducing completed fertility by the same amount, though births postponed long enough may never occur.
This is why an annual rebound deserves attention without automatically becoming a reversal. South Korea’s TFR rose from 0.72 in 2023 to 0.75 in 2024, and births increased. The movement may reflect timing, policy, marriage patterns, cohort size, or a more durable change. One year cannot tell us which.
Replacement, births, and population
The most direct measure of generational replacement is the net reproduction rate, or NRR. It estimates the average number of daughters per woman in a hypothetical cohort under the fertility and mortality rates of a period, including survival through the reproductive years. An NRR of 1 means that one generation of women is replacing itself with one daughter generation under those conditions.
Replacement fertility translates the same underlying relationship into a TFR benchmark. In a low-mortality population with the usual ratio of male to female births, the TFR associated with replacement is often about 2.1 births per woman. It is not exactly two because slightly more boys than girls are born and some girls do not survive through the reproductive years. The benchmark is higher where early mortality is higher. NRR, rather than a raw TFR, is the direct daughter-replacement measure.
Replacement is a population relationship, not an instruction to a household. A woman does not reproduce at 2.1, and the moral worth of a family does not turn on its position relative to a national average.
Nor does crossing 2.1 make population change direction immediately. The number of births depends on both fertility rates and the number of women at each childbearing age. A small cohort can have a slightly higher TFR and still produce fewer babies. A large cohort can have a lower TFR and produce more. Deaths depend on the number of people at older ages, while migration adds or removes people at many ages.
These distinctions give us a way to read the map ahead:
- Annual births count the new lives recorded in a year; the count depends on both rates and the number and ages of potential parents.
- Period TFR summarizes one period’s age-specific childbearing rates as a hypothetical lifetime.
- Completed cohort fertility records the childbearing accumulated by a generation of women.
- NRR estimates daughter-generation replacement under a period’s fertility and mortality schedules.
- Replacement TFR approximates the fertility level associated with an NRR of 1 under stated mortality and sex-ratio conditions.
- Age structure, mortality, and migration determine how those measures appear in the population total.
The measures do not compete. They describe different parts of the same movement.
Citations
- Eurostat, “Fertility statistics”, updated 2026.
- United Nations, Department of Economic and Social Affairs, Population Division, World Population Prospects 2024: Methodology Report, 2024.
- Statistics Korea, Birth Statistics in 2024, 2025.