The price
of time.
Across the world, and within every single country, money buys years. This is a survey of how much — and why the same dollar buys wildly different amounts of lifespan depending on where, and who, you are.
A number that sounds wrong until you check it
The conversation that prompted this was about US Social Security — and a claim that low-income earners live seven or eight years less than high earners. That number turns out to be an understatement.
The most rigorous version of this comes from Raj Chetty and colleagues at Stanford, Harvard, and the US Treasury, who linked 1.4 billion tax records to Social Security death records between 1999 and 2014. Comparing the richest 1% of Americans to the poorest 1% at age 40, they found a life expectancy gap of 14.6 years for men and 10.1 years for women. Zoom out to lifetime-earnings quintiles going back to men born in 1930, and the gap was already 5 years — and for men born in 1960, it has only widened, because gains in longevity over the late 20th century went almost entirely to the top.
This isn't a US idiosyncrasy, though the US shows it most starkly. It's the sharpest local expression of a pattern that holds, in some form, in nearly every country and at nearly every zoom level: national wealth, individual income, education, race, and geography all leave a measurable fingerprint on how long you live. What follows is a tour through each layer. (It’s a companion to Cheap Things, Expensive Life, which looked at how unequally the US distributes wealth in the first place.)
The income ladder is a lifespan ladder
Chetty's data lets you place any American on an income-to-longevity gradient. It rises steeply at the bottom, and — this is the counter-intuitive part — keeps rising all the way to the top, with no plateau.
A few features of this data are worth sitting with. Life expectancy differences are almost entirely about the bottom, not the top — going from the 10th to 14th percentile of income (roughly $14,000 to $20,000 a year) buys about the same extra life expectancy as going from the 95th to 100th percentile (roughly $224,000 to $1.95 million). And the gap is growing: between 2001 and 2014, the richest 5% of Americans gained about 3 years of life expectancy. The poorest 5% gained essentially nothing — 0.32 years for men, 0.04 for women.
Geography compounds it. Among low-income Americans specifically, where you live can swing life expectancy by up to 4.5 years — poor New Yorkers live meaningfully longer than poor residents of Gary, Indiana, mostly tracking differences in smoking, obesity, and exercise rather than differences in access to medical care.
The "hypothetical" example isn't hypothetical
The line you'd expect from an article like this — that a low-income American lives no longer than someone in a much poorer country — turns out to be understated by the actual data.
This last comparison, from the Institute for Health Metrics and Evaluation's 2025 county-level study, is the cleanest way to see it: if US college graduates were their own country, they'd rank 4th in the world for life expectancy — just behind Japan and Switzerland. Americans without a high school diploma would rank 137th, among countries like Iraq and Kyrgyzstan. Same passport, same healthcare system, a 36-year gap in life expectancy exists between counties depending on which of these two groups dominates the population.
The global relationship between wealth and lifespan
This is the "Preston Curve," first described by demographer Samuel Preston in 1975: national GDP per capita and life expectancy are tightly related — but with steep diminishing returns. Below is that relationship using 2024–2026 data across 56 countries spanning the full income range.
The shape matters more than any single country's position on it. Going from $1,000 to $5,000 in GDP per capita is associated with a huge jump in life expectancy — often 10+ years — because that range is where clean water, basic nutrition, vaccination, and maternal care become widely available. Going from $50,000 to $100,000 per capita buys almost nothing further in aggregate life expectancy. Money's return on lifespan is front-loaded: it does its heaviest lifting when it's lifting people out of absolute deprivation, not when it's making already-comfortable people more comfortable.
The exceptions are as interesting as the rule. Some countries sit well above the curve for their income level — Costa Rica, Cuba, and Sri Lanka are classic examples, all with life expectancies rivaling much richer nations, largely credited to strong, low-cost primary care and public health investment relative to GDP. Others sit well below — historically Russia and South Africa, and, as the next section covers, the United States underperforms its own income level too.
Spending the most, buying the least
If the Preston Curve were destiny, the world's richest large country by GDP per capita would also lead the world in life expectancy. It doesn't — not by a wide margin.
In 2024 the US spent about $14,775 per person on healthcare — nearly $5,000 more than the next-highest spender, Switzerland, and nearly three times the OECD average of ~$5,000. For that money, US life expectancy sits at around 79 years, two full years below the OECD average and third-lowest among OECD members, ahead of only Mexico and Türkiye.
Where does the money go instead of years? A large share goes to administrative overhead — estimated at up to 30%+ of US health spending going to billing, insurance administration, and compliance, roughly double the share in comparable countries. The US also spends far more per capita on inpatient/outpatient care and drugs without correspondingly better outcomes on the metrics that move life expectancy most: obesity (43% vs. 26% OECD average), opioid mortality (223 per million vs. 30 per million OECD average), and gun violence, none of which more spending on hospitals and specialists fixes directly.
This is the piece that complicates a simple "money buys life" story. At the system level, how money is spent matters more than how much. Universal, low-friction primary-care-oriented systems — common across the peer countries in the chart above — convert healthcare dollars into life-years far more efficiently than the US's fragmented, high-overhead, specialist-heavy model.
The other dimensions that predict lifespan
Income is the headline variable, but it travels with — and is partly a proxy for — several other factors that independently move the needle.
The steepest US gradient of all
By 2019, US college graduates lived 11 years longer on average than adults without a high school diploma — up from an 8-year gap in 2000. The divide is still widening: life expectancy for the least-educated group hasn't moved in two decades, while it rose 2.5 years for graduates.
73.5 → 84.2 yrs across education tiersA gap that narrowed, then widened again
The Black–white life expectancy gap fell from over 14 years in 1900 to about 3.6 years by 2019 — a genuine achievement. COVID reversed roughly a decade of that progress in two years. As of 2023: Asian Americans 85.2 yrs, Hispanic 81.3, white 78.4, Black 74.0, American Indian/Alaska Native 70.1.
70.1 to 85.2 yrs across US racial/ethnic groupsRural America is falling behind
A 60-year-old rural man can now expect to live 2 fewer years than an urban counterpart — a gap that's nearly tripled in two decades, driven by smoking, obesity, and shrinking rural healthcare access. For women the gap is smaller (about 6 months) but still widening.
Rural-urban gap ~3x wider than 20 years agoUniversal coverage, not just GDP, predicts outcomes
Countries with universal, primary-care-first systems (Japan, Spain, most of the OECD) consistently outperform their GDP-predicted life expectancy. The US, with its market-based, high-out-of-pocket system, is the clearest outlier in the opposite direction among wealthy nations.
70+ countries now have universal coverageThese factors aren't independent of income — education correlates strongly with earnings, race correlates with historical and ongoing wealth gaps via redlining and segregation, and rural areas tend to be poorer. But researchers have tried to isolate them, and each retains explanatory power on its own. For instance, one Urban Institute study found that up to 87–89% of the national Black-white life expectancy gap for men and women is attributable to within-county inequality — meaning it's not simply that Black Americans live in different (poorer) places than white Americans; even within the same county, race predicts a mortality gap that income alone doesn't erase.
What actually predicts how long you live
Ranking these factors by how much life expectancy variation they explain, roughly from most to least powerful:
(e.g. Chad → Norway)
(no diploma → college grad)
(bottom 1% → top 1%, men)
(AIAN → Asian American, 2023)
(men, age 60)
Which country you're born in still dwarfs every within-country factor — nothing an individual American does moves the needle 30 years. But within any given country, especially the US, income, education, and race turn out to be comparably powerful to each other, and each is large enough that the "hypothetical" framing this article opened with — a poor person in a rich country living no longer than an average person in a much poorer one — is, for a meaningful slice of the population, simply accurate.
The uncomfortable finding isn't that money buys time. It's that how unevenly a country distributes its wealth predicts how unevenly it distributes life itself — and the US, despite outspending every peer nation on healthcare, has built a system where that distribution is unusually steep.