observations with both measures after the project’s stated cleaning steps
Health economics · 2021 cross-section
What does more
health spending buy?
Across countries and economies, higher spending is linked to longer lives—yet the United States sits strikingly below the fitted global pattern.
years associated with a 10% rise in per-person health spending
of cross-sectional variation explained by the log-linear fit
years: the U.S. gap below its fitted value
01 · THE PATTERNSpend more. Live longer.
Spend more. Live longer.
With one costly exception.
Every dot is one observation with available 2021 data. Search, select, and switch scales to inspect how the pattern changes.
Quick picks
Loading World Bank observations…
Other observationsUnited StatesOLS fit
The log view matches the model used in the project. Switching to linear reveals how concentrated most observations are at lower spending levels.
02 · THE OUTLIER
The United States spends the most—and misses the line.
At roughly $12,080 per person, the United States is the highest-spending observation in the dataset. Its life expectancy was 76.3 years; the fitted model predicts about 84.0 years.
Actual76.3
Fitted84.0
This gap is descriptive. It does not identify why the U.S. differs, and the model does not estimate a causal effect.
03 · THE MODEL
Life expectancy = 47.6 + 3.86 × ln(health spending)
10% more spending+0.37 years95% CI: +0.34 to +0.40
Slope precisionSE 0.17p < 0.001
Model fitR² 0.689cross-sectional fit
How to read the coefficient
Because spending enters as a natural logarithm, equal percentage increases have comparable predicted associations. A 10% increase corresponds to about 0.37 additional years of life expectancy—roughly 4.4 months.
The confidence interval reflects uncertainty around the fitted association, not proof that spending itself causes the difference.
04 · METHOD
From World Bank rows to one fitted relationship.
The website mirrors the data workflow in Jackson’s R project and adds an interactive layer for exploring each observation.
- 01
Source
2021 World Development Indicators for current health expenditure per capita and life expectancy at birth.
- 02
Reshape
Pair the two indicators by country or economy, remove missing values and nonpositive spending, then exclude the named broad regional and income groups in the project.
- 03
Estimate
Fit ordinary least squares with life expectancy as the outcome and the natural log of spending as the predictor.
- 04
Interpret carefully
Describe a cross-sectional association. Do not treat it as a causal estimate or a complete model of population health.
WHAT THIS STUDY CAN SAY
Countries and economies that spend more per person on health tend to have longer life expectancy in the 2021 cross-section.
WHAT IT CANNOT SAY
That higher spending alone causes longer lives. Income, public health, demographics, institutions, inequality, and many other factors may shape both variables.
SOURCE & REPRODUCIBILITY
Built from the project’s original data and R analysis.
Data: World Bank, World Development Indicators. Values shown are from 2021 and use current U.S. dollars for spending.