What this tells policymakers
A map of where risk is concentrated, and why
Forced labor is hard to count and easy to hide, which is why the conditions that enable it are worth measuring directly. This index does not estimate how many people are exploited in a given country. It measures how strongly the structural environment favors exploitation, so that attention and resources can be directed before the harm is visible rather than after.
What the index is for
Prevalence figures answer one question: how many people are in forced labor now. They are essential, and they are also slow, uneven, and absent for much of the world. A structural index answers a different question. It asks where the conditions that produce forced labor are present, whether or not a case has yet been documented. That distinction matters most for prevention. The places where conditions are worst are often the places where evidence is thinnest, because weak institutions both enable exploitation and fail to record it.
Used well, the index is a triage instrument. It points to where vulnerability, weak enforcement, and financial opacity hold together, and it does so consistently across countries that are otherwise difficult to compare. It is a starting point for inquiry, not a verdict on any single country.
How to read it
Three readings carry most of the index’s practical value, and each comes with a discipline attached.
Read the structure, not a body count
A higher score means more of the enabling conditions are present together. It does not mean more victims, and a lower score does not certify a country as clean. The index measures the environment in which exploitation becomes more likely, rather than the exploitation itself.
Read inside the country, not only the national figure
A national score averages a whole country into one number, and that average conceals the regions where risk actually concentrates. Where the data allow, the subnational maps surface those regions directly. For programming and enforcement, the regional view is frequently the more actionable one. The national figure tells you which countries to attend to; the regional surface tells you where inside them to look first.
Read the corridors, not isolated countries
Conditions enabling forced labor often span neighboring states, and they do not respect the borders that policy is organized around. A corridor that crosses three countries is not three separate problems. Where risk clusters regionally, a response confined to one country addresses a fraction of it.
What it does not say
The index correlates with weak governance, and it is meant to. Weak governance is among the most established structural drivers of forced labor, so an index that scored fragile states as low-risk would be measuring something other than what it claims to. That relationship is reported rather than engineered away; and the index has been confirmed not to be merely a restatement of it: about a third of the variation in the score is independent of governance, and a forced-labor-specific signal survives once governance is held constant. The one thing it cannot do is benchmark itself against a clean count of actual cases, because no such governance-independent count exists. That gap is the reason a structural measure is useful in the first place.
The index is not a prevalence estimate, not an accusation against any government, and not a substitute for the people who know a place from the ground. It is most useful read alongside that knowledge. Where a country’s data are too thin to rank it fairly, the build says so rather than assigning a misleading number. The honest answer to “how does this country rank” is sometimes that it cannot be ranked yet, and the index says that plainly.
Why isn’t the index checked against a count of actual cases?
This is the most important thing to understand before using it. The index was tested against an external estimate of how many people are in forced labor, and after netting out governance on both sides, no significant association was found. This is reported plainly; but it is uninformative rather than damning, because the benchmark itself is heavily entangled with governance (its own correlation with rule-of-law runs roughly 0.5 to 0.9). There is no clean, governance-independent count of forced-labor prevalence anywhere to validate against. That absence is the gap a structural measure exists to fill: where conditions are worst, evidence is thinnest, because weak institutions both enable exploitation and fail to record it.
From "where is it worst" to "where could action be taken"
The risk score answers where conditions concentrate. The natural next question for a policymaker (where could those conditions be changed) is the index’s second product. Because the model locates not only the drivers of risk but the conditions that modulate or could defeat it, it points to candidate levers: a recruitment-fee ban, a labor inspectorate, beneficial-ownership transparency, status regularization. These are flagged as candidate intervention points, where the structure is sensitive, not as forecasts that acting there will work, which only evaluation on the ground can establish.
See the Intervention page →: the levers, by phase, with a note on who could act on each.
This index measures the structural conditions associated with forced labor, not its prevalence. It is a tool for identifying where risk is concentrated and why, to be read alongside, not in place of, on-the-ground knowledge.