Skip to content

CALCCC LAB · DATA STORY 04

Life expectancy vs. health spending: 15 countries, one chart

More spending goes with longer lives — until it doesn’t. The United States spends the most of these 15 countries and still trails Japan, Spain, and South Korea.

Our World in Data · UN · OECD · 2023 7 min read No sign-up · no downloads

Key takeaway

In 2023, life expectancy at birth among these 15 countries ranged from 72.0 years in India to 84.7 years in Japan. Health spending per person ranged from about $312 (India) to $12,099 (United States). Higher spending is associated with longer lives, but the curve flattens — and the United States is a clear outlier: the highest spending and a life expectancy below Japan, Spain, South Korea, Italy, and Switzerland. Correlation is not causation.

Rich countries generally live longer, but the relationship is not a straight line. As health spending rises, extra spending appears to buy smaller and smaller gains in life expectancy — and some high-spending countries trail peers that spend far less.

This chart compares 15 countries in 2023, the latest year with a common observation in the Our World in Data series. Each dot is a country; hover or select one to see its values. The countries were chosen to span the spending range, not sampled at random.

Interactive chart

Each dot is a country

Hover, tap, or tab to a dot to see its values. The horizontal axis is logarithmic.

Life expectancy at birth against health spending per person, 15 countries, 2023 Life expectancy ranges from 72.0 years in India to 84.7 years in Japan. Health spending per person ranges from about 312 to 12,099 international dollars. Higher spending is associated with longer lives, with a flattening at the top and the United States below the trend. 70 75 80 85 $300 $500 $1,000 $2,000 $5,000 $10,000 Life expectancy (years) Health spending per person (international $, log scale) United States Switzerland Norway Germany Canada Australia United Kingdom Japan South Korea Italy Spain Chile China Mexico India
Period life expectancy at birth and health spending per person (PPP, 2020 international $), 2023. Association is not causation. Source: Our World in Data, UN WPP 2024, OECD (2026).

United States

Life expectancy79.3 years
Health spending$12,099 / person

Use the dropdown to highlight a different country. The 15 countries were chosen to span the spending range, not sampled at random.

What the chart shows

The vertical axis is period life expectancy at birth. The horizontal axis is health expenditure per capita in international dollars at 2020 prices, adjusted for purchasing power. The x-axis is logarithmic because spending spans nearly 40-fold, from about $312 to $12,099. A log scale makes percentage differences easier to see and stops the United States from squashing everyone else into the left edge.

Japan (84.7 years, about $4,960) and Spain (83.7 years, about $3,903) sit above the trend. The United States (79.3 years, about $12,099) sits below it. Chile (81.2 years, about $3,003) and China (78.0 years, about $1,305) show how much life expectancy can improve at modest spending levels.

The high-spending plateau

The plateau is the interesting part. Going from roughly $300 to $3,000 per person is associated with a large gain in life expectancy; going from $3,000 to $12,000 is associated with a much smaller one. That pattern is consistent with diminishing returns, but it is not proof of one. Countries also differ in inequality, diet, air quality, road safety, obesity, smoking, health-system design, and age structure.

The United States is not a counterexample to the value of health care; it is a reminder that spending is an input, not an outcome. High spending can coexist with shorter lives if other factors — overdoses, firearms, road deaths, metabolic disease, unequal access — pull the other way.

Why correlation is not causation

A scatter plot shows association. It cannot tell you whether spending caused the difference in life expectancy, whether longer-lived populations demand more care, or whether both are driven by income. The comparison is also ecological: country averages say nothing about individuals. A country with high average life expectancy can contain large gaps between groups.

Period life expectancy is a snapshot, not a forecast. It applies one year’s age-specific death rates to a hypothetical person; it does not follow a real generation. COVID-19 affected 2020–2022 values, and the UN’s 2024 projections excluded those years when fitting trends.

How to read the data

The 15 countries span the spending range but are not a random sample, so the chart is illustrative. Health spending covers all financing schemes: government, compulsory insurance, voluntary insurance, and out-of-pocket payments, converted to international dollars at 2020 prices. The latest common year in this series is 2023.

If you take one thing from the chart, take this: more spending is associated with longer lives up to a point, but the countries with the best outcomes are not simply the biggest spenders. Policy, behaviour, inequality, and environment are part of the story, and this chart cannot separate them.

Method and assumptions

How the numbers were built

Data: Our World in Data, “Life expectancy vs. health spending”, 2023 values (downloaded 19 November 2026). Life expectancy: UN World Population Prospects (2024). Health spending: OECD Health Expenditure and Financing Database (2026).

  • Life expectancy at birth (period estimates, both sexes, age 0): UN World Population Prospects (2024), processed by Our World in Data.
  • Health expenditure per capita (total, all financing schemes): OECD Health Expenditure and Financing Database (2026), in international dollars at 2020 prices (PPP), via Our World in Data. Latest common year: 2023.
  • The 15 countries span the spending range but are not a random sample. The chart uses a logarithmic x-axis and a linear y-axis, with no regression or causal model.
  • Not modelled: income, inequality, lifestyle, environment, health-system quality, age structure, or individual outcomes. Country averages are ecological comparisons.

Sources

Where the numbers come from

  1. Our World in Data — “Life expectancy vs. health spending”, 2023 values (downloaded 19 November 2026) The chart source for the 15 country values, including both underlying datasets.
  2. United Nations — World Population Prospects (2024) Source of the period life expectancy estimates used by Our World in Data.
  3. OECD — Health Expenditure and Financing Database (2026) Source of health expenditure per capita, total financing schemes.

All sources retrieved 19 November 2026. Figures are reported as published; derived values are labeled in the method notes.

FAQ

Frequently asked questions

Does spending more on health make people live longer?

The chart shows an association, not proof of cause and effect. Higher-spending countries generally live longer, but income, inequality, lifestyle, environment, and health-system design also differ.

Why is the United States an outlier?

It has the highest spending among these countries but a lower life expectancy than several that spend much less. Researchers point to unequal access, metabolic disease, overdoses, firearms, and road deaths. The chart shows the pattern, not the explanation.

Why use 2023 data?

2023 is the latest common year in the series used here. Newer values are incomplete for some countries, and mixing years would confound time effects with country differences.

What is period life expectancy?

A snapshot: the average years a newborn would live if that single year’s age-specific death rates stayed the same for life. It is not a forecast for a real generation and is affected by unusual years such as 2020–2022.

What does “international dollars at 2020 prices” mean?

A purchasing-power-adjusted unit: spending is converted so a dollar buys the same basket across countries, and adjusted for inflation to 2020 prices. It compares real resources, not exchange rates.

Why these 15 countries?

They span the spending range from about $300 to about $12,000 per person, making the shape of the relationship visible. The selection is illustrative, not a random sample or a health-system ranking.

This is an educational data story about population averages. It is not medical advice and it cannot predict how long any individual will live.