SHARE · ELSA · HRS — Europe, England & the United States · ages 50+
Does health follow your rank, or your riches?
Older people who sit higher in their country's income distribution report better health. Switch the horizontal axis between relative position (income rank) and absolute purchasing power (PPP income); compare 28 European countries plus England and the United States; and use the matched-income spotlight to ask it directly — is it better to be rich in a poor country, or poor in a rich one?
Self-rated health by income rank
How to read this. Each point is a within-country income group (a decile = tenth, or a quintile = fifth): poorest on the left, richest on the right. Groups are computed within each country and survey wave, so "rank" means position among one's compatriots. The vertical value is the mean of the chosen measure for that group.
Measures. Self-rated health (1–5, reversed so 5 = excellent) is the only measure harmonised across all three surveys, so England and the United States appear only for it. Life satisfaction (SHARE CASP-12, 12–48) and depression (EURO-D, 0–12, higher = worse) are SHARE-only, because ELSA/HRS use different well-being and CES-D scales.
Age adjustment. "Age-adjusted" directly standardises each group to a common 5-year age distribution, so cross-country differences aren't driven by one place being older than another. "Raw" is the unadjusted mean.
Absolute income. Household income is converted to PPP-adjusted international dollars (World Bank 2017 price levels, US = 1), so €1 in Sofia, £1 in London and $1 in Ohio sit on a common purchasing-power scale. Each group is plotted at its median income. The axis is piecewise: it gives the crowded $0–$50k range most of the width and compresses everything above $100k into a short tail (so tick spacing narrows toward the right); the very few groups above $250k sit at the edge, with their true value in the tooltip. The matched-income spotlight reads off every country's health at one income — the spread at equal income is the relative / country-context effect.
Survey wave. The default ~2015 view shows each place at its wave nearest 2015; countries that joined later (much of Central & Eastern Europe, at 2019) and the non-SHARE comparators are drawn dashed with their actual year. SHARE single-wave options cover SHARE countries only; "All waves" pools each place across its own rounds (a repeated cross-section, not a within-person panel).
Sources. SHARE Rel 9.0.0 (28 countries, Israel excluded); ELSA and HRS via their harmonised files (RAND HRS). Household income (SHARE net; ELSA/HRS total). Descriptive means, unweighted. Built for a PSE master's thesis on relative income and health.