Europe's Demographic Turning Point

The most consequential fact about European demography is not a projection. It is a completed observation, and it dates from 2012.

In that year, across the European Union as a whole, deaths exceeded births - and they have done so in every year since. The EU's population has continued to grow in most of that period, but the growth has come from net migration. Natural change, the balance of births against deaths, has been negative for over a decade.

That is the turning point. Everything else in this article is either the mechanism behind it or the uncertainty in front of it.

What is measured, and how well

Status: OFFICIAL. The figures in this section come from Eurostat and national statistical institutes and describe events that have already happened.

The total fertility rate across the EU has been below the replacement level of approximately 2.1 births per woman for decades - not in some member states, in all of them. Eurostat recorded an EU total fertility rate of 1.38 in 2023, the lowest in its comparable series, continuing a decline from the modest recovery of the mid-2000s.

The EU's median age now exceeds 44 years and has risen consistently. The share of the population aged 65 and over has grown in every recent year, and the old-age dependency ratio - retirement-age people relative to working-age people - has risen accordingly.

These are not contested numbers. European statistical systems are among the most complete in the world, combining full population registers in many countries with censuses and continuous vital registration. When this article says deaths exceeded births in 2012, that is a count, not a model.

The three components, separated

Demographic argument goes wrong most often by merging three distinct things. They should be kept apart.

Fertility is the number of children born per woman. Europe's is low and has been for a long time. The decline is associated with later first births, extended education, housing costs, the economics of childcare and changed family formation patterns. Research consistently finds a gap between the number of children Europeans say they want and the number they have, which suggests the constraint is at least partly circumstantial rather than purely preferential - though the size of that gap and what closes it are actively debated.

Mortality and ageing. Life expectancy in Europe rose substantially over the twentieth century and, after the disruption of the pandemic years, resumed broadly upward. Longer life is an achievement. Its demographic consequence is that a large cohort born in the post-war decades is now moving into old age simultaneously.

Migration. Net migration has been positive for the EU as a whole and is currently the sole source of population growth. It changes the total and the age structure; it does not by itself change the fertility trajectory, since migrant fertility tends to converge toward the destination country's rate over a generation.

ADVERTISEMENT

Merging these three produces most of the bad arguments in this field. Ageing is driven principally by past fertility decline and rising longevity, not by migration. Migration levels do not explain why fertility fell. Keeping the components separate is the whole methodological discipline.

Why the trend is largely locked in

There is a further property of demographic change that deserves emphasis, because it explains why the next thirty years are more predictable than the next thirty years of almost anything else.

Population structure has momentum. The number of women who will be of childbearing age in 2045 is already fixed: they have been born. A country whose birth cohorts shrank in the 2000s will have a smaller cohort of potential parents in the 2030s, and even a sharp recovery in fertility per woman would produce fewer births in absolute terms.

This works in both directions and it is why demographers are comparatively confident about the medium term and comparatively humble about the long one. A policy that raises fertility today changes the labour force in twenty years and the pensioner ratio in sixty. Nothing available to a government changes the age structure of the next two decades, because that structure is already alive.

Variation inside the aggregate

The EU figure conceals a wide spread, and the spread matters more than the average for anyone reading a specific country.

Fertility varies meaningfully between member states, and the drivers differ. Several central and eastern member states have experienced population decline driven not only by low fertility but by sustained emigration of working-age people, which removes future parents as well as current workers. Some western and northern states have offset natural decline through immigration and have younger age structures as a result. A few countries have seen fertility fall furthest from the highest recent levels.

Treating the continent as a single demographic unit is therefore useful for framing and misleading for policy. The pension problem in a country losing young workers to emigration is not the same problem as in one receiving them.

What is projected, and what that word means

Status: PROJECTED. The following are not predictions. They are the arithmetic consequences of stated assumptions.

Eurostat's population projections and the United Nations World Population Prospects 2024 both indicate that, on current assumptions, the EU's population would decline over the coming decades and age substantially further, with the working-age share falling relative to the retired share.

The critical qualification: a projection is a conditional statement. It says that if fertility, mortality and migration follow specified paths, then the population follows a particular curve. The UN publishes variants - low, medium and high fertility among them - precisely because the assumptions are uncertain, and the spread between variants widens sharply with distance. A medium-variant figure for 2050 is reasonably robust. A medium-variant figure for 2100 carries a range so wide that quoting the central value alone is misleading.

UN WPP 2024 projects to 2100. It does not extend beyond that. Any figure for a later date - including any 2126 figure appearing in this publication - is a NATION CONTINUITY model with its own stated assumptions, and it is never attributable to the United Nations. This magazine labels those figures accordingly, and readers should treat any source that does not as unreliable.

What the numbers do not settle

This is the part usually omitted, and it is where the demography stops and the politics begins.

Low fertility and ageing are established. What they mean is not, because the same demographic inputs are compatible with radically different outcomes depending on decisions not yet made: pension design, retirement age, labour force participation among women and older workers, productivity growth, care system funding, migration policy, family policy.

Nor do the numbers settle anything about cultural continuity, which is this magazine's subject. A smaller population can transmit a language, a craft and a body of civic memory perfectly well if the institutions to do so are funded. A larger one can lose them if they are not. The demographic constraint is real; it is a constraint on resources and on the number of people available to transmit and receive, not a determinant of the outcome.

The responsible statement is narrower than the alarmed one and harder to dismiss: Europe's population structure is changing in a documented direction, the change is largely locked in for the next several decades by births that have already happened or not happened, and the policy questions it forces are genuine. That is enough. It does not require amplification, and it does not license projection quoted as prophecy.