Aging population
Dependency, the oldest old, the workforce that carries them and what the state already pays — the demographic pressure on the US economy
Thesis
Is population aging becoming an economic constraint?
- The dependency ratio and the share over 80 rise without pause.
- Working-age growth slows; older workers offset part of it.
- Pension and health spending takes a growing share of GDP.
Index
Five indicators, standardized and equal-weighted, updated monthly — 100 = January 1979, higher means more demographic pressure
Aging Pressure Index
149.3
1 mo
↑ +1.3
1 yr
↑ +10.1
Source: Liquiqui, from OECD / BEA
Where each constituent stands today, in standard deviations from its own average since 1979
- Old-age dependency ratioAnnual · Jan 1, 2025+2.6
- Population 80+ (share)Annual · Jan 1, 2025+2.0
- Pensions + health / GDPQuarterly · Apr 1, 2026+1.8
- Employment 55-64 (inv.)Monthly · Sep 1, 2026-1.8
- Working-age growth (inv.)Monthly · Sep 1, 2026+2.2
Notes
- What is happening
- The index stands at 149.3 in September 2026 against a base of 100 in January 1979. It drifted without direction through the 1980s and 1990s — 117 in 1990, 114 in 2000 — while the boomers filled the workforce, then climbed in one movement from 2008: 127 in 2010, 139 at the end of 2019. Its all-time high, 151 in April 2020, was the pandemic: older workers' employment rate lost four points in two months and spending jumped against a collapsed GDP, before both recovered. Today the dependency ratio, a working-age population shrinking on a year earlier, the share of the population aged 80 and over and age-related spending are the most stretched constituents (+2.6, +2.2, +2.0 and +1.8 standard deviations), and older workers' record employment rate is the only one pulling the other way (−1.8).
- Why it matters
- Aging is not one number but a tension between two forces: the burden — more retirees per worker, more of the very old, more pension and health spending — and the offsets, a workforce that keeps growing or older people who stay in work. This index adds the first three at equal weight and subtracts the two offsets, each standardised against its own history first, to answer one question: is the pressure building or easing?
- What's driving it
- The baby boomers passing 80, a birth rate below replacement since 2007, and immigration slowed since 2025.
- What to watch
- 2026 opens the phase in which the boomers reach 80. Demography is the most predictable series on this site: the people who will be 80 in 2040 are already born, and only immigration can still move the denominator. The one genuine unknown is the employment rate of older workers — it has the most room to move and the most effect.
Signals
Each source of demographic pressure, scored by how far it has moved — and the arrow points the way the pressure is heading
| Criterion | Current Reading | Trend | Comment |
|---|---|---|---|
Old-age dependencyin the index How many people aged 65 and over for every hundred of working age? | High | ↑ | 29.6 per hundred in 2025, from 19.5 in 2010 and 12.5 in 1950. Ten points in fifteen years, as the boomers cross 65 — the last of them do so in 2029. |
The oldest oldin the index Is the 80+ population, where care needs concentrate, growing faster than the rest? | High | ↑ | 15.0 million in 2025, 4.4% of the population, against 3.3% in 2000. The first boomers turn 80 in 2026, so this is where the next fifteen years' acceleration lies. |
Working-age populationin the index Is the 15-64 population still growing, and how fast? | High | ↑ | 210 million in September 2026, down 0.9% on a year earlier. Growth ran 2.2% a year in 2000 and 0.6% by 2010; since 2020 it has swung between −0.9% and +1.6% with immigration and with the population controls the BLS applies each January. The underlying trend is a plateau. |
Older workersin the index Are 55-64 year-olds staying in work? | Low | ↓ | 65.6% employed in September 2026, a record, from 57.9% in 2000. Participation is 67.1%, level with its record. Every point gained here is a cohort carrying part of its own cost — the strongest offset on the page. |
Age-related public spendingin the index What share of GDP already goes to Social Security, Medicare and Medicaid? | High | ↑ | 12.4% of GDP in the second quarter of 2026 — 5.0 points Social Security, 4.0 Medicare, 3.4 Medicaid — against 11.2% at the end of 2019, 8.0% in 2000 and 6.0% in 1980. The trustees' 2026 report dates the retirement trust fund's depletion to late 2032, and health is the one line on this page with no natural ceiling. |
Productivity offsettable only Is output per hour rising fast enough to make up for a workforce that has stopped growing? | Moderate | → | GDP per hour grew 1.9% a year over 2019-2025, better than the 0.8% of the 2010s, while total hours grew only 0.5% a year. All of the growth is coming from productivity now. Kept out of the index: it moves with the cycle and with technology, not with demography. |
Raw data
Old-Age Dependency Ratio
Old-age dependency ratio
29.6%
1 yr
↑ +0.8%
Source: OECD
Notes
- What is happening
- 29.6 people aged 65 and over per hundred of working age in 2025, against 19.5 in 2010, 18.8 in 2000 and 12.5 in 1950. The line was almost flat from 1980 to 2010, then gained ten points in fifteen years: that is the boomers, born 1946 to 1964, crossing 65.
- Why it matters
- The old-age dependency ratio divides those aged 65 and over by those aged 15-64: in a single line, how many retirees each potential worker carries. Public pensions and Medicare are pay-as-you-go, so the ratio is also, mechanically, the price of those systems.
- What's driving it
- The last boomers reach 65 by 2029 while the cohorts entering working age are the smallest since the 1970s.
- What to watch
- The last boomers reach 65 in 2029; the ratio should pass 33 by then and level off through the 2030s before a second, slower climb driven by longer lives. Japan stands at 55 and Italy at 40: the United States ages later and more slowly, thanks to higher fertility and immigration.
Population Aged 80 and Over
Population 80+ (millions)
15.0
1 yr
↑ +0.54
Source: OECD
Notes
- What is happening
- 15.0 million people aged 80 and over in 2025, 4.4% of the population, against 9.3 million and 3.3% in 2000 and 5.2 million in 1980. Growth has been steady — roughly 200,000 a year — because the cohorts reaching 80 so far are the small ones born through the Depression and the war.
- Why it matters
- The cost of aging is not linear in age: health and long-term care spending per person rises steeply after 80. The size of this group therefore matters more than that of the 65+ for Medicaid, which pays for most nursing-home care, and for the care workforce.
- What's driving it
- Life expectancy at 65 gained three years since 1990, and the large 1940s cohorts are now crossing 80.
- What to watch
- The first boomers turn 80 in 2026. The pace moves from about 200,000 to over 500,000 a year, and the Census Bureau projects more than 25 million people aged 80 and over by the early 2040s. That is when long-term care demand becomes the story.
Working-Age Population
Working-age population (millions)
210.4
1 yr
↓ -1.9
Source: OECD
Notes
- What is happening
- 210.4 million in September 2026, against 178 million in 2000 and 198 million in 2010. Annual growth fell from 2.2% in 2000 to 0.6% in 2010 and has hovered around zero since 2020: +1.6% in January 2025, −0.9% in September 2026. Those swings reflect the population controls the BLS applies each January as much as actual immigration — read the series as a trend, not month by month.
- Why it matters
- The population aged 15-64 is the denominator of everything else: the tax base, the potential workforce, the people who contribute. Its growth was for decades the quiet engine of the US economy — close to 2% a year through the 1990s — and its plateau is the concrete form aging takes.
- What's driving it
- Fewer births since 2008 and the collapse of net migration since 2025 leave the 15-64 population shrinking on a year earlier.
- What to watch
- Without immigration this population would already be shrinking: the birth cohorts of 2005-2010 barely replace those leaving. Any year in which net migration falls below about 500,000 turns the line down — and that is now happening: the Census Bureau puts it at 1.3 million in 2025, from 2.7 million in 2024, and projects about 320,000 for 2026. It is the policy variable most directly tied to this page.
Employment Rate, 55-64
Employment rate, 55-64
65.6%
1 mo
↑ +0.4%
1 yr
↑ +1.2%
Source: OECD
Notes
- What is happening
- 65.6% in September 2026, a record, against 57.9% in 2000 and 51.6% in 1985. The rise reflects the full-retirement age moving from 65 to 67, the disappearance of defined-benefit company pensions, and less physical work. The pandemic cost two points, since recovered.
- Why it matters
- The employment rate of 55-64 year-olds is the most powerful offset to aging: an older person in work contributes instead of drawing, and the group is 40 million strong. Each point of this rate is worth roughly 400,000 jobs.
- What's driving it
- The full-retirement age moved to 67, defined-benefit pensions gave way to 401(k)s, and work shifted from physical to office jobs.
- What to watch
- Japan is at 80% and Germany at 75%, so the room to rise is real. August and September 2026 are the first two readings above 65%; holding there would say the trend is intact, while a fall back below 62% would say early exits — often health-related — are winning again.
Labor Force Participation, 55-64
Participation rate, 55-64
67.1%
1 mo
↓ -0.1%
1 yr
↑ +0.6%
Source: OECD
Notes
- What is happening
- 67.1% in September 2026, level with August's record, against 59.4% in 2000 and 54.0% in 1985. Women drove the rise, closing much of the gap with men's participation between 55 and 64.
- Why it matters
- The participation rate adds job-seekers to those in work: it measures the will to stay in the labour market regardless of the cycle. The gap with the employment rate is unemployment among older workers, structurally low but long when it hits.
- What's driving it
- Women's participation after 55 caught up with men's, and better health keeps people looking for work later.
- What to watch
- A point and a half above the employment rate, so unemployment among older workers of about 2% — within the norm. Participation, more than employment, says whether pension reforms change behaviour: its plateau since 2010 suggests most of that effect is already in.
Employment and Labor Force
Employment (millions)
163.2
1 mo
↑ +0.4
1 yr
↓ -0.5
Labor force (millions)
170.3
1 mo
↑ +0.5
1 yr
↓ -1.0
Source: OECD
Notes
- What is happening
- 163.2 million employed and a labour force of 170.3 million in September 2026, against 136.6 and 144 million in 2000. Since late 2019 employment has gained 4.4 million in nearly seven years — half the pace of the 2010s — and it is 0.5 million below a year earlier, despite low unemployment.
- Why it matters
- The absolute levels say what rates hide: the US employs more people than ever, and the aging question is not a shortage of workers today but how fast their number can grow tomorrow.
- What's driving it
- Older workers and, until 2024, immigration added to employment even as the native working-age population stopped growing.
- What to watch
- The gap between the two lines is unemployment: 7.1 million people. What matters for this page is the slope of the labour force: below 0.5% a year, economic growth rests entirely on productivity.
Labor Productivity
GDP per hour worked (2020 USD)
85.6
1 yr
↑ +1.5
Source: OECD
Notes
- What is happening
- 85.6 dollars of GDP per hour worked in 2025 (at 2020 prices and purchasing-power parities), against 76.6 in 2019 and 56.9 in 2000. Growth ran 1.9% a year over 2019-2025, against 0.8% over 2010-2019 and 2.3% in the 2000s: a clear rebound after the lost decade, whose cause the AI Capital Spending page tries to find.
- Why it matters
- When the workforce stops growing, output per hour is the only source of growth left — and the only way to fund more retirees without raising taxes. It is the variable that decides whether aging is a problem or a transition.
- What's driving it
- Capital deepening, software and the first AI gains lift output per hour, offsetting a workforce that no longer grows.
- What to watch
- At 1% productivity growth and 0% workforce growth the economy grows 1%; at 2%, it grows 2%. The whole difference between a manageable aging and a fiscal crisis sits in that percentage point. The CBO assumes about 1.5% in its long-term projections.
Total Hours Worked
Hours worked (billions)
293.8
1 yr
↑ +1.1
Source: OECD
Notes
- What is happening
- 293.8 billion hours in 2025, against 285 billion in 2019 and 261 billion in 2000. Growth has fallen to 0.5% a year since 2019, from 1.5% over 2010-2019 and 1.1% over 1990-2007. The 2008 recession wiped out 19 billion hours and the pandemic 18 billion; each recovery has been slower than the last.
- Why it matters
- Total hours worked is the product of three things: how many people work, how many hours each, and how many are of an age to. It is the most honest measure of an economy's labour effort — and the one aging erodes first, before employment itself.
- What's driving it
- Headcount growth near zero and slightly shorter working weeks: hours grow only as fast as older workers stay on.
- What to watch
- Flat hours with a stable working-age population and record employment rates mean there is no slack left: the next recession will take hours that nothing replaces. Sustained negative growth in hours would be the first sign of aging biting into GDP.
Methodology
1Sources
OECD (dependency ratio, population 80+), BLS via FRED (employment rate 55-64, working-age population), BEA via FRED (Social Security, Medicare and Medicaid / GDP).
2Normalisation
Ratios and shares in percent; working-age growth smoothed over two years; quarterly and annual inputs carried forward on the monthly calendar.
3Z-scores
Each constituent minus its mean since 1979, divided by its standard deviation; capped at ±3. Employment of older workers and working-age growth enter inverted.
4Weighting
Five constituents, equal weights.
5Index
Average of the five z-scores × 20 points, rebased to 100 in January 1979. Updated monthly.
6Interpretation
Higher means more demographic pressure. A slow, structural series: read the level, not the month.