How This Research Was Done
The approach behind the dashboard: why no one has counted Americans abroad, why we count from the receiving side, and the rules every page follows. Adapted from AER’s research.
The Uncounted Diaspora
Two presumptions structure most public discussion of Americans abroad: that there are about nine million of them, and that we know who they are. Neither survives scrutiny. The nine-million figure has circulated in State Department communications for two decades; it is not a counted number, and the Department has acknowledged in writing that no supporting data series exists.
The structural reason is fundamental. The United States has no federal civil register on the European model: no equivalent of the Nordic Folkeregister or the Spanish Padrón. The Census Bureau has excluded Americans abroad from the decennial count since 1976. What exists instead is a federation of partial instruments: absentee-ballot registrations, IRS filings from abroad, Social Security payments to foreign addresses, each a partial view of a partial subset, none designed to be a count.
The asymmetry is striking: the American statistical apparatus knows who comes in, in granular detail, and elects not to know who goes out.
Counting From the Other Side
If the American state will not count its emigrants, the only alternative is to ask the receiving states. That is what this project does: a systematic compilation of what 99 host countries publish about their US-national populations, researched country by country from national statistical offices, migration ministries, residence-permit administrations, civil registries, and Eurostat.
The single most important methodological fact is that countries do not count Americans by the same definition. Some count citizenship (most European registers), some count birthplace (the census tradition), some count permit categories. These produce categorically different numbers for the same country, and the difference is real information, not error.
We do not normalise to a single definition, because there is no defensible normalisation. Every page reports what each country publishes, against the definition that country uses, with the caveat in front of the number rather than buried behind it.
Counted, Not Modeled
Every figure in this dashboard is something a government actually published: census tabulations, population registers, residence permits, naturalization records. We do not estimate, model, or interpolate. Where a country publishes nothing, the page says so plainly: the absence is itself a finding about that country's statistical apparatus.
The research restricts itself to publicly available data. Where a country's administrative system plausibly holds the numbers internally but does not publish them (several Gulf states, for example, manifestly enumerate US nationals for visa management), we record that finding and pursue it through formal information requests to the relevant authorities.
How Each Country Was Researched
Each country page is the product of one structured exercise, carried out with local-language proficiency: identify every institution in the jurisdiction that produces migration-related data, map its published datasets, and assess for each whether US nationals can be isolated within it: by what definition, with what time-series depth, and with what reliability.
Each country is then graded on the Data Visibility Index: five kinds of publication (resident US citizens, US-born residents, annual arrivals, naturalizations, permits), each scored independently from 0 to 5. The grade measures data publication, not population size: a low score means the public record is thin, not that few Americans live there. Every score is published and open to challenge.
From the Research to This Dashboard
The full research exists as a master volume of 99 country chapters with four featured analyses; this dashboard is its data layer. The volume holds the prose, the dashboard holds the numbers, and the two reconcile through the per-country data-quality grade that bridges them.
Every figure on every page names its source, and every country page links the exact publishing authority and dataset. If a number looks wrong, the path to checking it is on the same page.
What Is Not in Scope, and Why
The dashboard covers 99 countries out of roughly 193 UN member states. The boundary is deliberate. Countries are outside the launch scope for one of three reasons, and the scope is reviewed as conditions change.
Countries where US sanctions regimes, travel bans or suspended consular services make ordinary American residence impractical to measure or pursue.
Countries currently without a functioning national migration administration capable of producing usable statistics.
The remaining countries, across Sub-Saharan Africa, the Pacific, Central and South Asia, MENA, Latin America and South-East Asia, where the measured American resident population is very small and no descent-, retirement- or investment-based pathway currently draws one. Several candidates from this group are held under review for future coverage.
Metric FAQ: What Each Number Counts
The definitions that govern every page, organized as answers; which kind of official source counts each metric, what it counts, whether it is a stock or a flow, and how the per-capita figure is computed.
Comparability labels: each metric on a country page carries a High / Medium / Low comparability chip, derived mechanically from that country’s domain-coverage record: High where the domain is published as an official annual series, Medium where the publication is partial or one-off, Low where nothing official is published.
The full series, every year, every country
The public view shows a limited year window for each series. Researchers and journalists can request access to the complete annual record for every country in the project.