An open, growing record
Click any country. Where WMFC has coded the history, you get the workforce story: what portability did to the people who teach and the institutions that train them. Where we haven't yet, you get an honest blank — coverage you can watch fill in, not a confident guess.
Coverage depth, not Readiness Index categories. The shading here records how far our research has got in each country. It is unrelated to the Readiness Index categories, which code individual institutions. The four European and Chilean equilibrium cases are written up in full in International Precedents; the Americas were coded in August 2026 and live in the panel only.
Shading encodes how deeply WMFC has researched a country — a fact about our coverage, not a judgment about the country's policy. It is deliberately not a "more choice / less choice" scale, and uses one hue only. The stone countries are the empty track: the absence read against the coded ones. England is recorded under the United Kingdom. Equirectangular projection; geometry simplified for file size.
This is a baseline, not a finished map. It ships mostly empty on purpose. A country shaded stone means "no WMFC entry yet" — absence of evidence, not evidence of absence. The map's value is that it shows its own coverage honestly rather than papering over the gaps with thin summaries. The empty track beside the coded countries is the finding.
Every coded finding carries a source and the month it was checked, and every entry is read through one lens only: supply and workforce — who is allowed to operate, and who is trained and paid to staff it. The coded observations describe observable policy behavior, so either side of the choice debate can cite the same country.
New in August 2026: North and South America. Every country in both continents has been read and classified — 26 coded, 6 shortlisted with a stated reason for each. Coding a country is not the same as finding portability in it: Cuba, Mexico, Brazil, Uruguay, Costa Rica, El Salvador, Panama and Venezuela are coded entries recording that public money does not follow the child. A null that has been checked is a finding; a blank is not.
The durable asset here is the data spine, not the picture. Each country is one record — coded observation, founding year, mechanism, sourced findings, checked dates — so the map, the country briefs, and a future downloadable dataset all read from the same file. Coverage grows by adding records, never by rebuilding the map.