Global Payment Currency Share 2026
By Axis Intelligence Research
Co-author: Sarah Davis | Last updated: August 25, 2026 | License: CC BY 4.0
The US dollar took 50.99% of global payment value on Swift in July 2026, up 3.18 percentage points from July 2024. On the international basis that strips out payments made inside the eurozone, the same dollar moved 0.12 points in two years, from 59.46% to 59.58%. Two numbers, one currency, one month.
Quick Answer
The dollar held 50.99% of global Swift payment value in July 2026, the euro 21.78%, sterling 6.97%, the yen 3.35% and the renminbi 3.10%. Excluding intra-eurozone traffic, the dollar held 59.58% and the euro 13.36%. According to Axis Intelligence Research, the dollar’s widely reported two-year gain of 3.18 percentage points shrinks to 0.12 points on the ex-eurozone basis, meaning almost the entire move is a change in eurozone-internal payment behaviour rather than a shift in international currency choice.
Key Findings
- According to Axis Intelligence Research, the Eurozone Exclusion Effect (EZE) for the US dollar fell from 11.65 percentage points in July 2024 and July 2025 to 8.59 points in July 2026, a 3.06-point compression that accounts for 96% of the dollar’s apparent gain on the global measure.
- According to Axis Intelligence Research, the Axis Currency Breadth Index (CBI) reads 59.28 points for the dollar against 13.88 for the euro, 5.20 for the yen and 3.95 for the renminbi, placing the dollar 4.27 times ahead of the euro across payments, trade finance, foreign exchange turnover and official reserves combined.
- The renminbi ranked second in Swift trade finance at 8.43% of value in July 2026, ahead of the euro at 6.24%, after the two currencies traded the second position twice in three years, per the Swift Global Currency Tracker.
- According to Axis Intelligence Research, the renminbi is the only major currency used more in international payments than it is held in reserves, with a Payment-to-Reserve Balance of 1.18 against 1.04 for the dollar, 0.90 for the yen and 0.67 for the euro.
- The top three currencies took 79.74% of global Swift payment value in July 2026, up from 77.28% two years earlier, while the twenty ranked currencies together covered 98.26% of value.
What Share of Global Payments Is in US Dollars in 2026?
The dollar accounted for 50.99% of global payment value on Swift in July 2026, measured across MT 103, MT 202, pacs.008 and pacs.009 messages, according to the Swift Global Currency Tracker published on 19 August 2026. The euro followed at 21.78% and sterling at 6.97%.
Swift publishes two rankings each month, and they are not two views of the same population. The global table counts every payment message on the network, including payments sent between two institutions inside the eurozone. The international table removes those. Since intra-eurozone traffic is euro-denominated by construction, the two tables answer different questions: one asks what currency the network carries, the other asks what currency the world reaches for when it crosses a border that matters.
Global Payment Currency Share, July 2026
| Rank | Currency | Share of global payment value (%) | Share excluding intra-eurozone (%) | Source |
|---|---|---|---|---|
| 1 | USD | 50.99 | 59.58 | Swift Global Currency Tracker, August 2026 |
| 2 | EUR | 21.78 | 13.36 | Swift Global Currency Tracker, August 2026 |
| 3 | GBP | 6.97 | 5.22 | Swift Global Currency Tracker, August 2026 |
| 4 | JPY | 3.35 | 4.87 | Swift Global Currency Tracker, August 2026 |
| 5 | CNY | 3.10 | 2.34 | Swift Global Currency Tracker, August 2026 |
| 6 | CAD | 2.76 | 2.60 | Swift Global Currency Tracker, August 2026 |
| 7 | HKD | 1.84 | 1.43 | Swift Global Currency Tracker, August 2026 |
| 8 | AUD | 1.58 | 1.88 | Swift Global Currency Tracker, August 2026 |
| 9 | SGD | 1.12 | 0.76 | Swift Global Currency Tracker, August 2026 |
| 10 | CHF | 0.91 | 1.80 | Swift Global Currency Tracker, August 2026 |
| 11 | PLN | 0.73 | 0.70 | Swift Global Currency Tracker, August 2026 |
| 12 | SEK | 0.64 | 0.71 | Swift Global Currency Tracker, August 2026 |
| 13 | NOK | 0.53 | 0.63 | Swift Global Currency Tracker, August 2026 |
| 14 | NZD | 0.33 | 0.41 | Swift Global Currency Tracker, August 2026 |
| 15 | DKK | 0.31 | 0.49 | Swift Global Currency Tracker, August 2026 |
| 16 | MXN | 0.29 | 0.50 | Swift Global Currency Tracker, August 2026 |
| 17 | ZAR | 0.29 | 0.25 | Swift Global Currency Tracker, August 2026 |
| 18 | THB | 0.26 | Not ranked | Swift Global Currency Tracker, August 2026 |
| 19 | HUF | 0.25 | 0.31 | Swift Global Currency Tracker, August 2026 |
| 20 | MYR | 0.23 | Not ranked | Swift Global Currency Tracker, August 2026 |
The ex-eurozone column carries its own top 20, and three currencies appear there that miss the global list: the Czech koruna at 0.41%, the Turkish lira at 0.39% and, in earlier months, the Saudi riyal. The Swiss franc is the clearest illustration of what the basis change does to a single currency. It sits tenth globally at 0.91% and roughly doubles to 1.80% once eurozone-internal payments come out, because a large share of franc traffic is priced against a euro leg that the global table dilutes.
Sarah Davis: The two-column table is the reason so much published commentary on dollar share is arithmetically fine and analytically empty. A treasury desk quoting 50.99% and a policy note quoting 59.58% are both reading Swift correctly and are not describing the same world. If you are pricing a correspondent relationship, the ex-eurozone number is the one that maps to your problem, because intra-eurozone euro payments are not competing for your dollar nostro. If you are modelling Swift’s own message revenue, the global number is the one that matters, because those payments are still traffic. Nobody has to pick a winner between the two series. Everybody has to say which one they used, and most do not.
Which Currencies Gained and Lost Payment Share Since 2024?
Three consecutive August editions of the tracker, each reporting the preceding July, give a clean same-month comparison free of seasonal distortion. Axis Intelligence Research pulled the July 2024, July 2025 and July 2026 rankings from the source PDFs rather than from any intermediate compilation.
Three-July Series, Global Payment Basis
| Currency | July 2024 (%) | July 2025 (%) | July 2026 (%) | Change 2024 to 2026 (pp) | Source |
|---|---|---|---|---|---|
| USD | 47.81 | 47.94 | 50.99 | +3.18 | Swift trackers; change Axis-calculated |
| EUR | 22.47 | 23.11 | 21.78 | -0.69 | Swift trackers; change Axis-calculated |
| GBP | 7.00 | 7.24 | 6.97 | -0.03 | Swift trackers; change Axis-calculated |
| JPY | 3.62 | 3.56 | 3.35 | -0.27 | Swift trackers; change Axis-calculated |
| CNY | 4.74 | 2.88 | 3.10 | -1.64 | Swift trackers; change Axis-calculated |
| CAD | 2.42 | 3.04 | 2.76 | +0.34 | Swift trackers; change Axis-calculated |
| HKD | 1.63 | 2.02 | 1.84 | +0.21 | Swift trackers; change Axis-calculated |
| AUD | 1.80 | 1.57 | 1.58 | -0.22 | Swift trackers; change Axis-calculated |
| SGD | 1.34 | 1.47 | 1.12 | -0.22 | Swift trackers; change Axis-calculated |
| CHF | 0.92 | 0.92 | 0.91 | -0.01 | Swift trackers; change Axis-calculated |
Three-July Series, Excluding Intra-Eurozone Payments
| Currency | July 2024 (%) | July 2025 (%) | July 2026 (%) | Change 2024 to 2026 (pp) | Source |
|---|---|---|---|---|---|
| USD | 59.46 | 59.59 | 59.58 | +0.12 | Swift trackers; change Axis-calculated |
| EUR | 12.80 | 13.40 | 13.36 | +0.56 | Swift trackers; change Axis-calculated |
| GBP | 4.96 | 5.37 | 5.22 | +0.26 | Swift trackers; change Axis-calculated |
| JPY | 5.10 | 5.12 | 4.87 | -0.23 | Swift trackers; change Axis-calculated |
| CNY | 3.32 | 2.11 | 2.34 | -0.98 | Swift trackers; change Axis-calculated |
| CAD | 2.76 | 2.72 | 2.60 | -0.16 | Swift trackers; change Axis-calculated |
| AUD | 2.01 | 1.75 | 1.88 | -0.13 | Swift trackers; change Axis-calculated |
| CHF | 1.72 | 1.73 | 1.80 | +0.08 | Swift trackers; change Axis-calculated |
| HKD | 1.21 | 1.52 | 1.43 | +0.22 | Swift trackers; change Axis-calculated |
| SGD | 0.73 | 0.79 | 0.76 | +0.03 | Swift trackers; change Axis-calculated |
Set the two tables beside each other and the dollar’s story inverts. Globally it gained 3.18 points. Internationally it gained 0.12. The euro lost 0.69 points globally and gained 0.56 internationally. Sterling was flat globally and up 0.26 internationally.
The Axis Eurozone Exclusion Effect (EZE)
EZE stands for Eurozone Exclusion Effect. It measures how much of a currency’s reported Swift payment share is an artefact of whether eurozone-internal payments are counted, expressed as the gap between the two rankings Swift publishes in the same month.
Formula
EZE = International payment share (excluding intra-eurozone) - Global payment share
Inputs: the two currency rankings published side by side in each monthly Swift Global Currency Tracker, both value-based, both covering MT 103, MT 202, pacs.008 and pacs.009 messages.
Worked example, US dollar, July 2026
EZE_USD = 59.58 - 50.99
= 8.59 percentage points
EZE Readings, Three Julys
| Currency | July 2024 (pp) | July 2025 (pp) | July 2026 (pp) | Source |
|---|---|---|---|---|
| USD | +11.65 | +11.65 | +8.59 | Axis Intelligence Research, from Swift Global Currency Tracker |
| EUR | -9.67 | -9.71 | -8.42 | Axis Intelligence Research, from Swift Global Currency Tracker |
| GBP | -2.04 | -1.87 | -1.75 | Axis Intelligence Research, from Swift Global Currency Tracker |
| JPY | +1.48 | +1.56 | +1.52 | Axis Intelligence Research, from Swift Global Currency Tracker |
| CNY | -1.42 | -0.77 | -0.76 | Axis Intelligence Research, from Swift Global Currency Tracker |
| CHF | +0.80 | +0.81 | +0.89 | Axis Intelligence Research, from Swift Global Currency Tracker |
| CAD | +0.34 | -0.32 | -0.16 | Axis Intelligence Research, from Swift Global Currency Tracker |
| AUD | +0.21 | +0.18 | +0.30 | Axis Intelligence Research, from Swift Global Currency Tracker |
| SGD | -0.61 | -0.68 | -0.36 | Axis Intelligence Research, from Swift Global Currency Tracker |
| HKD | -0.42 | -0.50 | -0.41 | Axis Intelligence Research, from Swift Global Currency Tracker |
Interpretation. A large positive EZE means a currency’s global share is being held down by the weight of euro traffic inside the eurozone, and that its true international standing is understated by the headline number. A large negative EZE means the reverse. The dollar’s reading sat at exactly 11.65 points in both July 2024 and July 2025, then dropped to 8.59 in July 2026. According to Axis Intelligence Research, that 3.06-point compression explains 96% of the dollar’s 3.18-point gain on the global measure, because the gain came from intra-eurozone payments falling as a proportion of network traffic rather than from anyone outside the eurozone switching into dollars.
Scope. EZE is computed on value, not message counts, and it moves with any change in the relative volume of eurozone-internal traffic on Swift, including migration of that traffic to domestic and regional rails. It is a decomposition diagnostic, not a measure of currency preference on its own.
Sarah Davis: The de-dollarization debate is fought almost entirely on the global column, and the global column is the one contaminated by a denominator effect. Watch what a European bank actually does. Euro payments between two eurozone institutions increasingly clear on TARGET services and on instant rails, and every euro of that migration mechanically lifts the dollar’s percentage of what remains on Swift without a single treasurer changing a settlement currency. The ex-eurozone series is the cleaner read on behaviour and it has been sitting between 59.46% and 59.59% for three Julys running. A 13-basis-point range over two years is not a trend. It is a flat line with a rounding error attached. Anyone building a de-dollarization thesis needs to explain why the series designed to isolate international choice refuses to move.
Why Do Dollar Share Figures Differ So Much Between Sources?
Because they measure different markets. A currency’s share of payment messages, of trade finance instruments, of foreign exchange turnover and of central bank reserves are four separate populations with four separate methodologies, and the dollar’s reading ranges from 44.6% to 79.86% depending on which one is quoted.
Currency Share Across Four Markets
| Currency | Swift international payments (%) | Swift trade finance (%) | BIS FX turnover (%, 200% basis) | IMF official reserves (%) | Source |
|---|---|---|---|---|---|
| USD | 59.58 | 79.86 | 89.2 | 57.13 | Swift, Jul 2026; BIS, Apr 2025; IMF COFER, 2026Q1 |
| EUR | 13.36 | 6.24 | 28.9 | 20.03 | Swift, Jul 2026; BIS, Apr 2025; IMF COFER, 2026Q1 |
| JPY | 4.87 | 1.40 | 16.8 | 5.44 | Swift, Jul 2026; BIS, Apr 2025; IMF COFER, 2026Q1 |
| CNY | 2.34 | 8.43 | 8.5 | 1.99 | Swift, Jul 2026; BIS, Apr 2025; IMF COFER, 2026Q1 |
| GBP | 5.22 | 0.13 | 10.2 | Below 5, not separately published in the source used | Swift, Jul 2026; BIS, Apr 2025; ECB, Jun 2026 |
Foreign exchange turnover shares sum to 200% because every trade involves two currencies. The BIS Triennial Central Bank Survey recorded 9.6 trillion dollars of daily OTC foreign exchange turnover in April 2025, up 28% on 2022, with the dollar on one side of 89.2% of trades. Official reserve shares come from the IMF COFER release of 1 July 2026, covering 13.10 trillion dollars of reserves.
The Axis Currency Breadth Index (CBI)
CBI stands for Currency Breadth Index. It compresses those four markets into a single ranked reading, so a currency can be compared on how broadly it is used rather than on whichever single market flatters it most.
Formula
CBI = 0.30 x Swift international payment share
+ 0.20 x Swift trade finance share
+ 0.25 x (BIS FX turnover share / 2)
+ 0.25 x IMF COFER official reserve share
Weights and why they sit where they do. Payments carry the largest weight at 0.30 because payment messaging is the only monthly, high-frequency observation in the set. Reserves and foreign exchange turnover take 0.25 each as the two stock-and-flow anchors of international currency status. Trade finance takes 0.20 because it is the smallest population by message count and the most concentrated by commodity. The foreign exchange input is halved to rebase the survey’s 200% convention onto a 100% scale before weighting; without that step the turnover lens would swamp the other three.
Worked example, renminbi, July 2026
CBI_CNY = 0.30 x 2.34 = 0.702
+ 0.20 x 8.43 = 1.686
+ 0.25 x 4.25 = 1.063
+ 0.25 x 1.99 = 0.498
= 3.95 index points
CBI Readings, July 2026
| Currency | CBI (points) | Global payment share (%) | Gap, CBI minus payment share | Widest lens | Narrowest lens | Source |
|---|---|---|---|---|---|---|
| USD | 59.28 | 50.99 | +8.29 | Trade finance, 79.86 | FX turnover rebased, 44.60 | Axis Intelligence Research |
| EUR | 13.88 | 21.78 | -7.90 | Reserves, 20.03 | Trade finance, 6.24 | Axis Intelligence Research |
| JPY | 5.20 | 3.35 | +1.85 | FX turnover rebased, 8.40 | Trade finance, 1.40 | Axis Intelligence Research |
| CNY | 3.95 | 3.10 | +0.85 | Trade finance, 8.43 | Reserves, 1.99 | Axis Intelligence Research |
Interpretation. The gap column is where the index earns its keep. The dollar reads 8.29 points higher on CBI than on the payment measure alone, so the single most quoted statistic in this field understates the dollar. The euro reads 7.90 points lower, so the same statistic overstates the euro by almost exactly the amount it understates the dollar, which is the eurozone denominator showing up a second time. According to Axis Intelligence Research, the dollar takes 72.02% of the combined CBI of these four currencies while taking 50.99% of global payment value, and the difference between those two figures is the difference between a payments question and a monetary-system question.
Scope. CBI mixes observation dates: payments are monthly, reserves quarterly, the foreign exchange survey triennial. It covers the four currencies for which all four lenses are separately published by primary sources; sterling is excluded because the reserve share is not broken out in the release used here. The weights are a judgement, they are disclosed in full, and any reader who prefers a different set can recompute the reading from the accompanying dataset.
Which Currency Ranks Second in Trade Finance?
The renminbi, at 8.43% of Swift trade finance value in July 2026, against 6.24% for the euro. That ranking has changed hands twice in three years.
Trade Finance Currency Share, Three Julys
| Currency | July 2024 (%) | July 2025 (%) | July 2026 (%) | Source |
|---|---|---|---|---|
| USD | 83.22 | 79.62 | 79.86 | Swift Global Currency Tracker |
| CNY | 6.00 | 7.02 | 8.43 | Swift Global Currency Tracker |
| EUR | 5.83 | 7.70 | 6.24 | Swift Global Currency Tracker |
| JPY | 1.46 | 1.56 | 1.40 | Swift Global Currency Tracker |
| AED | 0.46 | 0.60 | 0.76 | Swift Global Currency Tracker |
| SAR | 0.82 | 0.72 | 0.75 | Swift Global Currency Tracker |
| THB | 0.48 | 0.75 | 0.65 | Swift Global Currency Tracker |
| VND | 0.19 | 0.29 | 0.56 | Swift Global Currency Tracker |
| IDR | 0.47 | 0.44 | 0.39 | Swift Global Currency Tracker |
| GBP | Not ranked | 0.17 | 0.13 | Swift Global Currency Tracker |
The renminbi ranked second in July 2024, slipped behind the euro in July 2025, and took the position back in July 2026 at its highest reading of the three. Its 2.43-point gain over the period is the largest move by any currency in the trade finance table. Sterling’s 0.13% against a 6.97% global payment share is the same phenomenon photographed from the other side.
Sarah Davis: Trade finance is where currency internationalisation shows up first, and it shows up in a form that is hard to fake. A letter of credit commits a bank to a currency for the life of the instrument, usually months, on the back of a commercial invoice that was priced in that currency long before anyone sent a payment message. The renminbi at 8.43% of trade finance against 2.34% of international payments is a ratio of 3.6, and that ratio is the interesting object, not either number alone. It says Chinese counterparties are increasingly writing the invoice in renminbi and then, in a large share of cases, settling the resulting exposure somewhere Swift does not observe. The euro runs the opposite way at 0.47, invoicing less than it pays. Europe’s currency is a settlement currency for European business. China’s is becoming a pricing currency for Asian and Gulf trade. Those are different jobs and they show up in different message types.
Is the Renminbi Gaining Ground in Global Payments?
Not on the payments measure, and clearly on several others. The renminbi held 3.10% of global payment value in July 2026, below the 4.74% it reached in July 2024, and 2.34% on the ex-eurozone basis against 3.32% two years earlier.
Renminbi Position Across Swift Measures
| Measure | July 2024 | July 2025 | July 2026 | Source |
|---|---|---|---|---|
| Global payments rank | 4th | 6th | 5th | Swift Global Currency Tracker |
| Global payments share | 4.74% | 2.88% | 3.10% | Swift Global Currency Tracker |
| Ex-eurozone payments rank | 5th | 6th | 6th | Swift Global Currency Tracker |
| Trade finance rank | 2nd | 3rd | 2nd | Swift Global Currency Tracker |
| Foreign exchange spot rank | 5th | 6th | 5th | Swift Global Currency Tracker |
| Hong Kong share of offshore renminbi payments | 82.84% | 75.56% | 75.71% | Swift Global Currency Tracker |
| United Kingdom share of offshore renminbi payments | 4.60% | 6.16% | 6.89% | Swift Global Currency Tracker |
Away from Swift, the renminbi’s foreign exchange turnover share reached 8.5% in the April 2025 BIS survey, its highest on record and a continuation of the rise the BIS has tracked since 2013. Its official reserve share was 1.99% in the first quarter of 2026, up from 1.95% at the end of 2025.
The Axis Payment-to-Reserve Balance (PRB)
PRB stands for Payment-to-Reserve Balance. It divides a currency’s international payment share by its official reserve share, separating currencies that the world uses from currencies that the world holds.
Formula
PRB = Swift international payment share / IMF COFER official reserve share
Worked example, renminbi
PRB_CNY = 2.34 / 1.99 = 1.18
| Currency | International payment share (%) | Official reserve share (%) | PRB | Source |
|---|---|---|---|---|
| CNY | 2.34 | 1.99 | 1.18 | Axis Intelligence Research, from Swift and IMF COFER |
| USD | 59.58 | 57.13 | 1.04 | Axis Intelligence Research, from Swift and IMF COFER |
| JPY | 4.87 | 5.44 | 0.90 | Axis Intelligence Research, from Swift and IMF COFER |
| EUR | 13.36 | 20.03 | 0.67 | Axis Intelligence Research, from Swift and IMF COFER |
Interpretation. A PRB above 1.0 marks a currency doing more transactional work than its reserve standing implies. Below 1.0 marks a store of value that circulates less than it is held. According to Axis Intelligence Research, the renminbi’s 1.18 is the highest of the four and the euro’s 0.67 the lowest, which is the inverse of how the two currencies are usually positioned in the de-dollarization conversation.
Sarah Davis: The renminbi is the most transactional and the least reserved of the major currencies, and that combination is exactly what you would expect from a currency whose use is being built out through trade corridors rather than through central bank portfolios. The ECB’s June 2026 review put the point plainly, noting that the renminbi’s role as an official store of value is limited at present, and reserve managers left their renminbi positions unchanged through 2025. Meanwhile the same review records CIPS settlement rising by more than a third in the days around the outbreak of the war in the Middle East, and President Xi calling on 1 February 2026 for the renminbi to become a global reserve currency. Read those together and the sequencing is legible: build the corridor, price the invoice, clear the trade, and worry about reserve managers last. The measure to watch is not the payments column. It is whether the trade finance share and the reserve share start moving in the same direction, because until they do, the renminbi is a settlement option rather than a monetary alternative. Our ISO 20022 statistics breakdown covers the message-standard layer that makes those corridors comparable in the first place.
What Do Reserve and Foreign Exchange Data Say About Dollar Dominance?
The dollar held 57.13% of global official foreign exchange reserves in the first quarter of 2026, up from 56.42% at the end of 2025, with the IMF attributing around half of the increase to the dollar’s appreciation against other reserve currencies rather than to reserve manager purchases. The euro fell to 20.03% from 20.38%, the yen to 5.44% from 5.84%, and the renminbi rose to 1.99% from 1.95%.
Official Reserve Currency Composition
| Currency | 2025Q4 (%) | 2026Q1 (%) | Change (pp) | Source |
|---|---|---|---|---|
| USD | 56.42 | 57.13 | +0.71 | IMF COFER, July 2026 release |
| EUR | 20.38 | 20.03 | -0.35 | IMF COFER, July 2026 release |
| JPY | 5.84 | 5.44 | -0.40 | IMF COFER, July 2026 release |
| CNY | 1.95 | 1.99 | +0.04 | IMF COFER, July 2026 release |
| Other currencies | 6.25 | 6.18 | -0.07 | IMF COFER, July 2026 release |
Total foreign exchange reserves stood at 13.10 trillion dollars, slightly below the 13.15 trillion recorded at the end of 2025. The “other currencies” line, which the IMF does not break out, fell for the first time after seven consecutive quarterly increases beginning in the first quarter of 2024.
Gold complicates the picture in a way currency shares alone cannot show. The ECB puts gold at 27% of total official foreign reserves at the end of 2025, ahead of both the euro at 15% and US Treasuries at 22%. Correct for the price surge by holding gold at its end-2023 valuation and gold falls to 16%, level with the euro at 16%, while Treasuries return to 26%. Central bank gold purchases eased to around 850 tonnes in 2025 from more than 1,000 tonnes annually between 2022 and 2024.
Sarah Davis: There is one honest disagreement in this dataset worth stating rather than smoothing. The ECB reports the euro at 28.5% of global foreign exchange transactions in the April 2025 BIS survey; the BIS preliminary release of 30 September 2025 puts it at 28.9%. Axis Intelligence Research publishes both as reported and does not average them. The gap is 0.4 points and it almost certainly reflects the December 2025 final revision against the September preliminary, but we have not seen a document that says so, and inventing a reconciliation is worse than showing the seam. For anyone building a model off these numbers, the practical instruction is to pick a vintage and label it, because the reserve series was itself restated in the third quarter of 2025 when the IMF eliminated the unallocated category and imputed the full composition, a change that moved the euro’s historical share down by roughly 0.8 points in mid-2025. Comparing a pre-revision figure with a post-revision figure produces a trend that exists only in the methodology. Domestic rails sit outside all of this, and our wire transfer statistics work shows what happens to currency questions when the payment never crosses a border.
Methodology
Collection. Every figure in this dataset was retrieved from a primary source document during the production session on 25 August 2026. Five source documents were used: the Swift Global Currency Tracker August 2026 edition reporting July 2026 data; the Swift RMB Tracker August 2025 edition reporting July 2025; the Swift RMB Tracker August 2024 edition reporting July 2024; the BIS Triennial Central Bank Survey statistical release on OTC foreign exchange turnover in April 2025; and the IMF COFER data brief of 1 July 2026 covering the first quarter of 2026. The ECB’s June 2026 review of the international role of the euro supplies the euro-specific indicators, the gold shares and the reserve methodology note. No figure derives from secondary aggregation, and every external link in this article was fetched and confirmed live before publication.
Swift data basis. The tracker reflects statistics computed from MT 103, MT 202, pacs.008 and pacs.009 messages for payments, MT 300 for foreign exchange, and MT 400 and MT 700 for trade finance, exchanged between financial institutions on Swift. Currency shares are value-based. Swift states that because financial institutions have multiple means of exchanging transaction information, the tracker does not represent complete market or industry statistics, and that shares can move for reasons unrelated to currency usage, such as changes in message-type adoption. The RMB Tracker was renamed the Global Currency Tracker in February 2026; the underlying series is continuous.
Same-month comparison. All three years are read from August editions reporting the preceding July, which removes month-of-year effects from the comparison. Where a currency appears in one year’s ranked 20 and not another’s, the table records it as not ranked rather than as zero.
Formula: Eurozone Exclusion Effect (EZE).
EZE = International payment share (excluding intra-eurozone) - Global payment share
Both inputs are taken from the same monthly tracker edition, so no cross-vintage comparison is involved.
Formula: Currency Breadth Index (CBI).
CBI = 0.30 x Swift international payment share
+ 0.20 x Swift trade finance share
+ 0.25 x (BIS FX turnover share / 2)
+ 0.25 x IMF COFER official reserve share
Weights sum to 1.00. The foreign exchange input is divided by two to rebase the survey’s 200% two-sided convention onto a 100% scale. Inputs are July 2026 for the Swift measures, April 2025 for the BIS measure and the first quarter of 2026 for the IMF measure. Readings are computed for the four currencies with all four lenses separately published.
Formula: Payment-to-Reserve Balance (PRB).
PRB = Swift international payment share / IMF COFER official reserve share
Arithmetic. All derived figures were computed in Python from the source values and are reproducible from the accompanying CSV. Every calculated row carries a method note identifying its formula, and every prose number in this article exists as a CSV row at the same value and as-of date.
Disclosed source discrepancy. The ECB reports the euro’s share of global foreign exchange transactions in the April 2025 BIS survey as 28.5%, while the BIS preliminary statistical release reports 28.9%. Both readings are published in this dataset as reported by their sources, with no reconciliation applied. The CBI uses the BIS figure, since the other turnover inputs come from the same BIS release.
What this reading does not capture. Swift is a messaging network, not a settlement system. Payments cleared bilaterally, on domestic real-time gross settlement systems, on correspondent books, through CIPS or over closed networks do not appear in these shares, and message value is not the same as settled value. Reserve shares reflect voluntary reporting by 147 monetary authorities with imputation for the remainder. The foreign exchange survey is a single month conducted every three years, and April 2025 coincided with unusually elevated volatility.
About This Dataset
This dataset contains 403 observations covering currency shares of Swift global payments, Swift international payments excluding intra-eurozone traffic, Swift trade finance, Swift foreign exchange spot rankings, offshore renminbi payment weights, BIS foreign exchange turnover, IMF official reserve composition and ECB euro internationalisation indicators, for July 2024, July 2025 and July 2026 with reserve and turnover readings dated to their own release cycles, at global and selected national coverage.
License: CC BY 4.0
Citation line: Axis Intelligence Research, Global Payment Currency Share 2026, 2026
How to Cite This Research
APA Axis Intelligence Research. (2026). Global payment currency share 2026: US dollar 50.99%, euro 21.78%, and the ex-eurozone reading that reverses the de-dollarization story. Axis Intelligence. https://axis-intelligence.com/global-payment-currency-share/
MLA Axis Intelligence Research. “Global Payment Currency Share 2026: US Dollar 50.99%, Euro 21.78%, and the Ex-Eurozone Reading That Reverses the De-Dollarization Story.” Axis Intelligence, 2026, axis-intelligence.com/global-payment-currency-share/.
Chicago Axis Intelligence Research. “Global Payment Currency Share 2026: US Dollar 50.99%, Euro 21.78%, and the Ex-Eurozone Reading That Reverses the De-Dollarization Story.” Axis Intelligence. 2026. https://axis-intelligence.com/global-payment-currency-share/.
Frequently Asked Questions
Which Swift currency share figure should a treasury team use for corridor pricing?
The international series excluding intra-eurozone payments. A corridor decision is a question about which currency counterparties outside a single monetary union reach for, and intra-eurozone euro traffic is euro-denominated by construction, so including it answers a different question. The practical consequence is large: the same July 2026 month yields a dollar share of 50.99% on the global basis and 59.58% on the international basis, an 8.59-point gap that is entirely definitional.
Does a rising dollar share on Swift mean the world is choosing dollars more often?
Not by itself. The dollar’s global share rose 3.18 percentage points between July 2024 and July 2026 while its ex-eurozone share rose 0.12 points. The gap between those two movements is a denominator effect: as eurozone-internal euro payments fall as a proportion of Swift traffic, every other currency’s percentage rises without anyone changing behaviour. A genuine shift in currency choice would show up in both series.
Why does the renminbi rank higher in trade finance than in payments?
Because trade finance instruments follow the invoice currency and payments follow the settlement currency, and those diverge when a counterparty prices in one currency and settles in another. The renminbi held 8.43% of Swift trade finance value in July 2026 against 2.34% of international payment value, a ratio of 3.6. Letters of credit and guarantees also commit to a currency months earlier than the corresponding payment message, which makes the trade finance series a leading indicator of invoicing behaviour.
How can a currency’s share be 89.2% in one dataset and 57.13% in another?
They count different markets on different conventions. The 89.2% figure is the dollar’s share of BIS-surveyed foreign exchange turnover, where shares sum to 200% because each trade has two currency legs, so the comparable one-sided figure is 44.6%. The 57.13% figure is the dollar’s share of official reserves reported to the IMF. Neither is wrong, and quoting one against the other without naming the basis produces a comparison with no meaning.
What changed in the IMF reserve series in 2025 and does it break historical comparisons?
Starting in the third quarter of 2025, with revisions back to the first quarter of 2000, the IMF eliminated the unallocated category from COFER and imputed the currency composition of those reserves so that shares sum to 100% of world reserves. The revision lifted the “other currencies” category materially and, per the ECB, moved the euro’s share down by roughly 0.8 percentage points in mid-2025. Comparing a figure published before the change with one published after it will produce an apparent trend that is a methodology artefact.
Is CIPS displacing Swift as the venue for renminbi payments?
The available evidence points to growth alongside Swift rather than substitution within it. The ECB records CIPS settlement activity rising by more than a third in the days surrounding the outbreak of the war in the Middle East, and reports of vessels paying in renminbi via CIPS to transit the Strait of Hormuz in March and April 2026. Since CIPS clearing does not appear in Swift’s currency shares, renminbi activity moving to that route would depress the Swift reading while renminbi usage rose, which is a reason to read the Swift payments series as a floor rather than a total.
Why did Hong Kong’s share of offshore renminbi payments fall?
Hong Kong accounted for 82.84% of offshore renminbi payment value in July 2024, 75.56% in July 2025 and 75.71% in July 2026, a net decline of 7.13 percentage points concentrated in a single year. The United Kingdom moved the other way over the same period, from 4.60% to 6.89%. The pattern is a broadening of the offshore renminbi base rather than a contraction of Hong Kong’s absolute activity, since these are shares of a growing pool and Hong Kong remains the dominant clearing centre by a wide margin.
Does gold’s share of official reserves signal reserve managers moving away from currencies?
Mostly it signals the gold price. Gold reached 27% of total official foreign reserves at the end of 2025, above both the euro at 15% and US Treasuries at 22%, but holding gold at its end-2023 valuation puts gold at 16%, level with the euro, and Treasuries at 26%. Central bank purchases also eased to around 850 tonnes in 2025 from more than 1,000 tonnes annually in the preceding three years. The allocation shift is real but far smaller than the market-price headline implies.
What would actually constitute evidence of de-dollarization in this data?
Three series moving in the same direction at once: a falling dollar share on the ex-eurozone payments measure, a falling dollar share in Swift trade finance, and a falling COFER reserve share adjusted for exchange rate valuation effects. As of the readings here, the first has been flat within 0.13 points across three Julys, the second stands at 79.86%, and the third rose to 57.13% with the IMF attributing about half the quarterly increase to valuation. Any two of the three moving without the others is a composition effect until proven otherwise.
How often is the underlying source data updated?
Swift publishes the Global Currency Tracker monthly, generally in the third week of the month, reporting the preceding month’s data. The IMF publishes COFER quarterly with roughly a three-month lag. The ECB publishes its review of the international role of the euro annually in June. The BIS runs its Triennial Survey every three years, most recently in April 2025, with final turnover data released the following December.
Related research: SWIFT payment statistics | ISO 20022 statistics | Cross-border payments statistics | Wire transfer statistics | ACH payments statistics