Cross-border B2B payments: Why it breaks and where stablecoins actually help
The number of active correspondent banks worldwide fell by 22% between 2011 and 2019, and the steepest regional decline was Latin America at 34%, according to BIS data drawn from SWIFT payment messages. Over the same period, the value and volume of payments running through those rails kept growing. Fewer banks are moving more money, and the corridors that lost the most coverage are the ones where a supplier now waits a week for funds that left the buyer’s account on day one.
That gap is the case for cross-border stablecoin payments, and the case is narrow on purpose.
It applies to corridors where the correspondent network has thinned out, where capital controls sit between the payer and the payee, or where two trading countries have no direct banking relationship at all. Blockchain settlement being faster than banking in general is a claim that does not survive contact with a domestic euro transfer. If your business does not touch one of those corridors, the case is weak. If it does, the case is close to unanswerable.
The reason more European finance teams have not worked out which category they fall into is that the evaluation usually stalls on two things that have nothing to do with corridors. One is a piece of vocabulary. The other is a bank risk that stopped being a real risk several years ago.
What cross-border stablecoin payments solve that SEPA does not
Uve Poom, Co-founder and COO at CryptoSwift, put the European position plainly on a recent panel hosted by Hipther, and moderated by Silvia-Kairet Põld (CMO of the Estonian Web3 Chamber), alongside Swapin CEO, Evald-Hannes Kree. Inside the eurozone, SEPA moves money in under ten seconds, so for real-world payments across borders in Europe, stablecoins are not needed.
That is worth conceding early, because it is what makes the rest of the argument legible. SEPA’s ten seconds are a property of a shared clearing system with a single currency, a single rulebook, and a settlement guarantee that every participating bank has already agreed to. The speed is a feature of the zone, not banks.
Once a payment crosses out of that zone, the guarantee does not travel with it. What handles the payment instead is correspondent banking, which is a chain of bilateral relationships between institutions that each apply their own compliance review, their own cut-off times, and their own risk appetite. Every link in that chain is a place the payment can sit.
The Financial Stability Board’s 2025 progress report on the G20 cross-border payments roadmap states: satisfactory improvements at the global level are unlikely to be achieved in line with the 2027 timetable, and the policy work completed so far has not translated into tangible gains for end users.
Add capital controls and the chain gets longer.
Hannes uses a client selling goods into Argentina as the working example. The buyer cannot send euros out of the country. They can send US dollars, and that transfer has to pass through a government approval layer before it moves, then through the correspondent chain, then through the receiving bank’s own review.
Realistically that is a week, sometimes several. The cost is not only the fee. It is a week of float on a receivable, an FX position held open across that week, and a reconciliation that cannot close until the money lands.
The same invoice settled in stablecoins clears in minutes, and the seller receives euros in their existing bank account. The comparison worth making is against that corridor, not against SEPA. SEPA was never built to serve a payment leaving Buenos Aires, and stablecoins lose any comparison against a domestic euro transfer.
This is also why the strongest early adopters are rarely the businesses you would guess. Car dealers exporting to third countries. Construction suppliers. Commodity traders. Companies with no interest in digital assets whatsoever, who found that a customer in a difficult jurisdiction could pay them in three minutes instead of three weeks.
Why the word custodial stops the evaluation before it starts
Traditional businesses hear that stablecoins work. Then they hear that they will need a wallet, and that the wallet is either custodial or non-custodial, and the evaluation stops there. It sounds like a technical decision with legal consequences that nobody in the finance team is qualified to make, so the project goes back in the drawer.
Wallet infrastructure was built for people who wanted to hold digital assets, then marketed largely unchanged to companies whose actual requirement was for an invoice to get paid. The vocabulary of the first product has been sitting on top of the second one ever since.
The two models a finance team is choosing between are these.
Hold the asset. The business sets up a wallet, receives stablecoins into it, holds a balance denominated in USDC or USDT, and takes on key management, treasury exposure, internal access controls, and the question of when to convert. That is a treasury decision and should be evaluated as one.
Do not hold the asset. The customer pays in stablecoins. A regulated provider receives, screens, and converts them, and the business receives euros into its existing bank account on the same day. No wallet, no keys, no exposure to a token.
Most companies asking about stablecoin payments want the second one, and very few know it exists. The objection they raise in the first meeting is an objection to a product they were never required to buy.
Whether banks still penalise businesses that accept stablecoin payments
The second stall is the bank. A compliance officer at a legacy bank sees on-ramp and off-ramp activity moving through a business account and reads it as exposure they cannot explain.
Five years ago it was a reasonable thing for a business to fear, because accounts were closed and payments were returned. Hannes says that those conditions have largely gone.
For an EU-registered company receiving fiat from a regulated European provider, with transaction screening and Travel Rule data attached, a returned payment is now a rounding error rather than a live risk. The provider is licensed, the counterparty data exists, and the flow is documented end to end. That is a payment a bank can explain to its own supervisor.
What auditors need before a stablecoin payment is usable
The Travel Rule obliges the institutions on either side of a transfer to exchange information about who is paying and who is being paid. That data cannot sit on a public blockchain, since nobody wants counterparty names and invoice numbers published on chain, so it moves through a separate network between the regulated institutions involved.
It helps with bookkeeping. A payer name, a payee name, a reference field, and an invoice number are elementary properties of a fiat payment. Remove that and a stablecoin transfer arrives as an amount and a wallet address. An accountant cannot match it to an invoice, an auditor cannot identify the counterparty, and no finance team can close a month against a ledger of hashes.
What has to change in European rules for business volumes to move
One specific restriction limits how far this can scale in Europe. Transfers from a regulated custodial wallet to a self-hosted wallet above 1,000 euros are permitted only where the destination wallet belongs to the same user. Below that threshold, payments to third parties are fine.
For consumer transfers that threshold is workable. For business payments it is not, because 1,000 euros is not a commercial invoice.
The regulatory logic behind it is sound. Supervisors treat a self-hosted wallet the way they treat cash, because ownership cannot be established from the address alone, and a large cash payment to an unidentified party is worth questioning.
The argument from the compliance infrastructure side is that the European Union digital identity framework changes that. Under EUDI, a user will be able to hold verified identity credentials and cryptographically prove control of a self-hosted wallet.
Once that proof exists, the wallet has a named, verified owner attached to it, the screening and paper trail supervisors want it to become possible, and the cash analogy no longer holds. Lifting the threshold for wallets with proven ownership would open business-scale payments without giving up the control the threshold was written to preserve.
How to tell whether stablecoin settlement is worth evaluating for your business
Four conditions worth checking:
One, you sell into a market with capital controls, where the buyer’s funds cannot legally leave in your currency and have to be converted and routed through an approval layer first.
Two, you trade with counterparties in a corridor the correspondent network has retreated from, where payments take days, arrive short, or fail for reasons nobody in either bank can explain.
Three your receivables cycle is long enough that a week of settlement float has a measurable cost, whether that shows up as FX exposure, working capital, or a month-end that cannot close.
Four, your customers are already asking to pay this way, and the reason you have said no is that somebody mentioned wallets.
Swapin processes stablecoin payments for European businesses with euros arriving in an existing bank account and no wallet on the merchant side.
