Births down over 40%: Japan should start with the marriage entry point
- Main cause
- The AI sees two overlapping factors: fewer marriages (down about 38%) shrank the entry point, and the births-per-marriage indicator fell.E-01E-02E-03E-04
The AI's own summary of its diagnosis and policy. Figures are checked against the evidence.
Live births (Japan) 1995 → 2025 −43%
Japan's births fell from about 1.19 million in 1995 to about 0.67 million in 2025, a decline of about 44%.
Marriages fell from about 0.79 million in 1995 to about 0.49 million in 2025, a decline of about 38%.
The total fertility rate fell from 1.42 in 1995 into 2005, recovered to 1.45 in 2015, and stood at 1.14 in 2025.
A rough indicator comparing births with marriages in the same year fell from 1.50 in 1995 to 1.37 in 2025 (it is not the number of children per couple).
Between 1995 and 2024, the share of people aged 65 and over rose from about 15% to about 30%. The female labour-force participation rate also rose, reaching about 56% in 2024.
Cabinet Office general-account spending (settlement) rose from about 2.7 trillion yen in FY2015 to about 4.8 trillion yen in FY2024. However, the scope of the Cabinet Office's responsibilities changed over this period, including the creation of the Children and Families Agency in 2023, so this rise cannot be read directly as growth in spending on children's policy.
The roughly 44% fall in births appears to combine two things: a shrinking 'entry point' (marriages down about 38%) and a lower births-to-marriages indicator. Framing the problem only as support for married couples is likely to miss the shrinking entry point.
The rise in women's labour-force participation alongside a falling fertility rate suggests that combining work and children may still be hard in Japan. These changes happened at the same time, however, and this data alone cannot show which, if either, drives the other.
Policy options fall into three directions. (1) Stable incomes and jobs for young people, which may help at the marriage 'entry point' but raises costs for employers and working-age people. (2) Housing and childcare capacity, which tends to work in cities but raises questions of allocation between urban and rural areas. (3) Larger cash benefits for families with children, extending the 3.6 trillion yen 'acceleration plan', which tends to ease costs for existing families more than it raises the number of births. The AI's view is that (1) should be the core, combined with (2) and (3).
Another view holds that marriage and childbearing are personal choices, and falling births simply reflect changing values and lifestyles. From this standpoint, designing policy around a 'target' number of births is itself the problem, and the priority should be redesigning social security and local systems for a shrinking population.
How it is made: claude-opus-5-5 · prompt 8dc838ce54de