Methodology
How we score the risk of every fiat currency on earth.
12 scoring factors
Each factor is normalized to 0–100 and weighted. Missing factors redistribute their weight to available ones. All factors are scored at full strength — there is no hidden dampening or special treatment.
01Inflation Rate~13%▶
Annual consumer price inflation. Measures how fast prices are rising and purchasing power is eroding.
02Debt-to-GDP~9%▶
Central government debt as a percentage of GDP. Measures fiscal sustainability and the government's ability to service its obligations.
03FX Volatility~8%▶
Recent exchange rate movement against the US dollar. Captures near-real-time currency stress. For the USD itself, an inflation-based proxy is used since measuring USD against USD is circular.
04Global Currency Role~8%▶
Structural factor measuring the degree to which a currency is embedded in the global financial system — capturing reserve holdings, trade invoicing, payment settlement, FX turnover, and offshore funding. Reserve share alone is insufficient because it misses these other dimensions of international usage. Uses a compressed 10–85 risk scale.
05Capital Controls~7%▶
Score from 0 to 100 measuring how difficult it is to convert or move money out of this currency. Combines formal restrictions (capital account controls, outbound transfer limits) with FX market liquidity depth. Thin FX markets amplify control risk — even mild controls become serious when you can't find a counterparty.
06Reserve Adequacy~6%▶
Foreign exchange reserves measured in months of import cover.
07Current Account~6%▶
Current account balance as a percentage of GDP — the net flow of trade and income with the rest of the world.
08Governance & Institutional Quality~6%▶
Composite measure of institutional quality, rule of law, and regulatory effectiveness.
09GDP Growth~5%▶
Annual real GDP growth rate. Inverted — negative growth increases risk.
10NPL Ratio~5%▶
Bank non-performing loans as a percentage of total loans. A direct measure of banking system health.
11Peg Fragility~5%▶
Score from 0 to 50 measuring the risk of an exchange rate peg breaking suddenly. Only applies to hard pegs and currency boards (e.g. Hong Kong, Bahrain, Bulgaria). Free-floating currencies always score 0.
12Black Market Premium~4%▶
Gap between official and parallel/black market exchange rates. Covers countries with known currency controls or parallel markets.
Tracked but not scored
We track M2 money supply as a percentage of GDP for every country, and it's visible as a column in the rankings table. However, it is not included in the composite risk score.
The reason: M2/GDP is an ambiguous indicator. A high ratio can signal either excessive money printing (risk) or deep financial development (stability). Japan has M2/GDP above 250% — one of the highest in the world — yet its currency is structurally stable. Meanwhile, countries with low M2/GDP ratios like some African economies may simply have underdeveloped banking systems, not conservative monetary policy.
Rather than assigning a weight to a factor that can mean opposite things in different contexts, we display it as contextual information and let the other 12 factors — which have clearer directional signals — drive the score. If you believe money supply growth should be weighted, you can add it via the custom model builder.
Why Global Currency Role matters
Reserve share alone (e.g. IMF COFER allocation percentages) is insufficient to capture a currency's structural resilience. A currency's global embeddedness extends far beyond central bank reserve holdings — it includes trade invoicing, cross-border payment settlement, FX market turnover, and offshore debt issuance. The USD, for example, is used to settle roughly 88% of all FX transactions (BIS) and denominate about half of global trade invoicing — far exceeding its 58% reserve share.
The Global Currency Role factor is intended to capture structural international embeddedness, not short-term market confidence or crisis-period flow dynamics. It captures some structural conditions associated with safe-haven currencies, but does not directly model flight-to-safety behavior during crises. A currency scoring well on this factor has persistent demand that makes it harder to destabilize — not a guarantee of safety.
We use a compressed 10–85 scale rather than 0–100 because the risk difference between a secondary international currency (like the Australian dollar) and a purely domestic one (like the Bangladeshi taka) is real but not as stark as the difference between either and a crisis currency. The compression prevents the factor from overwhelming the model for ordinary currencies.
Shared currencies (EUR, CFA, etc.)
Countries that share a currency — most notably the 20 eurozone members using the EUR — all receive the same Global Currency Role score, the same capital control score, and the same FX market depth assessment. The shared currency provides structural stability: deep FX markets, reserve status, and free capital movement within the bloc.
However, fiscal health is country-specific. Germany and Greece both hold euros, but their debt burdens, banking systems, inflation rates, and governance quality differ substantially. A Greek citizen and a German citizen experience different purchasing power erosion, different fiscal policy risk, and different institutional quality — even though their currency is identical.
This is why we score by country, not by currency. The currency structure provides the floor (shared benefits), but the country-specific factors determine where each nation lands on the risk scale. The same logic applies to CFA franc countries in West and Central Africa, and to dollarized economies like Ecuador and El Salvador.
Data opacity penalty
Countries that don't report economic data to international bodies are almost always hiding bad news. North Korea, Turkmenistan, Eritrea, and similar opaque economies tend to have the worst currency outcomes.
Rather than giving these countries a free pass (showing “No Data”), we apply an opacity penalty: the fewer indicators a country reports, the higher its penalty. A country with zero data gets a baseline risk score of ~65 (High risk) from opacity alone.
Additionally, when less than half of the factors are available, a “confidence nudge” pushes the score upward — reflecting the additional uncertainty of sparse data.