Inequality Explorer · World Bank PIP + WID.world · 1985–2026
0 = everyone equal · 100 = one person has everything. Dots are actual surveys; lines connect them. The spotlit country reads at full strength.
The Other Ladder · WID.world
The mountains above are what people earn in a year. This is what they own. Every country splits harder by wealth than by income, and the top of the range is where the two measures come apart.
Share of all net personal wealth held by the top 10% (solid) and top 1% (thin). Dashed spans are WID model imputations, not measured national data.
Top 10% share of income vs. top 10% share of wealth, same people, same year. The gap is the cliff.
Net personal wealth: everything households own (housing, land, deposits, equities) minus their debts. Shares are of the adult population, wealth split equally within couples. A negative bottom-50% share means that half of the country owes more than it owns; the bar then shows only the positive holdings (renormalized) while the legend carries the true signed shares. Countries marked modeled carry WID regional imputations rather than national wealth surveys — treat their levels as indicative. Countries measured only through an earlier year carry their last WID estimate forward; the split notes when that applies. The gold shard compares the top 1% share to an equal 1% share, on a square-root scale so small multiples stay visible.
Sources. Income: World Bank Poverty & Inequality Platform (PIP) — survey mean & median income/consumption (2021 PPP $ per person per day), Gini, extreme-poverty headcount at the $3.00/day line, decile shares; version 20260324, 2021 PPPs. Population: World Bank WDI. Wealth & top income shares: World Inequality Database (WID.world), 2025 update, series through 2024 — net personal wealth and pre-tax national income, adults, equal-split. Basemap: Natural Earth 110m. Pulled 2026-07-15 via API.
Method. Mountain shapes are log-normal reconstructions from each survey's mean + Gini (the standard Gapminder-style approximation) — real distributions have fatter tails, so treat the far right of any mountain as indicative. All stats shown come straight from the surveys; values between survey years are linearly interpolated (monotone cubic, so playback glides through survey points), the nearest survey year is labeled, and outside a country's survey span the nearest survey is carried flat. Years beyond the last survey — through 2026 — use the World Bank's own PIP nowcasts (growth-extrapolated means and poverty counts with the last survey's distribution shape) and are tagged nowcast. Wealth series end at WID's 2024 ceiling and hold there when the slider runs past it. Wealth series are WID Distributional National Accounts; per-country source citations appear under the wealth split.
Income ≠ wealth. The income Gini (PIP, post-tax/consumption surveys) and the wealth Gini (WID, net personal wealth) measure different things: the US sits near 41 on income and near 83 on wealth. This page shows both, on purpose. Likewise, "top 10% take home" on the cards and table is the PIP survey decile; the dumbbell uses WID pre-tax national income so income and wealth share one methodology.
Level bands use the classic Factfulness cutoffs ($2 / $8 / $32); the dashed line marks the World Bank extreme-poverty line ($3.00/day, 2021 PPP). Argentina is urban-survey-only.
Book a call and we'll map your AI Chief of Staff — the operating layer that turns your best work into something that runs without you.