Imagine 100,000 baby girls entering Japan’s 2025 life table. If the mortality rates observed that year remained fixed throughout their lives, 94,533 would reach 65. More than 88,000 would reach 75. Just over half—50,800—would celebrate a 90th birthday. The age at which only half remained alive would be 90.17.

That imagined cohort is a statistical instrument, not a forecast for any individual child. But it expresses the magnitude of Japan’s modern survival achievement better than a single average can. The Ministry of Health, Labour and Welfare’s abridged life tables put female life expectancy at birth at 87.33 years in 2025, up 0.20 from 2024. Male life expectancy reached 81.35, an increase of 0.25. Both sexes improved at every age shown in the ministry’s summary.

The movement reverses the pandemic-era decline more decisively than the near-flat readings of 2023 and 2024. It also keeps Japanese women at the front of the ministry’s country comparison: above South Korea’s 86.6 and Spain’s 86.53 among the listed sovereign countries, while Hong Kong is separately noted at 88.73. Japanese men stand seventh in that table. Such rankings require caution because national methods and reference years differ. The larger truth is less fragile: Japan remains one of the world’s longest-lived societies.

81.35 yearsMale life expectancy at birth in 2025, up 0.25 year
87.33 yearsFemale life expectancy at birth, up 0.20 year
5.98 yearsThe female–male gap, 0.05 year narrower than in 2024
50.8%Share of women expected to reach 90 under 2025 mortality rates
19.68 yearsRemaining life expectancy for a 65-year-old man
24.52 yearsRemaining life expectancy for a 65-year-old woman

What the headline number does—and does not—mean

Life expectancy at birth is the average number of years a hypothetical newborn would live if the age-specific death rates measured in one period never changed. It is called a period measure. It summarizes the mortality environment of 2025 without being distorted by how many people happen to be young or old.

It does not say that every boy born in 2025 will die around 81, or every girl around 87. Medical treatment, epidemics, climate, behavior and social conditions will change over their lifetimes. If mortality continues to improve, many newborns will live longer than the period figure. Nor is the average the most common age of death. Because a small number of earlier deaths pull an average downward, the life-table median is higher: 84.13 years for men and 90.17 for women.

Life expectancy is not the age written on a newborn’s future gravestone. It is a compact reading of the risks faced by people of every age in one particular year.

The anatomy of the 2025 rebound

The ministry decomposed the increase by cause of death. For men, lower cancer mortality contributed about 0.10 year to the 0.25-year gain; the easing of COVID-19 mortality contributed 0.08, heart disease 0.05 and cerebrovascular disease 0.02. For women, heart disease contributed 0.06 year, COVID-19 0.06, cancer 0.05 and cerebrovascular disease 0.01. Lower suicide mortality added 0.01 year for men and 0.02 for women.

Other forces moved against the improvement. Changes in pneumonia mortality subtracted roughly 0.02 year for men and 0.01 for women, while “senility” as a recorded cause subtracted about 0.01 for each. Those small negatives are a reminder that extreme longevity changes the composition of death. When more people survive cancer or vascular disease into advanced old age, frailty, aspiration pneumonia, dementia-related decline and multiple overlapping conditions become more visible.

The rebound therefore has two stories. One is cyclical: fewer deaths attributed to COVID-19 after the severe disruption of the early 2020s. The other is structural: continued progress against cancer, heart disease and stroke. Neither should be read as a declaration of victory. In the same life table, cancer, heart disease and cerebrovascular disease together still account for about four in ten eventual deaths under the model.

A nation that once expected fewer than 50 years

Japan did not inherit its longevity. It manufactured it through policy, rising living standards, scientific practice and millions of household decisions. The earliest national complete life tables, covering 1891–1898, put life expectancy at only 42.8 years for men and 44.3 for women. In 1935–1936 the figures were still 46.92 and 49.63. War then wrecked food supply, housing and public health.

In 1947, the first postwar life table recorded 50.06 years for men and 53.96 for women. By 1960 the figures had surged to 65.32 and 70.19. A gain of roughly 15 to 16 years had arrived in little more than a decade. No “longevity gene” changed that fast. Death retreated because conditions did.

The first stage was a broad assault on early death. Safe water, sewage and housing improved. Nutrition recovered. Antibiotics, vaccination and organized control programs drove down tuberculosis and other infections. Maternal and child-health services expanded, with public-health nurses carrying prevention into homes and communities. Fewer infants and young adults died, producing enormous gains in life expectancy even before the diseases of old age were mastered.

YearMenWomenWhat the era reveals
1891–189842.844.3Infectious disease and early-life mortality kept lives short.
194750.0653.96Japan emerged from war with severe material and health damage.
196065.3270.19Nutrition, sanitation, tuberculosis control and child survival transformed the curve.
198073.3578.76Stroke mortality was falling; medical access and living standards widened.
200077.7284.60Japan had become a global longevity leader, especially for women.
202081.5687.71Pre-pandemic gains approached a historic peak.
202281.0587.09COVID-19 and wider disruption produced a setback.
202581.3587.33Survival rebounded, led by lower cancer, cardiac and COVID-19 mortality.

1961: a social contract around access

Universal health insurance, achieved in 1961, is central to the story, but chronology matters. Much of the first great leap had already occurred through public health and social recovery. Insurance then made access to physicians and hospitals a national expectation rather than a privilege tied to wealth, employer or location. Patients could seek treatment early; providers could build routine systems for managing chronic illness; and local authorities could connect screening, prevention and care.

The Japanese model never eliminated cost or inequality, and it has been repeatedly reformed. Yet relatively broad access, controlled fees and dense medical provision helped translate economic development into population health. It also made the next phase possible: reducing deaths in middle and old age rather than only preventing death among the young.

This is why Japan’s history resists a single-secret explanation. Insurance without sanitation would have arrived too late for many children. Hospitals without blood-pressure control would not have stopped the stroke epidemic. Diet without effective treatment would not have carried as many people through cancer and heart disease. Longevity accumulated because the layers reinforced one another.

The long campaign against stroke

In the 1950s and 1960s, cerebrovascular disease—especially hemorrhagic stroke—was a defining Japanese killer. Traditional diets could be rich in fish and plant foods but also extraordinarily high in salt, particularly where preserved vegetables, miso, soy sauce and salted fish helped households through winter. High blood pressure was common, treatment was uneven and cerebral hemorrhage struck people in midlife.

Japan’s response reached far beyond specialist wards. Community screening found hypertension before symptoms appeared. Public-health campaigns pressed for less salt. Antihypertensive medicines became more widely used, nutrition improved and homes became warmer. Stroke mortality fell sharply after the mid-1960s. Researchers studying the postwar rise consistently identify this decline as one of the decisive sources of Japan’s later longevity.

The campaign also shows why averages are political. A person cannot personally install a water system, set a national fee schedule, train a public-health workforce or make affordable blood-pressure care available in a rural town. The largest survival gains were not simply rewards for individual virtue; they were returns on collective capacity.

Diet matters, but the “Japanese secret” is a myth

Food belongs in the explanation, just not as folklore. Japan historically combined low intake of red meat and saturated fat with frequent fish, soy foods, vegetables and tea. Obesity remained uncommon by international standards. Those patterns likely helped keep ischemic heart-disease mortality comparatively low even while stroke and stomach cancer were major problems.

But there was never one timeless Japanese diet. Postwar intake changed radically as rice consumption fell and animal foods, dairy and fat increased. Some of that change improved protein and micronutrient intake; excess salt remained dangerous. Today, inexpensive highly processed foods, less physical activity and metabolic disease create new pressure, while household income, isolation and access shape what older people can actually eat.

Nor can Okinawa, green tea, fish or fermented food carry the causal weight often placed on them. Genetics, early-life conditions, medical access, education, work, social connection and smoking all interact over decades. The more useful lesson from Japan is not a superfood. It is a health system and a food culture that, for long periods, kept several large risks relatively low at the same time.

Women lead the world; men still pay a larger toll

The female advantage was 5.98 years in 2025, down from a peak of 6.96 in 2005. Biology contributes to sex differences in survival, but behavior and social history enlarge or shrink them. Japanese men have carried heavier exposure to smoking, alcohol, dangerous or stressful work and delayed use of care across much of the postwar period. Women, meanwhile, benefit from lower mortality at most ages but bear a greater share of disability and very old age.

The 2025 table makes that asymmetry vivid. Under the year’s mortality rates, 26.7% of men but 50.8% of women would reach 90. A 90-year-old man could expect another 4.38 years; a woman the same age, 5.66. At 65, the figures are 19.68 and 24.52 years. Retirement policy, household savings, housing and care planning must therefore treat life after 65 as a long stage of adulthood, not a short epilogue.

Japan’s male improvement of 0.25 year, larger than the female gain, is encouraging. But ranking seventh in the ministry’s international table is not a failure, and female leadership is not a reason for complacency. The relevant questions are which premature deaths remain preventable and who is missing the protection enjoyed by the national average.

The postcode behind the average

The newest prefectural life tables available for this comparison cover 2020. Men in Shiga had life expectancy of 82.73 years, followed by Nagano at 82.68 and Nara at 82.40. Aomori was lowest at 79.27—a 3.46-year distance from Shiga. For women, Okayama led at 88.29, followed by Shiga at 88.26 and Kyoto at 88.25; Aomori was lowest at 86.33, a 1.96-year gap.

Those differences are not a league table of personal discipline. They can reflect smoking and drinking patterns, occupation, income, education, transport, winter conditions, diet, preventive services, medical access and the statistical volatility of smaller populations. The rankings also change with age: Okinawa, for example, ranked only 43rd for male life expectancy at birth in 2020 but climbed to second for remaining life expectancy at 75.

A national gain that leaves persistent regional and social gradients untouched is an incomplete victory. The next longevity strategy must be able to see smaller places and harder lives—not merely lift the top line.

Long life is not yet healthy life

Japan’s latest official healthy-life-expectancy estimates refer to 2022 and measure years lived without limitations in daily activity. They were 72.57 years for men and 75.45 for women. Compared with that year’s ordinary life expectancy, the gap was 8.49 years for men and 11.63 for women. Both gaps had narrowed significantly since 2019, but they remain the central challenge hidden inside the triumph of longevity.

Healthy life expectancy is derived differently and should not be mechanically subtracted from the new 2025 figures as though the surveys covered the same year. Still, it changes the question. A society can lower mortality while leaving people to live longer with frailty, pain, sensory loss or dementia. Women’s larger gap is particularly important: the group that lives longest also spends more years, on average, with activity limitations.

The next national milestone should not be 88 alone. It should be more years in which an 88-year-old can choose, move, remember, connect and live with dignity.

That shifts attention toward fall prevention, muscle and bone health, oral care, hearing, dementia support, mental health, social participation and homes that work for bodies at 85. It also demands support for family caregivers and a stable long-term-care workforce. Medical success can postpone death; healthy longevity requires the design of everyday life.

From 153 centenarians to nearly 100,000

When Japan began honoring centenarians under the Elderly Welfare Act in 1963, the country counted 153 people aged 100 or older. By September 2025, the official total had reached 99,763. Women numbered 87,784—about 88% of the total. The figure is both a celebration and a demographic signal.

The population aged 65 or older stood at a record 36.24 million in 2024, 29.3% of the country. Japan’s Statistical Handbook projects the share to reach 38.7% by 2070. At the same time, low fertility and population decline are shrinking the working-age base that finances and staffs pensions, health care and long-term care.

Longevity did not cause low fertility, and older people are not a burden category. Many work, care for relatives, volunteer, pay taxes and sustain communities. The policy problem is the ratio and the design: how to distribute the costs and gains of longer life fairly when the age structure is changing faster than institutions and workplaces.

The longevity paradox

Japan’s success creates the conditions for its next difficulty. Preventing a stroke at 62 may allow a person to live into the years when cancer, frailty or dementia becomes the dominant risk. A hospital can save a life that then depends on accessible housing and home care. A daughter’s long life may overlap with years of caring for a mother in her nineties. Each victory changes rather than abolishes need.

The OECD’s 2025 Japan profile captures the double picture. Japan’s combined life expectancy was about three years above the OECD average, while preventable and treatable mortality were both substantially lower. Yet self-reported poor health was more common than the OECD average, and Japan’s suicide rate remained higher. Excellent population survival can coexist with loneliness, mental distress and uneven wellbeing.

Public policy must therefore avoid two mistakes. The first is fatalism: treating longer lives as an unaffordable accident. The second is triumphalism: assuming a high average proves that every community and generation is flourishing. The goal is not simply to contain the cost of age, but to invest in the capabilities that let people contribute and live independently longer.

What Japan’s next health revolution requires

Five tests beyond the headline
  • Close the healthspan gap: measure function, cognition and participation alongside survival.
  • Reduce unequal risk: target smoking, hypertension, suicide and delayed care where they remain concentrated.
  • Build age-ready communities: connect primary care, rehabilitation, long-term care, transport, housing and social contact.
  • Protect the care workforce: improve staffing, pay, technology and working conditions without turning care into a conveyor belt.
  • Prepare for shocks: strengthen vaccination, infection control and continuity of care for older people during disasters and epidemics.

Japan has already shown that coordinated prevention can redraw a mortality curve. The tools now must become more precise. Municipal data can identify neighborhoods where hypertension, frailty or isolation cluster. Primary care can detect declining mobility before a fall. Better vaccination and ventilation can protect nursing homes. Hearing aids, sidewalks, buses and community meals may preserve independence as meaningfully as another hospital procedure.

Technology will help only when it serves human capacity. Monitoring, care robots, telemedicine and artificial intelligence may ease labor shortages and extend expertise, but they cannot replace trust, touch or local knowledge. The oldest population in the world is also a test market for whether innovation can enlarge autonomy rather than merely reduce staffing.

A number built one prevented death at a time

In 1947, only 0.9% of men and 2.0% of women in the life-table model survived to 90. In 2025, the shares were 26.7% and 50.8%. That change is so large that it can appear inevitable in retrospect. It was not. It was built through pipes and clinics, school meals and vaccines, public-health nurses and blood-pressure cuffs, insurance cards and cancer treatment, safer work and better homes.

The 2025 rebound deserves celebration because fewer people died than the previous year’s mortality pattern would have predicted. But the deepest reason to study Japan’s history is that it rejects both mysticism and resignation. Longevity is not a national character trait. It is a social outcome, repeatedly made and remade.

The average Japanese woman now stands near 87 years not because one secret was discovered, but because thousands of ordinary risks were made a little less lethal. The next revolution will be judged by something more demanding: whether those additional years belong to the people living them.

Sources and references

This report uses Japan’s official 2025 abridged life tables as its primary source. Historical, international and healthy-life-expectancy comparisons use the latest cited datasets available; reference periods and methods differ, so rankings should be interpreted cautiously.