Japan’s Singles Are Opting Out of Marriage—What the New Government Survey Actually Says

For the first time since Japan began asking unmarried adults this question in 1982, fewer than four in five people aged 18–34 say they intend to marry someday. The National Institute of Population and Social Security Research’s 17th National Fertility Survey, released 11 September 2026, is not a mood piece. It is a primary dataset—and it rewrites a line many Japan explainers still treat as settled.

Japan’s demographic story is usually told with births and deaths. The Year the Lines Crossed walks that chart through 2025: more deaths than births every year after a one-year pause, a fertility rate that no longer replaces the population, and a marriage count that still decides almost every birth. What the vital statistics cannot show is whether the next generation of adults still wants the institution those births run through.

On 11 September 2026, the National Institute of Population and Social Security Research (IPSS / 国立社会保障・人口問題研究所) published the results of the 17th National Fertility Survey (第17回出生動向基本調査)—also titled the National Survey on Marriage and Childbirth. Fieldwork was in June 2025. The sample: 6,756 valid responses from unmarried people aged 18–54, and 5,024 married couples in which the wife was under 55. The previous wave was 2021.

This article reads that survey the way The Japanist reads other Tier-1 sources: for what it measured, what it did not, and what English readers should actually take away.

The number that broke a forty-year streak

Among unmarried people aged 18–34—the core series IPSS has tracked for decades—75.1% of men and 77.8% of women said they “intend to marry someday” (いずれ結婚するつもり).

Both figures fell by more than six percentage points from the previous survey (81.4% for men, 84.3% for women). Both are below 80% for the first time since the unmarried series began in 1982.

That is the headline Japan’s English press led with, correctly. It is also easy to oversell. A three-quarter majority of young unmarried adults still say they plan to marry. The break is not that marriage intention vanished. The break is that the “almost everyone still says yes” floor, which held for more than forty years of this series, did not hold this time.

The complementary figure is sharper for people who want a never-marry narrative. Among the same age band, those who said they never intend to marry rose to 24% of men (+6.7 points) and 21.5% of women (+6.9 points)—above 20% for the first time. IPSS’s own summary adds a caution English coverage sometimes skips: among those never-intent respondents, roughly four in ten said there is still a possibility they might shift back to “intend to marry someday.” Preferences are softening; they are not a sealed fate for every individual in the cell.

Age detail sharpens the picture. IPSS notes especially large drops for women 30–34 (marriage intention 70.0%, from 77.5%) and for men and women in their late twenties. Never-marry intent among men 30–34 reached 34.1%; among women 30–34, 28.9%.

(IPSS overview; IPSS summary PDF; Japan Times)

“No benefits,” and what marriage used to buy

The survey does not only ask whether people want to marry. It asks whether marriage still looks like a good trade.

IPSS reports that the share who think marriage has advantages fell from the 60s into the 50s for men, and from about 70% into the 60s for women. Press coverage of the release put the “no benefits” share at 42.8% of men and 36% of women—about seven points higher than the previous wave.

Which hoped-for gains eroded matters more than the abstract “no benefits” line. Fewer people said marriage would let them have children or a family, gain peace of mind (精神的やすらぎ), or achieve greater financial security. The item that rose was living with someone you love (愛情感じる人と暮らせる)—companionship as the remaining pitch, while children, calm, and money look less automatic.

That is a cultural shift with a traveler’s implication: Japan’s low non-marital birth share means partnering and parenting are still tightly coupled in practice. When marriage looks less like the route to family, money, or rest, fertility does not get an easy off-ramp through cohabitation the way it does in much of Europe. The vital statistics and this survey are describing the same gate from two sides.

How people meet now

Among married couples in a first marriage of less than five years, 20.2% said they met online (ネット)—up from 11.0% in the previous wave—now essentially level with workplaces (20.9%) and introductions by friends or siblings (20.2%).

That is not a Silicon Valley morality play. It is a logistics fact. If “no suitable partner” (適当な相手がいない) remains the top reason unmarried people 25 and older give for still being single—even as that share declined slightly while never-intent rose—then matching channels matter. Online dating is no longer a fringe path into Japanese marriage; for recent newlyweds it is a mainstream one.

(Japan Times; IPSS couples survey summary)

Loneliness, money, and never holding a baby

IPSS’s public commentary on the release is unusually plain. An institute official told reporters that more people believe they will not feel lonely even if they keep living alone, while more people have never interacted with infants—and that both trends link to lower desire to marry and have children.

The overview’s unmarried lifestyle block points the same way: more respondents report having little money for things they want, and more say solo life is not lonely. Desired number of children among the unmarried kept falling (1.64 for men, down from 1.83; 1.70 for women, down from 1.79). About one in six unmarried men and women said that even if they marry, they do not want children. Support for the norm that “one should have children after marriage” also weakened.

Contact with infants is not a soft lifestyle detail in this framing. The share of never-married 18–34-year-olds who said they often had contact with babies or small children fell to 32.5% of men and 42.8% of women. People with less contact show lower marriage intention and lower desired children. That is IPSS’s measured association, not a causal proof—and it is one of the more actionable findings for anyone designing family policy or community programs, because exposure to young children is a different lever from housing subsidies alone.

Partnered life is thin before marriage, too: 19.6% of unmarried men and 27.8% of unmarried women reported having a romantic partner; about one in three (roughly 35%) said they do not want to date.

Married Japan is not standing still either

The couples survey is not a story of a stable married majority quietly having two children while singles opt out.

The average ideal number of children among married couples was 2.18, down from 2.25. More telling for behavior: mean intended/planned children fell to 1.95—below 2.0 for the first time in the modern series going back at least to the 1977 survey. Ideal is still near replacement; planned children are not. The gap between what couples say they want and what the country records each year remains the policy problem Japan has been naming for a generation.

When asked why ideal exceeds planned, the top reason was still cost: 52.9% cited that raising and educating children costs too much (up from 46.0%). Psychological and physical childcare burden also rose.

Reproductive medicine is part of that gap’s texture: 22.8% of couples—about one in 4.4—had undergone infertility testing or treatment (essentially unchanged from 22.7%), while 42.1% said they worried about infertility (up from 39.2%). That is not evidence that technology will close the birth shortfall. It is evidence that a large minority of married Japan is already medicalizing the path to a child.

On gender ideals, both unmarried men and women most often chose “balance work and childrearing” as the preferred women’s life course—about four in ten. The survey is not returning a 1950s housewife script as the modal dream. It is returning a dual-earner ideal that Japan’s workplaces and care infrastructure still make expensive to live.

What this survey is not

It is not proof that “Japan is dying,” a phrase that flattens a slow, measured shift into catastrophe tourism.

It is not a forecast that three-quarters of today’s 18–34-year-olds will remain unmarried. Intention is not behavior; the never-intent group itself reports room to change; and people marry after 34.

It is not a claim that dating apps caused the birth decline. Online meeting rose among recent marriages—the people who did marry—while overall marriage intention fell. Those can be true together.

It is not a substitute for the vital statistics. Births, deaths, and marriage registrations still decide the annual ledger. This survey decides whether the next ledger has a willing denominator.

And it is not a moral ranking of Japanese young adults. Wanting solitude without loneliness, delaying partnership, or doubting that marriage delivers money or rest are preferences shaped by housing costs, work hours, gender expectations, and weak exposure to infants—as IPSS itself emphasizes—not a national character flaw for foreigners to diagnose from a café in Shibuya.

What English readers should watch next

If you follow Japan for travel, living, or business, treat this release as a baseline, not a vibe:

1. The next vital-statistics marriage count — intention falling is one series; registered marriages are another. Watch whether the 2025–2026 registration totals move with the attitude shift. (Related: The Year the Lines Crossed)
2. Housing and working-hour politics — when “financial security” and “peace of mind” stop looking like marriage benefits, policy that only lectures people to marry harder will miss the survey’s own language.
3. Infant-contact and community design — IPSS tied low touch with babies to low marriage and birth desire. That is a research prompt for local governments, not a reason for tourists to panic.
4. How couples meet — online matching at ~20% of recent first marriages is now a structural fact of Japanese partnering. Platforms, fraud, and manners around dating will matter more in everyday Japan coverage.
5. Immigration and labour debates — they sit beside this chart, not instead of it. A country with fewer marrying cohorts and a low non-marital birth share does not get the next generation from tourism. (Japan Wanted Foreigners…)

Bottom line

The 17th National Fertility Survey did not invent Japan’s demographic squeeze. It measured something the birth totals only implied: for the first time in this unmarried series, the default “yes, someday” answer to marriage no longer clears 80% among 18–34-year-olds.

Most young unmarried adults still say they intend to marry. A rising minority says they never will. A larger share sees no benefit in the institution. Recent newlyweds increasingly met online. Married couples still name an ideal near two children, while a substantial fraction have already sought infertility care. More people say living alone is not lonely, and fewer have ever held a baby.

That is the survey. Everything else is commentary. The Japanist’s job is to keep those two layers separate—and to update the old line that “Japanese singles still almost all want to marry” the moment the official series says otherwise.

Primary sources: National Institute of Population and Social Security Research,『第17回出生動向基本調査（結婚と出産に関する全国調査）　結果の概要』, published 11 September 2026 (概要 PDF; press release; results page). English title on the Institute’s PDF: The 17th Japanese National Fertility Survey, 2025 — Summary of the Survey Results. Secondary: The Japan Times, 11 September 2026. Fieldwork: June 2025 (as of 30 June). Valid responses: 6,756 unmarried (42.7% valid rate); 5,024 couples (47.9%). Response rates fell versus the 2021 wave, and this was the first wave offering an online option—worth keeping in mind when comparing waves. IPSS does not require citation permission; The Japanist attributes figures to the Institute’s published summary.