Why women live longer than men: the gap held through famines, epidemics and slavery in seven populations

A girl born in the United States in 2023 could expect to live to 81.1, and a boy to 75.8. In the words of the federal statisticians who compile the figures, the difference in life expectancy between females and males was 5.3 years. Some version of that gap appears in nearly every country that keeps records.

The usual explanation is behavior. Men have smoked more, take more risks and die more often from accidents and violence. All of that matters. A team of demographers led by Virginia Zarulli at the University of Southern Denmark wanted to know whether it is the whole story, and found a grim way to check. They looked for populations in which conditions were so bad that almost nobody lived long enough for lifestyle to make a difference.

Why extreme hardship is a test of why women live longer

There was already evidence that the gap does not depend entirely on behavior. It persists among nonsmokers. It persists in religious groups such as active Mormons or cloistered monks and nuns, where men and women lead far more similar lives than the general population does. It also shows up among newborns and infants, when differences in behavior are unlikely to matter much.

Zarulli and colleagues, writing in the Proceedings of the National Academy of Sciences in 2018, turned the question around. They describe the reverse situation, when both men and women experience high, perhaps extreme, levels of mortality risk, as an untapped source of information. In a famine or an epidemic the differences in habit that separate men from women in ordinary times are swamped. Everyone is starving, or everyone is exposed to the same infection. If women’s longer lives were purely a product of how the sexes behave, the gap should close.

The authors spelled out what would count against the biological view. A finding that men and women have similar life expectancies under these conditions would challenge the notion that the survival advantage of women is fundamentally biologically determined in all environments.

Seven populations where almost everyone died young

The team assembled life tables, the demographer’s record of how many people die at each age, for seven populations with reliable enough data.

Population	Life expectancy in years, men	Life expectancy in years, women
Freed American slaves in Liberia, 1820 to 1843	1.68	2.23
Ukraine, famine of 1933	7.3	10.9
Plantation slaves in Trinidad, 1813 to 1816	15.18 to 19.45	13.21 to 20.58
Iceland, measles epidemic of 1882	16.76	18.83
Sweden, famine of 1773	17.15	18.79
Iceland, measles epidemic of 1846	17.86	18.82
Ireland, famine of 1845 to 1849	18.7	22.4

These are some of the worst conditions for which records exist. Among the freed slaves who sailed from the United States to Liberia and met diseases they had never encountered, about 43% died during the first year, and the authors describe the data as showing the highest mortality ever registered in recorded human history. In Ukraine, life expectancy had been above 40 for both sexes in the five years before the famine. In Iceland the measles epidemics roughly halved it within a single year.

What the crisis records showed about women’s survival

Life expectancy was higher for women than for men for all populations, with the partial exception of the Trinidad slaves. The advantage was sometimes small, about half a year in Liberia, and sometimes large: in relative terms women’s life expectancy exceeded men’s by almost 50% in Ukraine, 33% in Liberia and 20% in Ireland. A second measure agreed. The age to which the hardiest 5% of a population survived was higher for females than for males in all seven. “Our results confirm the ubiquity of a female survival advantage even when mortality is extraordinarily high,” the authors wrote.

Where the advantage came from is the most telling result. When the authors divided the gap by age, they found that the biggest contribution came from strikingly large mortality differences between male and female infants. In Liberia, Trinidad, Iceland and Ireland, the first year of life accounted for more than half of the whole difference between the sexes. After the first birthday, the gap between boys and girls and between men and women contributed less and less.

That matters because a baby has no habits. A newborn does not smoke, take risks or avoid the doctor. The authors drew the conclusion carefully: that females survived more than males even at the infant ages, when behavioral differences are minimal, lends support to a biological underpinning of the female survival advantage. The Duke University account of the study summarized the scale of the effect, saying that women still lived longer than men by six months to almost four years on average.

The exceptions are as instructive as the rule. On the Trinidad plantations, under one of the two scenarios the historians reconstructed, men lived slightly longer than women, and young women died at higher rates than young men. The authors suggest that young adult men had lower mortality than young women, perhaps because a premium was placed on their survival, since male slaves commanded higher prices. Their reading is that in famines and epidemics the female advantage holds across ages, whereas in slave populations, where the conditions of life were under some human control, it could be overridden.

What biology may lie behind women living longer than men

The historical records show that the advantage exists. They do not show what produces it, and for that the authors turn to laboratory and comparative research.

Sex hormones are one candidate. In the paper’s summary of the literature, estrogens have antiinflammatory, vasoprotective effects and enhance immune defenses, whereas testosterone seems to increase the mortality risk for certain diseases, although the evidence on this point is mixed.

Chromosomes are the other. Women carry two X chromosomes and men an X and a much smaller Y, so a harmful mutation on a man’s single X has no healthy copy to cover for it. Zoe Xirocostas and colleagues at UNSW Sydney tested that idea, known as the unguarded X hypothesis, across the animal kingdom. In mammals the male has the mismatched pair, and in birds it is the female. Across 229 species, the sex with two matching sex chromosomes lived 17.6% longer on average than the sex with the reduced one. In mammals, females tend to live longer than males, while in birds, males tend to live longer than females.

Two larger surveys have since filled in the picture. Jean-François Lemaître at the University of Lyon and colleagues compiled data from 134 populations of 101 wild mammal species and found the female’s median lifespan to be on average 18.6% longer than the male’s, whereas in humans the female advantage is on average 7.8%. By mammal standards, in other words, the human gap is modest. They found no consistent sex differences in the rate of aging itself, and they concluded that the size of the gap is shaped by local environmental conditions as well as by the costs of competing for mates.

In 2025 Johanna Staerk of the Max Planck Institute for Evolutionary Anthropology and colleagues, reporting in Science Advances, extended the comparison to zoos, where predators, hunger and many injuries are largely removed. Among 528 mammal species, 72% showed a female advantage in adult life expectancy, and among 648 bird species 68% showed a male advantage, as the chromosome theory predicts. Chromosomes alone could not account for the variation, though. The gap was smaller in monogamous species and larger where males are much bigger than females and compete for mates, which suggests that some of the male disadvantage is the price of traits evolved for that competition.

How much of the gap between men and women is behavior

A biological base does not make the gap fixed. Its size has changed a great deal, and the changes trace to how people lived.

Hiram Beltrán-Sánchez and colleagues followed 1,763 birth cohorts from 1800 to 1935 in 13 developed countries. They found that a large excess of female life expectancy in adulthood is a demographic phenomenon that emerged among people born in the late 1800s. Before then, adult men and women died at fairly similar rates. Afterward, deaths among men aged 50 to 70 pulled ahead, with heart disease the main condition involved. Smoking was part of it. The authors estimate that smoking-attributable deaths account for about 30% of excess male mortality at ages 50 to 70 for people born between 1900 and 1935, and that substantial excess male mortality remained after accounting for smoking.

A later paper by Zarulli looked at which ages drive the gap. Young men are the group people usually have in mind: between ages 15 and 40, death rates for men are usually two or three times higher than death rates for women. Deaths at those ages are rare, however, so the effect on life expectancy is limited. Higher male mortality between 15 and 40 accounts, depending on the country and time, for less than 25% of the life expectancy gap. Before about 1950 the decisive age was infancy, and since 1950 high male mortality after age 60 has largely determined the life expectancy gap. The paper adds that the gap has narrowed in recent decades in step with the convergence of smoking between men and women.

The United States figures show the same thing from another angle. At birth the difference is 5.3 years. Among people who have already reached 65, the difference in life expectancy at age 65 between females and males was 2.5 years.

Longer lives for women, and fewer of the extra years in good health

Outliving men is not the same as staying healthier than men. Zarulli and colleagues note that in most countries the difference between the sexes in healthy life expectancy is smaller than the difference in total life expectancy. They give the example of France and Sweden in 2015, where the gender gap in life expectancy at 65 was 4.1 and 2.6 years, while the gap in healthy life expectancy at the same age was only 0.9 and 1.1 years.

The authors put the point bluntly: women live more years than men and are able to do so even though they are in bad health for a substantial part of those extra years of life. Survival and health appear to be separate matters, and women’s advantage lies in the first.

Weak points in the evidence on why women live longer

The crisis data are reconstructions. Historians built them from parish registers, slave registers, censuses and ship lists, and the paper is candid about the weak points. Infant deaths in particular tend to go unrecorded when mortality rises, and infants are where most of the female advantage was found. In the Liberian records the sex of 181 of the 4,472 cases could not be determined. Causes of death were known for only one of the seven populations.

Infancy is also less free of social influence than it looks. A baby has no habits, but parents do, and the authors point to studies of preindustrial Europe in which sons and daughters were treated differently when food was short. That leaves room for a social contribution even in the first year of life, in either direction.

Nor do the historical records test any mechanism. Hormones and chromosomes are plausible explanations supported by animal and laboratory work, and the paper itself describes the evidence on testosterone as mixed. Zarulli and colleagues summarize their own result in two parts that have to be read together: the survival advantage of women has biological underpinnings, and it differs across environments and is modulated by social factors.

Women outlived men even where life expectancy fell below twenty years, which points to a biological base for the gap, while its size in any one place and time still depends on how men and women live.