Stablecoins Settle the Payment Leg. Most of Merchandise Trade Isn't the Payment Leg.
There are two questions hiding inside the phrase "stablecoins for trade," and they have different answers.
One is whether a stablecoin can move money from a buyer in one country to a seller in another faster and cheaper than a correspondent banking chain. The answer there is a qualified yes, with the qualifications concentrated at the edges where fiat converts. The other is whether a stablecoin can do what trade finance does: extend credit before the goods ship, guarantee payment against compliant documents, allocate risk between two parties who have never met. The answer there is no, and not for reasons that better technology fixes.
A WTO Secretariat report published in September 2026 is the first attempt to separate those two questions properly and follow each to its conclusion. What emerges is a split that runs along a line most stablecoin commentary ignores entirely: the goods and services divide. We've focused here on that distinction, on the regulatory stacking problem the report identifies, and on where the actual flows are going, rather than on the report's general survey of stablecoin mechanics.
The biggest number in trade finance is the one stablecoins can't touch
The scale of the financing problem is well measured. The Asian Development Bank's ninth Global Trade Finance Gap Survey, released on 15 January 2026 and reported by Reuters, put unmet demand for trade finance at $2.5 trillion in 2025, unchanged from 2023 and equal to roughly 10% of global trade, down slightly from 10.6%. Eighty per cent of the banks surveyed expect demand to rise further as supply chains reconfigure.
Set that against what stablecoins actually moved. The WTO report, citing recent industry analysis, finds that of the roughly $35 trillion in headline stablecoin transaction volume, genuine end-user payments amount to about $390 billion a year, or around 0.02% of global payment volumes. That figure more than doubled from 2024, so the direction is real. But the largest quantified problem in trade finance is a credit problem, and credit is the one function the report is unambiguous that stablecoins do not perform.
The report sets out the functions trade finance actually delivers: payment and settlement, credit provision, risk mitigation, documentation, and liquidity. Stablecoins contribute to the first. On credit, the report's entry is simply none. They do not finance production or shipment, cannot substitute for a legally enforceable payment undertaking such as a letter of credit, and do not address the commercial and documentary risks that arise earlier in the transaction.
Where finance is absent, trade is already conducted on payment terms
Here is the inversion that makes the report worth reading closely, and it is left slightly implicit.
The WTO-IFC regional studies the report draws on find that in surveyed emerging markets, including large traders such as Viet Nam and Mexico, only 2 to 20% of trade flows are supported by trade finance at all. In Mexico, only a quarter of importers and exporters have access to any credit, let alone trade finance. The majority of trade and traders are, in the report's word, unserved. Those firms self-finance from working capital, borrow informally from family, or transact cash in advance or on open account.
Cash in advance and open account are pure payment arrangements. No documentary guarantee, no bank undertaking, no collateral. Which means that for the majority of traders in exactly the economies where the inclusion argument is loudest, the financial requirement already is the payment, because the finance layer was never available to them. Stablecoins are relevant to that population not because they replace trade finance but because trade finance was never there.
That does not resolve the underlying problem. A firm paying cash in advance still bears the risk the goods never arrive, and a faster payment rail does nothing about that. But it does explain why the payments case and the trade finance case keep getting conflated, and why the conflation is understandable rather than merely sloppy.
Services trade is where the argument is strongest, and nobody frames it that way
The report's cleanest analytical move is to observe that letters of credit and documentary collections exist because a physical good is shipped and can be pledged, inspected or held as collateral while payment and delivery are reconciled. Digitally delivered services have no equivalent. There is no shipped good against which a bank can lend or issue a guarantee.
So for services trade, the entire financial requirement collapses into the payment itself. A freelancer, a small software firm, a business process outsourcing provider, or an independent creator exporting across borders does not need a letter of credit. They need payment that is cheap, fast and reliable, often in small and irregular amounts, which is the friction profile stablecoins are best matched to.
This is a more precise claim than the usual one, and it cuts in an unexpected direction. It suggests stablecoins are more relevant to developing-economy services exports than to their merchandise exports, which is close to the opposite of where most "stablecoins for trade" enthusiasm points. Merchandise trade is where the headline volumes are. Services trade is where the instrument actually fits.
The flows are not going where the inclusion story says
The geography in the report is worth stating plainly, because it sits awkwardly against the prevailing narrative.
Asia accounts for roughly $245 billion of stablecoin payment flows, around 60% of the global total, driven predominantly by a small number of financial and digital asset hubs, particularly Hong Kong, China; Japan and Singapore. North America follows at $95 billion and Europe at $50 billion. Latin America and Africa each account for less than $1 billion.
Those last two figures deserve to sit on their own. The regions that anchor nearly every financial inclusion pitch for stablecoins are, by payment volume, rounding errors. Business-to-business payments dominate what activity exists, at roughly $226 billion or 60% of stablecoin payment volume, having grown 733% year on year in 2025. This is corporate treasury and B2B settlement running through well-banked hubs, which is a real and fast-growing use case and not the one the inclusion argument describes. It is also the shape we found when the US Faster Payments Council argued the strongest B2B case is the one where the stablecoin is invisible to both counterparties.
The cost picture is similarly split. One study cited in the report estimates that sending $500 via stablecoins costs $5 to $15 against $20 to $30 through traditional routes. Another, examining a single stablecoin across a selection of corridors, found no systematic cost advantage at all, with total costs ranging from 0.3% to nearly 9% of the amount transferred, driven primarily by the fiat conversion phases rather than the on-chain transfer. That second study is the one we looked at when a central bank sent 200 USDC around the world and found the blockchain was the cheapest leg.
Compliance stacks, it does not harmonise
The report's most operationally useful finding concerns what happens when a single trade transaction touches several jurisdictions at once.
A stablecoin-settled trade must simultaneously satisfy the requirements of the exporter's jurisdiction, the importer's, the issuer's, the blockchain service provider's, the wallet provider's, and those of any intermediary. The report is explicit that this is not resolved by meeting the most permissive common standard. Certain requirements are jointly binding, which eliminates transaction architectures that would be perfectly viable under each framework taken individually.
The United Arab Emirates supplies the clearest example, and it rewards being stated precisely. Under the Central Bank of the UAE's Payment Token Services Regulation, a merchant may accept only a dirham-denominated payment token issued by a CBUAE-licensed issuer, or a foreign-currency payment token from a CBUAE-registered foreign issuer where that token is being used to buy a virtual asset or a virtual asset derivative. Settling an ordinary commercial invoice in USDT or USDC falls outside both permissions. A UAE importer therefore cannot lawfully settle a trade payable in either token, however impeccably the exporting counterparty is regulated at the other end.
A clarification we would add to the report's wording. The report describes this restriction as applying "on the mainland." That is accurate in UAE usage but invites a misreading, so it is worth spelling out. Mainland does not mean "outside Dubai and Abu Dhabi." Dubai and Abu Dhabi are mainland. The only geographic carve-out is the two financial free zones that sit inside them, the Dubai International Financial Centre and the Abu Dhabi Global Market, whose firms are supervised by the DFSA and FSRA rather than the central bank. The regulation excludes those financial free zones from its references to the UAE. Commercial free zones are not excluded, so an entity registered in DMCC or JAFZA sits inside the restriction rather than outside it. The test is where an entity is incorporated and which regulator supervises it, not where an office or a vessel happens to be, and moving into a financial free zone changes which rulebook applies rather than automatically permitting stablecoin settlement. Two further details: legal commentary indicates the regulation's transition period has now lapsed, so the restriction is live rather than prospective, and the central bank registered its first foreign payment token, USDU from Universal Digital, only in January 2026. Anyone structuring a transaction on this should confirm the current position with UAE counsel, since the regime is young and still being built out.
Brazil and the EU show the same stacking, and both also reward precision.
Brazil built its position in two steps. Resolutions 519, 520 and 521, published in November 2025 and effective 2 February 2026, brought crypto service providers inside the regulated perimeter, and Resolution 521 classified the purchase, sale, exchange and cross-border transfer of fiat-pegged stablecoins as foreign exchange operations, creating a licensed provider category for firms doing it. Resolution 561 followed on 30 April 2026 and takes effect on 1 October 2026. It bars regulated electronic foreign exchange providers, meaning payment institutions, e-money issuers and acquirers, from settling the offshore leg of a cross-border payment in stablecoins or any other crypto asset. That settlement must run through a conventional foreign exchange transaction or a non-resident real account.
In the EU, a non-euro e-money token used as a means of exchange faces a threshold of 1 million transactions per day or EUR 200 million in daily transaction value, whichever is reached first. Crossing it obliges the issuer to stop issuing new tokens until usage falls back below the line.
Two clarifications on these. First, Resolution 561 is narrower than a Brazilian ban on stablecoin settlement and broader than the report's line suggests. It closes one specific route, the eFX rail used by payment institutions, while licensed virtual asset service providers operating under Resolution 521 can still use stablecoins for cross-border payments. It is also not yet in force as we publish, since it applies from 1 October 2026, so a reader taking the report's table as a settled description of Brazil would be reading a position that changes in a matter of weeks. Second, the EU figure is a threshold, not a transaction cap. It does not stop users transacting once daily volume passes EUR 200 million; it stops the issuer minting new tokens until volumes fall. It has a second limb at 1 million transactions per day, it applies only to tokens denominated in a non-EU currency rather than to euro-denominated ones, and under the EBA's technical standards it counts only payments where both payer and payee sit in the same single currency area, not all transactions in the token. Describing it as a cap on non-euro stablecoin transactions, as summary tables often do, overstates its reach in one direction and understates it in another.
Each of these rules is defensible domestically. Together they remove combinations, and none of them can be read off a one-line table entry. That gap between the summary and the operative rule is why we maintain the free Stablecoin Regulation Tracker across 200+ markets, built from legislation and regulator guidance rather than secondary summaries.
Redemption rights fragment the same way. In advanced-economy frameworks, holders get an unconditional right to redeem on demand at par, free of discretionary fees, enforceable against a licensed issuer. In several developing-economy frameworks, redemption is contractual, undefined, or not yet in force. A buyer in such an economy holding a foreign-issued stablecoin to settle a payable has no locally enforceable redemption right, and the issuer owes no statutory par-value obligation under those rules. Only five of roughly 29 jurisdictions surveyed by the FSB had finalised a stablecoin framework as of 2025.
Then comes the perverse result. Because a compliant institution on one side of a transaction can absorb the compliance exposure of a regulatory gap on the other, the report notes that advanced economies with robust frameworks may simply avoid transacting with developing economies whose frameworks are underdeveloped. The asymmetry undermines the commercial case for engagement precisely where the payment friction is worst.
A trade-law lever nobody is discussing
The report contains an argument that has had almost no airing elsewhere: that the General Agreement on Trade in Services already reaches much of the stablecoin stack.
Stablecoins depend on issuers, payment providers, custodians, exchanges, wallet providers, clearing and settlement entities, and data processing firms. Where those activities are supplied internationally, they fall within GATS scope and within the specific commitments WTO members have already made. And the commitments exist in volume: between half and two-thirds of members have made commercial presence commitments in the relevant service categories, and roughly a third to a half have made cross-border supply commitments, albeit with limitations.
The WTO does not regulate stablecoins and says so. But market access and non-discrimination disciplines on payment and money transmission services, custody, settlement and clearing, and financial data processing are a different lever from prudential standard-setting, and they are already in force. The prudential carve-out in the GATS financial services annex preserves the right to regulate for stability. What it does not obviously preserve is the right to erect barriers to foreign stablecoin service suppliers that go beyond prudential purpose.
What this means for each group
Banks and payment providers. The report's functional table is effectively a scoping document. Payment and settlement is contestable. Credit provision, documentary examination and legally enforceable payment undertakings are not, at least under current legal infrastructure. Institutions can cede the first without exposure to the others.
Corporate treasurers. Jurisdictional stacking means a stablecoin arrangement that works in one corridor may be unusable in another for reasons unrelated to the counterparty's own compliance posture. The report notes that where treasuries cannot guarantee mutual recognition across a multi-economy corridor, they hold traditional fiat buffers anyway, which erodes the efficiency the arrangement was meant to produce.
Issuers and infrastructure firms. The GENIUS Act is identified as the only framework carrying a statutory mandate to develop technical interoperability standards. Elsewhere, interoperability is being pursued through pilots and public-private collaboration rather than binding rules, which leaves standards-setting as an open competitive position.
Policymakers. The correspondent banking retreat the report documents is stark: relationships fell roughly 30% between 2011 and 2022, with declines of 62.6% in the South Pacific, 52.1% in the Caribbean and 50.5% in South America, and small island developing states losing 41% of connections. That is the gap stablecoins are being asked to fill, and the report's caution is that foreign-currency tokens filling it carry currency substitution and monetary transmission consequences.
The useful conclusion is a scoping exercise
The report's own framing is that stablecoins complement rather than replace existing trade arrangements. That is correct and a little soft. The sharper version is that "stablecoins in trade" was never one question, and treating it as one has produced a debate where both sides are right about different things.
Where the financial requirement is the payment, which describes digitally delivered services and describes the large unserved majority of traders who already transact cash in advance or on open account, the instrument fits the problem. Where the financial requirement is credit, collateral and enforceable risk allocation, which describes most of merchandise trade by value, it does not, and no improvement in settlement speed changes that. The $2.5 trillion gap is in the second category.
What determines whether the first category grows is not the technology. It is whether the edges work: conversion at both ends, redemption rights that hold under local law, and compliance regimes that permit the combination rather than each separately allowing it. All three are legal and institutional questions, and all three sit outside the blockchain entirely.
Source: Stablecoins and world trade: Emerging role, opportunities and challenges, World Trade Organization Secretariat, September 2026. Principal authors Joscelyn Magdeleine and Kenza Zakarya of the Trade in Services and Investment Division, and Marc Auboin, Emmanuelle Ganne, Théo Mbise and Musa Sawaneh of the Economic Research and Statistics Division. The report was prepared under the WTO Secretariat's own responsibility and does not necessarily reflect the positions of WTO members.
Additional data: Trade finance gap figures from the Asian Development Bank's ninth Global Trade Finance Gap Survey, released 15 January 2026, as reported by Reuters. The regulatory detail in the compliance section was verified separately against published legal commentary and regulator sources, and is not drawn from the WTO report: the CBUAE Payment Token Services Regulation, including the financial free zone carve-out, the position of commercial free zones and the lapsed transition period; the sequence and scope of Brazil's Resolutions 519, 520, 521 and 561, including the 1 October 2026 effective date; and the structure of MiCA's threshold for non-euro e-money tokens. All other figures come from the WTO report and the studies it cites.
Methodology: We focused on the report's goods and services distinction, its treatment of trade finance functions, the regulatory stacking analysis and the GATS discussion, rather than its introductory survey of stablecoin types and market structure. The observation that unserved traders already transact on payment terms is our reading of the report's data rather than a claim it makes directly.
This article is for informational purposes only and does not constitute financial, investment, or legal advice.