The world is having fewer children. What changes next?

01

A quieter demographic revolution

The world is having fewer children per woman. In this World Bank release, global total fertility fell from 4.68612931000644 births per woman in 1960 to 2.18831256131236 in 2024. Rounded for the opening comparison, that is 4.69 to 2.19. The change reaches well beyond Europe or East Asia: the histories of Brazil, India, Iran and Nigeria all show substantial declines, although their current levels differ considerably.

Yet this is not a single global timetable. Some societies are discussing how to maintain schools and pensions with smaller incoming generations. Others still face rapidly expanding demand for classrooms, maternal care and first jobs. An average describes the direction of change; it does not describe the local queue outside a school or clinic.

This story follows two measures through 65 years, across 217 countries and territories. It uses a fixed, validated World Development Indicators snapshot dated 13 July 2026, acquired on 5 October 2026. The latest common data year is 2024. These series combine national statistics with international estimates; they are not a real-time birth register.

World Bank · SP.DYN.TFRT.IN · SP.DYN.CBRT.IN · 2026-07-13

02

Two measures, two denominators

Total fertility rate asks a hypothetical lifetime question: how many children would a woman have if she experienced the current age-specific fertility rates throughout her reproductive years? Its unit is births per woman. It is a period measure, not a count of the children that today’s women have already had, and not a forecast of their eventual family size.

The crude birth rate asks an annual population question: how many live births occur per 1,000 people in the midyear population? It includes everyone in the denominator. The age composition of that population therefore matters. Neither measure supplies the absolute number of births without additional information.

In 2024 Czechia and Germany both report fertility of 1.36 births per woman in this source. Their crude birth rates differ: 7.7 and 8.1 live births per 1,000 people respectively. The equality of one rate does not require equality of the other. This comparison illustrates the different denominators; these two indicators alone cannot identify how much of the difference comes from age structure.

World Bank · SP.DYN.TFRT.IN · SP.DYN.CBRT.IN · 2026-07-13

03

The long decline was not a straight line

The world series rises between 1960 and 1970: fertility is 4.8399741901259 in 1970, compared with 4.68612931000644 in 1960. By 1980 it is 3.73123413364959, by 2000 2.72979589581575, and by 2024 2.18831256131236. Calling the long-term pattern a decline is justified; calling every year a decline would erase the actual history.

The world crude birth rate moves from 31.9992347613353 live births per 1,000 people in 1960 to 16.2730070969907 in 2024. Read the two global charts together, but do not add their values or interpret their difference as a number of missing children. Their units and denominators are different.

Historical estimates can be revised when new census, survey or registration evidence arrives. The snapshot makes this article reproducible, but it does not imply that an older estimate is measured as precisely as its long decimal representation suggests. Exact source strings document what the provider published; decimal length is not a confidence interval.

World Bank · SP.DYN.TFRT.IN · SP.DYN.CBRT.IN · 2026-07-13

04

Most countries are below a reference. The world is above it.

In 2024, 119 of the 217 countries and territories with a fertility value are below 2.1 births per woman. That is 54.84%, calculated as 119 ÷ 217 × 100 and rounded to two decimals. In 1960 the comparable count is 5 of 216 available countries and territories. The missing historical geography is excluded rather than counted as zero.

The apparent contradiction disappears once the denominator is visible. A country count gives a small territory the same weight as India. The World Bank’s world aggregate is a separate published series, not our simple average of national rates. Neither the 119-country count nor 54.84% tells us the share of the world’s people living below the benchmark.

Approximately 2.1 is a familiar replacement reference for populations with low mortality. Replacement depends on survival and the sex ratio at birth; it is not universally exactly 2.1. Even a sustained rate near replacement does not instantly stabilise a population. Age structure, deaths and migration still affect the outcome.

World Bank · SP.DYN.TFRT.IN · SP.DYN.CBRT.IN · 2026-07-13

05

One transition, several destinations

South Korea moves from 5.994 births per woman in 1960 to 0.748 in 2024. Brazil moves from 6.051 to 1.614; Iran from 7.519 to 1.684; India from 5.921 to 1.962. These endpoint comparisons establish the size of the change, not its cause. The full histories reveal when the pace accelerated or slowed.

Nigeria falls from 6.364 to 4.382 births per woman over the same interval. A decline can therefore coexist with a level still well above the replacement reference. Israel reports 2.87 in 2024, compared with 3.866 in 1960. The common downward direction does not make national experiences interchangeable.

Choose a country in the explorer and inspect its whole series before attaching an explanation. A short rebound can disappear inside a long decline; a low recent period rate can partly reflect births being postponed. These data do not include maternal ages or completed cohort fertility, so they cannot separate timing changes from changes in eventual family size.

World Bank · SP.DYN.TFRT.IN · SP.DYN.CBRT.IN · 2026-07-13

06

Very low fertility and high fertility coexist

Six countries or territories are below 1 birth per woman in 2024: Macao (0.582), South Korea (0.748), Hong Kong (0.841), Puerto Rico (0.92), Singapore (0.97) and Ukraine (0.988). These are World Bank geography definitions. Our coverage includes territories; this is not a ranking restricted to sovereign states.

At the other end, Chad reports 6.028, Somalia 6.013, the Democratic Republic of the Congo 5.981 and Niger 5.935 births per woman. Across the full 2024 coverage, 25 countries or territories are at 4 or above, compared with 160 of the 216 available in 1960. The geography of high fertility has narrowed while important differences remain.

A ranking describes levels, not welfare. A low rate is not proof that people are happier or poorer; a high rate is not proof that children are wanted or unwanted. Conflict, displacement and weak registration can also complicate estimates. The dataset cannot diagnose a household’s motives or establish a policy success from a national position in a league table.

World Bank · SP.DYN.TFRT.IN · SP.DYN.CBRT.IN · 2026-07-13

07

Czechia in its neighbourhood

Czech fertility is 1.74 births per woman in 2020 and 1.36 in 2024. The change is −0.38 births per woman, calculated as 1.36 − 1.74. Its crude birth rate changes from 10.3 to 7.7 live births per 1,000 people over the same years. These are changes in rates; they do not state the number of children in a classroom cohort.

In the same 2024 release, Poland is at 1.14, Germany at 1.36 and France at 1.61 births per woman. The United States is at 1.6265. Comparing the same year and source avoids silently combining different vintages. It does not eliminate differences in how underlying national information is measured or estimated.

For Czech municipal budgets the useful next question is local: which birth cohorts live here, where are families moving, and when will they enter nursery and primary school? A national fertility average cannot justify closing a specific school. Municipal planning needs residence-based cohort counts, migration, catchment areas and realistic capacity constraints.

World Bank · SP.DYN.TFRT.IN · SP.DYN.CBRT.IN · 2026-07-13

08

Fewer children per woman does not mean immediate population decline

A large generation reaching reproductive ages can produce many births even when each woman has fewer children than earlier generations. That is the intuition behind demographic momentum. Conversely, an older population can have relatively few births even if its fertility rate temporarily improves. Fertility describes reproductive intensity; the population’s age structure determines how many people are exposed to those rates.

Population change also includes deaths and net migration. India’s 2024 fertility of 1.962 and crude birth rate of 15.981 do not, by themselves, tell us whether its population is shrinking. South Korea’s corresponding 0.748 and 4.7 do not supply a complete population balance either. We need deaths, migration and population stocks before making that claim.

Migration can change the size and composition of the population and the location of demand for services. It should therefore enter a demographic budget scenario explicitly. It cannot be inferred from a fertility chart or treated as a fixed substitute for every other adjustment.

World Bank · SP.DYN.TFRT.IN · SP.DYN.CBRT.IN · 2026-07-13

UN WPP 2024 · methodology ↗

09

What could explain the change?

Research points to several interacting conditions: the cost and availability of housing, income security, access to childcare, and the ability to combine work with family life. OECD’s 2024 review finds no single policy that explains the whole pattern. These are contextual explanations from research, not effects estimated by our two-indicator dataset.

The economics literature also emphasises how institutions and sharing care can make careers and parenthood more compatible. UNFPA’s 2025 report frames reproductive agency as the ability to realise one’s own choices, including whether, when and how many children to have. A demographic target and a person’s preferred family life are different things.

To test a particular explanation we would need additional variables, a credible comparison and a method that separates policy effects from other changes. A subsidy introduced before a rebound is not sufficient proof of causation. Neither is a cross-country correlation. This story establishes the demographic pattern and makes the explanatory questions visible.

OECD 2024 ↗ · Doepke et al. ↗ · UNFPA 2025 ↗

10

The budget consequences arrive on different clocks

Births first affect maternity care and early-years services. Later, the same cohorts reach schools, training and the labour market. Eventually they influence the balance between workers and retirees. The timing matters: a fertility change today is not an immediate addition to tomorrow’s workforce or an immediate saving in today’s pension bill.

Smaller cohorts can reduce some demand while making fixed costs harder to spread. A school building, bus route or hospital department cannot always shrink in exact proportion to the population. Meanwhile family moves can concentrate children in a growing district even when national births fall. Capacity, geography and minimum service standards belong in the calculation.

Pension and tax scenarios require employment, productivity, wages, contribution rules, retirement ages, health and migration assumptions. These rates alone cannot price the future pension gap or establish that a system will collapse. A useful public-finance scenario publishes those assumptions and tests alternatives instead of presenting a fertility headline as a fiscal forecast.

The same discipline applies where fertility remains high. Expanding educational and health access can be urgent, but the budget needed depends on the number and location of children, existing provision and unit costs. A rate is a signal to investigate, not a spending estimate.

11

Explore the evidence yourself

Change the year, measure and region. Search countries and territories, then select up to four histories. The ranking shows the highest available values within the filter; the table includes every matching geography, including explicit missing values. The scatter compares the two rates in the same year without assigning population-sized bubbles.

Chart tables and CSV exports preserve source decimals. Chart labels are rounded for readability. A blank historical value remains missing; it is never converted to zero. Country names and regional classifications follow the provider’s current catalog, so this explorer does not reconstruct every historical border.

12

The evidence, its precision and its limits

Coverage: 217 countries and territories, two indicators, 1960–2024. The global charts use the World Bank’s WLD aggregate. Regional aggregates are stored separately and never added to country rows. There are 48 source aggregate geographies; 47 contain numeric values. Some early country-years are missing. The source also returned 2025 cells, all without values; they are excluded from the explorer rather than presented as a new data year.

The acquisition accepted 34,980 source cells across the two indicators, including 34,258 numeric values and 722 nulls, with no rejected or duplicate rows. These totals cover the complete received grid, including aggregate rows and the empty 2025 cells, not only the countries and years displayed here. Validation preserves source strings, units, geography identifiers and provenance before publication.

Reproducibility: warehouse release 0b7239dd-8ca1-4ea8-93e6-a5c7632ed167; serving snapshot 7c4803a6-426e-4fb5-9e47-74b61117d29d. The snapshot fixes the source vintage rather than querying changing live values. Threshold counts are calculated from available country/territory fertility cells using strict inequalities. Endpoint changes subtract the earlier value from the later value. There are no projections or imputed missing cells in this story.

Sources to inspect

• World Bank · SP.DYN.TFRT.IN · exact fertility source endpoint ↗
• World Bank · SP.DYN.CBRT.IN · exact birth-rate source endpoint ↗
• World Bank · fertility definition and estimation limits ↗
• World Bank · crude birth rate definition ↗
• OECD · Society at a Glance 2024 · fertility and policy ↗
• Doepke et al. · The Economics of Fertility: A New Era ↗
• UNFPA · State of World Population 2025 ↗
• UN · World Population Prospects 2024 · methodology ↗

All demographic figures above refer to the dated World Bank snapshot. Contextual research is cited separately and does not change those figures. Read the sources alongside the exact table and chart downloads.