MSRI
Multidimensional Scam Risk Index

Methodology

How MSRI turns evidence into a score, and the limits on how that score may be used. Weights, bands, and thresholds on this page come straight from the scoring code.

Five dimensions, fourteen indicators

Each indicator is scored from 0 to 100. Indicators average into a dimension score, and dimensions combine by weight into the composite. All indicators are expressed as risk, so a higher number always means more risk. The one indicator measured as capacity (survivor referral access) contributes 100 minus its value.

Governance & rule of law

Weight 25%

The primary structural enabler. Whether weak or selective governance lets compounds operate with impunity: special economic zone oversight gaps, armed or elite control of territory, and gray-market finance history.

  • SEZ oversight gap
  • Armed / elite control
  • Gray-market finance history

Labor & migration

Weight 25%

Where trafficking risk enters the index. Conditions that feed forced recruitment and block escape: irregular migration inflows, active recruitment pipelines, and lack of worker grievance access.

  • Irregular migration inflow
  • Recruitment pipeline activity
  • Worker grievance access

Financial flows

Weight 20%

The financial infrastructure that sustains compounds and launders proceeds: cross-border micro-transactions, shell company density, and crypto on-ramp exposure.

  • Cross-border micro-transaction volume
  • Shell company / nominee density
  • Crypto on-ramp exposure

Digital platform

Weight 15%

Scam-linked online activity originating from or routed through the location: scam IP and domain density, and platform enforcement actions.

  • Scam IP / domain density
  • Platform enforcement actions

Protection capacity

Weight 15%

Whether survivors can get help without being treated as criminals: referral network access, non-criminalizing enforcement, and whether past crackdowns displaced compounds rather than dismantling them. Stronger protection lowers the composite score.

  • Survivor referral network access · inverted
  • Non-criminalizing enforcement
  • Displacement-safe response

MSRI = Governance × 0.25 + Labor × 0.25 + Financial × 0.2 + Digital × 0.15 + Protection gap × 0.15

Weights reflect the paper's rationale: governance failure and labor vulnerability are preconditions, so they carry the most weight. Partners can explore other weightings; the default stays the reference.

Risk levels

The composite measures how strongly the conditions that let compounds take root, scale, and persist are present. It is not the certainty that a compound exists.

critical70–100Strong enabling conditions across multiple dimensions. Prioritize monitoring and survivor protection.
high50–69Significant enabling conditions. Elevated monitoring and pre-positioned referral contacts.
medium30–49Some enabling conditions. Regular monitoring and baseline documentation.
low0–29Limited enabling conditions currently detected. Background monitoring.

Confidence and uncertainty

Every score is shown with an uncertainty band: a structured acknowledgment of data quality, not a statistical confidence interval. Each dimension's confidence sets its band, and the bands combine by weight.

±5 pts
High confidence
Multiple corroborated sources
±10 pts
Medium confidence
Partially corroborated, or secondary sources
±18 pts
Low confidence
Single or uncorroborated source

Operational rule: a location with two or more low-confidence dimensions is flagged. Its score must not be used for operational decisions without field verification.

When an indicator has no credible source, it's shown as not yet sourced. Missing data is always visible and never filled in silently.

What MSRI must not be used for

Targeting or identifying individuals
The MSRI scores places, not people. It must not be used to identify, track, or profile anyone, including suspected perpetrators.
Enforcement without verification
A high score is not grounds for a raid or arrest. The index is diagnostic, not evidentiary.
Ranking jurisdictions publicly
A public league table could cause diplomatic harm and push risk into unmapped locations.
Criminalizing victims of forced labor
Any enforcement coordination using MSRI data must embed victim identification and non-prosecution protocols.
Replacing field verification
No score, however confident, substitutes for assessment by practitioners with direct access.
Predicting incidents
The MSRI measures structural susceptibility. It does not predict when or whether a specific event will occur.