The Fertility Divide: Korea Rebounds as China Collapses

Overview

East Asia is routinely described as a single demographic crisis, and in 2025 the region's two most-cited cases moved in opposite directions.

China recorded 7.92 million births, down 17% from 9.54 million the previous year and the fewest since 1939. With 11.31 million deaths, the population fell by 3.39 million — a fourth consecutive annual contraction, and the steepest so far.

South Korea, long treated as the world's demographic worst case, went the other way. Its total fertility rate rose from 0.75 to around 0.8, births increased for fifteen consecutive months through late 2025, and the first quarter of 2026 produced 75,013 births — up 14.8%, the largest increase since records began in 1981.

Korea remains far below replacement and the rebound rests on fragile foundations. But a country widely presented as beyond recovery has posted its strongest birth figures in over four decades, which is worth stating precisely rather than folding into a general narrative of collapse.

China's Births Fell to a 1939 Low

China recorded 7.92 million births in 2025, a 17% fall from 9.54 million in 2024. It is the lowest annual figure of the modern era, and the lowest since 7.57 million births in 1939.

The 1939 comparison is worth sitting with. That was a wartime year, during the Second Sino-Japanese War, when China's population was roughly a third of today's. A country of over 1.4 billion people recorded fewer births than one of around 500 million did during a war.

The birth rate fell to 5.63 per 1,000 people, below the previous record low of 6.39 set in 2023. A 17% single-year decline is exceptionally steep for a demographic indicator, which normally moves gradually.

The Decline Is Accelerating

China's population has now contracted for four consecutive years, and the notable feature is that each contraction has not simply continued but deepened.

11.31 Million Deaths

The population arithmetic has two sides, and the death figure is doing as much work as the birth figure.

China recorded 11.31 million deaths in 2025 against 7.92 million births. Even if births had held at their 2024 level of 9.54 million, the population would still have fallen by nearly two million.

This is the part of demographic decline that policy cannot easily reach. Deaths are rising because the enormous cohorts born during China's mid-twentieth-century population boom are now reaching the end of life. That trajectory is fixed decades in advance, so the gap between births and deaths will widen for years regardless of what happens to fertility.

South Korea Is Going the Other Way

South Korea has held the lowest fertility rate of any major country for years, bottoming at 0.72 in 2023 — a figure so far below replacement that it implies each generation being roughly a third the size of the last.

It has since risen, reaching 0.75 in 2024 and approximately 0.8 in 2025. In absolute terms that remains extraordinarily low. As a direction of travel it is the first sustained improvement in over a decade.

Births rose for fifteen consecutive months through September 2025, with 190,000 babies born from January to September — a 6.9% year-on-year increase, the largest since 2007.

The Best Quarter Since Records Began

The first quarter of 2026 produced South Korea's strongest birth figures in the history of its statistical record. There were 75,013 births, an increase of 14.8% on the same period a year earlier.

Both the rate and the scale of that increase are the largest since national statistics began in 1981. For a country whose demographic trajectory has been treated internationally as a cautionary tale, this is a genuinely significant reversal.

It is also why treating East Asia as a single bloc obscures more than it reveals. In the same period that China's births fell 17%, South Korea's rose 14.8% — two countries with similar cultural pressures around marriage, housing costs and working hours, moving in opposite directions.

Why the Korean Rebound May Not Last

Analysts examining the recovery have been careful about what is driving it, and the mechanisms are partly structural rather than behavioural.

One factor is composition. A larger cohort of women is currently passing through peak childbearing age, so more births can occur even without any change in how many children the average woman has. Another is timing: marriages and births delayed during earlier years are now taking place, producing a temporary bulge as a backlog clears.

The constraint underneath is severe. The absolute number of women aged 15 to 49 is declining, and continues to decline. Once the delayed-marriage effect works through, a rising fertility rate applied to a shrinking population of potential mothers can still produce falling births. The assessment is that the figures are stabilising rather than rebounding in a durable sense.

What Replacement Level Actually Means

The benchmark that gives these numbers meaning is 2.1 births per woman — the rate required for a population to replace itself without migration.

It is slightly above two because not every child survives to reproductive age and slightly more boys are born than girls. In countries with higher child mortality the replacement threshold is higher still.

Against that benchmark, South Korea at 0.8 is running at roughly 38% of replacement. Absent migration, each generation would be around a third the size of the one before, which compounds brutally: three generations at that rate reduces a cohort to about 5% of its original size. That is why even a substantial percentage rebound leaves the underlying situation extreme.

Why East Asia Isn't One Story

The divergence matters because it bears on whether low fertility is reversible, which is the central open question in demographic policy.

If fertility decline were a one-way ratchet driven by irreversible cultural change, no country at Korea's level would post a 14.8% quarterly increase. That it happened suggests these rates respond to something — economic conditions, housing costs, policy support, or simply the timing of delayed decisions.

China's simultaneous 17% fall points the other way, and the two countries differ in ways that may explain it: the demographic shadow of the one-child policy, the size and age structure of the cohort now reaching parenthood, and a property market whose troubles bear directly on family formation. The honest conclusion is that fertility appears responsive rather than fixed, and that nobody has yet demonstrated a reliable way to move it.

What a Shrinking Cohort Locks In

Demographic momentum is the reason these figures matter well beyond the year they describe. Every person who will be forty in 2066 has already been born.

China's 7.92 million births in 2025 therefore fix the size of its 2045 workforce entrants and its 2085 retirees. No policy introduced later can change that number, only how many people migrate in or out of it.

This is what separates demographic decline from most economic problems. A recession can be reversed within years; a small birth cohort propagates through every subsequent stage of a society's life for a century. Even if fertility returned to replacement tomorrow, the smaller cohorts already born would continue moving through the population structure — which is why automation and immigration dominate the policy discussion. They are the only levers that operate on a useful timescale.

Conclusion

China's demographic decline accelerated sharply in 2025. Births fell 17% to 7.92 million — the fewest since 1939 — while 11.31 million deaths pushed the population down by 3.39 million, the steepest of four consecutive annual contractions.

South Korea moved the opposite way. Its fertility rate rose from 0.75 to around 0.8, births increased for fifteen straight months, and the first quarter of 2026 delivered a 14.8% rise — the largest since records began in 1981. The country most often cited as beyond recovery is the one currently recovering.

Both facts deserve their qualifiers. Korea's rebound is driven substantially by cohort composition and delayed births working through, against a shrinking population of women of childbearing age, and 0.8 remains roughly 38% of replacement level.

What is already fixed is the arithmetic. The children born in 2025 determine workforce sizes into the 2040s and pension burdens into the 2080s, and no later policy can revise that. Which is why the response has centred on automation and migration — the only interventions that work on a timescale shorter than a generation.

Data Source and Attribution

Caixin GlobalCNN (Korea)SCMP

Chinese birth, death and population figures come from China's National Bureau of Statistics as reported in January 2026, covering the 2025 calendar year, with historical comparisons to 1939 and the annual contraction series from 2022 onward. South Korean fertility rates, monthly birth trends and first-quarter 2026 figures come from Statistics Korea and contemporaneous reporting. Replacement level fertility of 2.1 births per woman is the standard demographic benchmark for populations with low child mortality.

FactsFigs reviews, cleans, and cross-checks every source dataset before shaping it into a data story. Each visualization is created and designed in FactsFigs Design Studio — an internal tool developed and owned by FactsFigs — and is the original work of a FactsFigs author, not an AI-generated copy of any existing graphic. Individual assets within a visual may or may not be produced with AI tools, but the design of the visual itself is solely FactsFigs' own.

Figures are estimates at the time of publication, provided for information only — nothing here is financial advice or a guarantee of accuracy.

2026-07-20