Fertility Rate 1.15 on a ¥7.3 Trillion Budget: Evaluating Three Years of Japan's Children and Families Agency

The budget of the Children and Families Agency began at ¥4.8 trillion in its inaugural FY2023, and reached ¥7.3 trillion in FY2025. Compared with the baseline figure of ¥4.7 trillion — the combined pre-establishment total of the relevant ministries in FY2022 — that represents an increase of roughly 55%.

Over the same period, Japan's birth indicators moved in the opposite direction.

The total fertility rate in 2024 fell to 1.15, a record low and a drop of 0.11 points from 1.26 in 2022. The number of births was 686,061, the first time it had fallen below 700,000 since records began in 1899. The trend into 2025 is no better: births in January–June were 3.1% below the same period the previous year at 339,000, a new record low.

Spending increases while demographic indicators worsen. This inverse correlation is not a coincidence. Three structural problems compound one another.

What the "¥7.3 Trillion" Actually Buys

Breaking down the FY2025 budget, nursery school and after-school club operations account for roughly ¥2.5 trillion, the child allowance (including expansion after the income ceiling was abolished) for approximately ¥2.1666 trillion, and parental leave benefits for roughly ¥1.06 trillion. These three items alone account for approximately 77% of the total.

The first layer of the problem is the "rebranding" structure. When the Children and Families Agency was established in April 2023, it absorbed the MHLW's Child and Family Bureau (responsible for childcare, child abuse, and social care) and the Cabinet Office's Child and Child-Rearing Headquarters (responsible for early childhood education, childcare grants, and child allowance) in their entirety. 350 of the agency's 430 staff were transfers from existing ministries, and virtually all of the ¥4.8 trillion inaugural budget consisted of programs that had simply been moved across.

The "doubling of the children's budget" proclaimed by the Kishida administration also requires scrutiny. More than 80% of the ¥3.6 trillion Acceleration Plan — approximately ¥3.0 trillion — was said to have been realized in FY2025, but this assessment is based on spending volume, not outcomes. A substantial share of the increase represents expanded coverage of existing programs — child allowance and parental leave — rather than new spending whose effect on birth numbers or marriage rates has been measured.

Failing to Address the Root Causes

Thirty years of structural failure, from the 1994 Angel Plan onward, are embedded in the cumulative record. Cumulative pro-natalist spending since 2004 has exceeded ¥66 trillion (Nikkei Shimbun). One broader estimate, covering all child-related public spending, places the figure above ¥130 trillion — a comparison that requires caution because the two figures use different scope definitions. Yet births have continued to fall throughout.

Why? The root causes identified by researchers lie in areas where current child-rearing policy has little traction.

First, the acceleration of non-marriage and later marriage. The average number of children born to married couples has held relatively stable at roughly two, so the decline in the fertility rate is driven primarily by fewer marriages, not by fewer births per married couple. Japan's out-of-wedlock birth rate is only 2–3% — compared with around 62% in France and 55% in Sweden — meaning that declining marriage translates almost directly into declining births.

Second, the compound effect of precarious employment and economic factors. Men in non-regular employment report a desire to marry at a rate of 49.5%, compared with 69.1% for men in regular employment. The national burden rate has nearly doubled from 24.3% in FY1970 to 46.8% (projected) in FY2023, and rising housing costs in major cities continue to suppress household formation among young adults.

The Children and Families Agency has no authority to directly intervene in any of these root causes — employment is under the MHLW, housing under the Ministry of Land, Infrastructure, Transport and Tourism, and education under the Ministry of Education. The agency can coordinate but cannot compel, and this structural constraint fundamentally limits its reach.

The Problem of Formalistic EBPM

The Children and Families Agency established an EBPM promotion office at its founding, setting approximately 320 indicators and preparing "EBPM sheets" for roughly 100 programs. Analysis by Deloitte Tohmatsu acknowledges this effort while also identifying its limits.

The majority of the indicators measure outputs — number of nursery school users, take-up rate of parental leave. Direct causal linkage to outcome indicators — whether fertility rates and marriage rates are actually improving as a result of specific policies — remains weak. Structural limitations reinforce the problem: the time-lag between nursery school expansion and its effect on fertility can exceed five to ten years, and because most of the budget consists of transferred programs, constructing a credible counterfactual is difficult.

The claim that "more than 80% of the Acceleration Plan has been realized" is, once again, a spending-based assessment, not an outcomes-based one.

South Korea as a Cautionary Precedent — and What the Successes Required

Japan's demographic trajectory lags South Korea's by roughly 15–20 years. The Korean experience is instructive: despite investing 280 trillion won (approximately ¥31 trillion) over 16 years from 2006 to 2021, the total fertility rate in 2023 fell to 0.72, the lowest in the world. The failure is attributed to housing cost inflation driven by concentration in Seoul, the divergence between women's rising educational attainment and persistent gender role expectations, and employment precarity among young adults — all of which closely parallel Japan's structural conditions.

Countries that have achieved partial success offer different lessons. Sweden, as analyzed by the Dai-ichi Life Research Institute, combined more than 40 years of sustained investment, 480 days of parental leave with 80% income replacement, multi-child bonus allowances, and strong promotion of paternity leave — recovering from 1.5 in 1999 to 1.98 in 2010. The decisive factor was policy consistency and comprehensiveness over time, not any single year's budget figure. France is often cited as a success story, but in 2024 its fertility rate fell to 1.62, the lowest in over a century — a reminder that no country has found a durable solution.

Professor Shintaro Yamaguchi of the University of Tokyo (economics) has argued that Japan's child-related spending needs to exceed 3% of GDP. The current ¥7.3 trillion represents roughly 1.3% of GDP — below the OECD average of 2.34%. Economic evidence also indicates that in-kind benefits such as childcare infrastructure are roughly five times more effective per yen at raising fertility than cash transfers, yet cash benefits (the child allowance) command a larger share of the ¥7.3 trillion than in-kind services (childcare facility expansion).

The Structural Paradox of the Support Contribution

The situation is further complicated by the Child and Child-Rearing Support Contribution, which began collection in April 2026. Collected as a surcharge on social insurance premiums, it will reach approximately ¥1 trillion per year at full implementation in FY2028. An employee earning ¥8 million annually will pay ¥767 per month in FY2026, rising to ¥1,350 per month at full implementation.

The structural paradox lies in the source of the revenue. One of the root drivers of declining marriage is the erosion of young workers' disposable incomes. Raising social insurance premiums increases labor costs for employers, compressing the headroom for wage increases. A pro-natalist policy funded by a mechanism that further squeezes the financial position of the young workers most constrained from forming households contains a structural self-contradiction that has not been adequately addressed.

Budget expansion functions as a political signal that "something is being done," but there is no automatic correlation between spending volume and policy effectiveness. Three questions deserve sustained attention. First, when will the cross-ministerial policy intervention needed to address the root causes — employment, housing, gender equality — actually begin? Second, can EBPM be transformed from the production of evaluation sheets into the genuine verification of causal mechanisms and iterative policy revision? Third, how can the political pressure for short-term results be reconciled with the decade-long time horizon required for policies to show demographic effects?

South Korea's experience teaches that policy design quality is more decisive than the scale of funding. ¥7.3 trillion is a starting point for the question, not an answer.

For an analysis of the funding structure of child-rearing policy, see "The 'Singles Tax' Unmasked." For the quality implications of the drive to eliminate childcare waiting lists, see "Zero Waiting-List, Quality in Crisis."

For further reading, 子育て支援の経済学(The Economics of Child-Rearing Support) by Shintaro Yamaguchi (Nihon Hyoronsha) uses empirical analysis to compare the birth-rate effects of childcare expansion, parental leave policy, and cash transfers — finding that in-kind services are approximately five times more effective than cash payments. The book provides the evidentiary foundation for the claim referenced in this article and won the 64th Nikkei Economic Book Culture Award.

References

Budget, Accounts, Taxation and Special Account Disclosure — Children and Families Agency (2025)

Overview of Vital Statistics 2024 (Final) — Ministry of Health, Labour and Welfare (2025)

Birthrate Keeps Falling, Cumulative Spending Exceeds ¥66 Trillion — Nikkei Shimbun (2024)

FY2025 Budget: Children and Families Agency at ¥7.3 Trillion — KSI Policy News (2025)

'Child-Related Spending Must Exceed 3% of GDP': Interview with Prof. Yamaguchi, University of Tokyo — Tokyo Shimbun (2022)

Lessons from Sweden's Pro-Natalist Policy — Dai-ichi Life Research Institute (2023)

Is There a Way Out of Falling Fertility? France's Birthrate Continues to Decline — Dai-ichi Life Research Institute (2024)

South Korea's Total Fertility Rate in 2023 Was 0.72 — JETRO (2024)

Challenging EBPM: The Children and Families Agency's EBPM Promotion Office Gets Underway — Deloitte Tohmatsu (2023)

¥130 Trillion Invested in Pro-Natalist Measures, Yet Births Are Down 33% — PRESIDENT Online (2025)

Child-Rearing Support Levy: An Employee Earning ¥8M Will Pay ¥767 per Month — Nikkei Shimbun (2025)

Chapter 2, Section 2: Declining Birthrate and Household Finances — Cabinet Office (2023)