How world population will change across the 21st century

A population projection is less a forecast than a structured consequence. Given a starting age structure and a set of assumptions, the arithmetic is largely fixed. Recent interactive tools make this visible: the user sets a small number of inputs and the model returns a full trajectory, with the underlying mechanics small enough to run inside a web browser.

The three inputs that move everything

Almost every projection rests on three rates. The total fertility rate sets how many children each cohort of women bears. Life expectancy at birth, together with age-specific mortality, sets how long people live. The net migration rate adds or subtracts people at each age. Holding the starting age structure fixed, these three quantities determine the path. A tool that exposes them invites a useful experiment: change one rate, hold the others, and observe which lever the future is most sensitive to.

• Fertility dominates the long run, because it compounds across generations.
• Life expectancy reshapes the top of the age structure and the old-age dependency ratio.
• Migration matters most for individual countries and least for the world total, since it redistributes rather than creates people.

The cohort-component method

The standard engine is the cohort-component method. The population is divided into age and sex groups; each year, survivors age into the next group at rates implied by mortality, births are added from the fertility schedule applied to women of childbearing age, and net migrants are layered on. Iterated forward, this produces the familiar population pyramid evolving over time. The method is transparent, which is its strength: every output can be traced back to an assumed rate rather than a hidden parameter.

A century of slowing growth and a peak

The central message across major projections is consistent. Global population continues to rise, but the pace has already slowed sharply from its twentieth-century high. United Nations figures place the peak before the end of the century, at just over 10.4 billion around the mid-2080s, after which the world total begins a gradual decline. This is a structural turn rather than a cliff: the deceleration reflects fertility falling toward and below replacement level across most regions.

Population momentum explains why the peak lags the fertility decline. A young age structure keeps a population growing for decades even after fertility reaches replacement, because large cohorts are still entering their reproductive years. The same momentum works in reverse where populations have already aged, locking in decline that no plausible near-term rise in births would offset.

Where the growth, and the decline, will be

The global total hides divergent national paths. Growth in high-income countries has effectively stopped, sustained in some cases only by migration. India is projected to reach a peak above 1.6 billion and then decline; China is set to contract substantially over the century; the United States grows slowly. Net additions to world population are concentrated increasingly in sub-Saharan Africa, where younger age structures and higher fertility persist. This redistribution, more than the headline total, will shape questions of labour supply, ageing and settlement that other briefs in this resource examine.

A single world figure can move very little while the map beneath it is redrawn entirely.

Two cautions follow for anyone using a projection tool. First, the further out the horizon, the wider the fan of plausible outcomes, because small differences in assumed fertility compound. Second, a tool's clarity can be mistaken for certainty: the smooth output line is the consequence of chosen inputs, not a measurement of the future. Read alongside our briefs on below-replacement fertility and on migration and settlement, the projections are best treated as a disciplined way to reason about demographic change, not a prediction of it.