moneycurio

Methodology & data sources

Every number on this site traces back to a public source. This page explains where the data comes from, how often it refreshes, and how each tool calculates its results.

Data sources

  • Income percentiles: World Bank Poverty and Inequality Platform (PIP) — national and world income distributions — plus World Bank comparative price levels.
  • Inflation (CPI): US BLS CPI-U series (1913–present), UK ONS, OECD.
  • Asset price history: Yearly closing prices from public market data; Bitcoin spot via CoinGecko. Approximate, delayed, and for illustration only.
  • Exchange rates: Frankfurter API (European Central Bank reference rates).
  • Average wages & price levels: OECD and national statistics offices.
  • Billionaire net worth: a small list compiled manually from Forbes, updated quarterly.

How fresh is the data?

Market prices and exchange rates refresh automatically every day; monthly series like CPI refresh when their source publishes. Each tool shows a "Data updated" date — currently 2026-09-07 for market data. If a source is temporarily unavailable we keep the last good value rather than guessing.

Per-tool calculation notes

As each tool launches, its exact formula and assumptions are documented here and in the "How this is calculated" section on the tool page itself.

Income percentile: your figure is converted from nominal dollars into purchasing-power (PPP) dollars using your country's price level, then compared against World Bank distributions of household income per person. That is what internationally comparable survey data measures, and it is not the same as a salary ranking: one salary supporting a family of four lands far lower per person than the same salary supporting one. The tool cannot know which case is yours, so read the result as an order of magnitude rather than a verdict.

How the data is kept honest

The refresh runs automatically four times a day and it is written to fail safely. If a source is unreachable or answers with something implausible, the job keeps the previous good value and records the failure instead of writing a guess — so a broken API shows up as slightly older data, never as a wrong number. A counter tracks consecutive failures per source, and anything stale beyond three days is flagged for us to look at by hand.

Where two independent sources exist, we cross-check them: Bitcoin and Ethereum prices are compared against CoinGecko spot after being fetched from market history, and asset series were verified against official closing values before launch. Our blog posts are unusual in one respect worth knowing — every figure in them is computed from these same data files at build time rather than typed in by hand, so a post cannot silently drift out of date while the tool it describes moves on.

Limitations

These tools are simplified models for curiosity and education. Global income comparisons use purchasing-power adjustments that carry real uncertainty; historical returns ignore taxes, fees and dividends unless stated. Average wages are averages, so most people earn less than the figure shown, and a national price level is a national average that can be far from a capital city's. Nothing on this site is financial advice.