The BIS Triennial Survey, formally the Triennial Central Bank Survey of Foreign Exchange and Over-the-counter (OTC) Derivatives Markets, is a global statistical survey run every three years by central banks and other authorities under the coordination of the Bank for International Settlements (BIS). It is the most complete source of data on the size of the foreign exchange market and is widely quoted by banks, regulators, payment companies and researchers.
The survey has two parts. The turnover part measures trading activity during the month of April in the survey year. The amounts-outstanding part measures the size of open OTC derivatives positions at the end of June. The 2025 edition was the 14th survey; it involved central banks and other authorities in 52 jurisdictions, which collected data from more than 1,100 banks and other dealers. It complements the BIS's half-yearly OTC derivatives statistics, which cover a smaller group of large dealers.
BIS Triennial Survey in plain terms
Think of a national census, but for currency trading. Every few years, officials ask every large shop in the market the same set of questions over the same month: how much did you sell, of what, and to whom? Adding the answers together gives a picture of the whole market that no single shop could provide.
The Triennial Survey does this for the world's currency and derivatives dealers. Central banks collect the answers in their own countries, the BIS puts them together, and the result shows how big the market is, which currencies are traded and which cities do the trading.
How the data are collected
Each participating authority asks its reporting dealers, mostly large banks, to record all trades booked by their sales desks in that location during April. Trades are reported by instrument (spot, outright forwards, FX swaps, currency swaps and options), by currency pair and by type of counterparty: other reporting dealers, other financial institutions, and non-financial customers.
Because a trade between two reporting dealers would otherwise be counted twice, the BIS adjusts the figures for double-counting, both within each country and across countries. Each trade involves two currencies, so currency shares add up to 200% rather than 100%. Results are also published on a locational basis, which assigns each trade to the place where the dealer's sales desk sits rather than to the nationality of the bank.
Key findings of the 2025 survey
Preliminary results were published on 30 September 2025, with fuller data released over the following months and final results in June 2026. The main headline figures were:
- Total size. Global FX trading averaged $9.6 trillion a day in April 2025, up 28% from 2022.
- Instruments. FX swaps remained the largest instrument at about $4 trillion a day; spot turnover rose 42% and outright forwards rose 60%.
- Currencies. The US dollar was on one side of 89% of trades, the euro on 28.9%, the Japanese yen on 16.8% and sterling on 10.2%; trading in the renminbi increased.
- Locations. Sales desks in the United Kingdom, the United States, Singapore and Hong Kong SAR accounted for 75% of FX trading; the UK alone held about 38%, and Singapore rose to 11.8%.
- Interest rate derivatives. OTC interest rate derivatives turnover reached $7.9 trillion a day, up 59%.
National releases add local detail. The Hong Kong Monetary Authority reported average daily FX turnover in Hong Kong of US$883.1 billion in April 2025, up 27.2% from 2022, keeping it the fourth-largest FX centre, with renminbi FX turnover of US$315.1 billion.
Uses of the survey
Central banks and the CPMI use the data to judge market structure, concentration and settlement risk, including how much FX trading still settles without payment-versus-payment protection such as Continuous Linked Settlement (CLS); the 2025 round also collected data on how FX trades are settled. Researchers use the currency shares to track the international role of the dollar and of rising currencies.
In the payments industry the figures are often quoted to show the scale of wholesale FX, the central place of dollar liquidity even where neither end customer uses dollars, and the importance of hubs such as Hong Kong for trade-linked flows, including models like payment backhaul and FX aggregation.
Limitations
The survey is a snapshot of a single month every three years, so seasonal effects, market stress or one-off events in April can affect comparisons. Turnover is measured in US dollars, so movements in exchange rates change the reported size of trading in other currencies. Coverage is limited to reporting dealers, which means some activity by non-bank market makers, retail platforms and informal channels such as a parallel market is not captured. Turnover also measures gross trading volume, not the underlying economic flows such as trade or remittances.