By the end of this century, the United States is projected to have more than one million centenarians — people aged 100 or older — for the first time in human history, in a demographic shift driven by sharply falling mortality among nonagenarians, and some demographers now believe that a majority of children born today in low-mortality countries can reasonably expect to reach their hundredth birthday

In 1950, the United States Census Bureau estimated that approximately 2,300 Americans were aged 100 or older. The total population of the country at the time was approximately 152 million, which made centenarians one of the rarest possible demographic categories — approximately one in every 66,000 Americans had reached the age. By 1990, the centenarian count had risen to approximately 37,000. By 2020, it had reached approximately 92,000. By 2024, it had passed 101,000. The trajectory has been steep and accelerating, and the demographic shift now visible in the centenarian numbers is one of the larger long-term changes in the structure of the American population — a category that was statistically negligible at the end of World War II is now projected to become, within the lifetimes of most people reading this, a substantial demographic constituency in its own right.

According to the Pew Research Center’s January 2024 analysis of Census Bureau projections, the US centenarian population is projected to more than quadruple from approximately 101,000 in 2024 to approximately 422,000 by 2054 — meaning that within the next 30 years, the proportion of Americans aged 100 or older will rise from approximately 0.03 percent of the population to approximately 0.1 percent. This is the official Census base-case projection. Higher-end estimates developed by demographic researchers — including the MacArthur Foundation Research Network on an Aging Society — have projected substantially higher numbers under more optimistic mortality-improvement scenarios, with some estimates exceeding 1.8 million US centenarians by mid-century and substantially larger numbers by 2100. The range between the Census base case and the higher-end projections is large, and the actual outcome will depend on whether mortality improvements at older ages continue at recent rates, accelerate, or plateau.

What is driving the projected growth

The proximate driver of centenarian growth is not improving mortality among the young or middle-aged — those mortality rates are already very low and have only modest room to fall. The driver is sharply declining mortality among nonagenarians: people aged 90 to 99 who, in previous generations, would typically have died before reaching 100. The mortality rate among 90-year-olds has been falling steadily across developed countries for decades, with each successive cohort of nonagenarians surviving slightly longer than the previous one. The result is that the share of any given birth cohort that eventually reaches 100 has been rising, and the absolute number of centenarians is rising even faster because the population at risk (the 90-year-olds in the country) has also been growing rapidly.

Per a CEOWORLD analysis of the global centenarian surge, the same pattern is visible internationally. The United Nations estimated approximately 722,000 centenarians worldwide in 2024 and projects that the number will rise to approximately 4 million by 2054 — a nearly six-fold global increase in 30 years. By mid-century, the largest national centenarian populations are projected to be China (approximately 767,000), the United States (approximately 422,000), India, Japan, and Thailand, with several Southeast Asian countries and parts of southern Europe showing particularly rapid centenarian growth on a per-capita basis. Japan, currently the country with the world’s highest median age, is projected to have approximately 40 centenarians per 10,000 people by 2054 — meaning that roughly one in every 250 Japanese residents will be aged 100 or older. Thailand is projected to have an even higher ratio, at approximately 49 per 10,000.

The contested longer-term projections

The substantive disagreement among demographers is not about whether centenarian numbers will rise — that is now essentially settled — but about how far they will rise, and how soon a majority of births in low-mortality countries will produce people who eventually reach 100. The optimistic position, associated with the demographer James Vaupel and colleagues at the Max Planck Institute for Demographic Research, holds that life expectancy in the most advanced countries has been increasing by approximately three months per year for the past 150 years and shows no clear sign of slowing — and that this trajectory implies that the majority of children born in low-mortality countries today will live to see their hundredth birthday. The skeptical position, associated most prominently with the American demographer S. Jay Olshansky and his collaborators, holds that life expectancy gains are now slowing substantially in the United States and other developed countries, that the easy mortality reductions from infectious disease control, smoking declines, and cardiovascular medicine have largely been achieved, and that further gains will be substantially harder to obtain.

The empirical evidence for the slowdown position has become more visible in recent years. US life expectancy peaked at approximately 78.9 years in 2014 and then declined — partly because of the opioid crisis, partly because of rising obesity-related mortality, partly because of COVID-19 mortality during the pandemic, and partly because of slowing gains in cardiovascular medicine. The current US life expectancy of approximately 76 years is substantially below the trajectory that the Vaupel projections would have predicted. As covered by a foundational MacArthur Foundation Research Network paper on demographic forecasts for the United States in the 21st century, the higher-end centenarian projections depend on assuming that the recent slowdown is temporary and that the longer-run Vaupel-style trajectory will resume. Whether it does is one of the central empirical questions in contemporary demography, and the answer will substantially affect both the centenarian counts in 2054 and the much larger projections for 2100.

The maximum lifespan question

As reported by a 2021 Bayesian analysis published in Demographic Research by Michael Pearce and Adrian Raftery at the University of Washington, the related question of what the maximum human lifespan will reach by 2100 is approachable with statistical methods even where the underlying biology is uncertain. The Pearce-Raftery analysis modelled the existing distribution of supercentenarians (people aged 110 or older) and the projected growth of that population through the rest of the century, and estimated that there is greater than 99 percent probability that the current record human lifespan — Jeanne Calment of France, who died in 1997 at 122 years and 164 days — will be broken before 2100. They estimate an 89 percent probability that someone will reach age 126, a 44 percent probability that someone will reach 128, and a 13 percent probability that someone will reach 130. These projections are not predictions about the typical lifespan but about the extreme tail of the survival distribution, which depends on the same nonagenarian mortality declines that drive the centenarian numbers.

The deeper question that the centenarian projections raise is the relationship between living longer and living well. The available evidence on healthspan — the proportion of life spent in good health, rather than in disability or substantial decline — suggests that healthspan has been increasing more slowly than lifespan. The result is that the gap between the two has been widening: people are living longer, but with more years of substantial disability, more years of cognitive decline, and more years of dependence on caregivers and the healthcare system. The economic implications are substantial. The current pension systems, healthcare systems, and family structures of most developed countries were built around assumptions about average lifespan and the typical duration of post-retirement life that no longer hold. The institutional response, across most of the developed world, has been slower than the underlying demographic shift. Whether the eventual centenarians — including, possibly, a substantial fraction of the children born this year — will arrive at their hundredth birthdays in good health, in supported decline, or somewhere in between will depend on whether the gains of the coming decades are made primarily in extending lifespan or primarily in extending healthspan. The two trajectories produce very different futures.