Stablecoin Payments: How the Money Actually Moves in 2026

Stablecoin payments explained from bank funding and on-chain settlement to FX, liquidity and local payouts, with 2026 data, worked examples and market analysis.

An editorial illustration of banks, wallets, payment terminals and stablecoin settlement connected in one payment flow

Last verified: 2026-08-23

A stablecoin payment can begin with dollars in a bank account and end with pesos, euros or reais in another bank account. The payer may never buy a token. The recipient may never open a crypto wallet. Somewhere between them, however, a stablecoin can replace part of the chain of bank balances, correspondent accounts and settlement windows that traditionally moves money across borders.

That is the useful way to think about stablecoin payments in 2026. Not as a new button at checkout, and not as $60 trillion of people buying things with USDT. Stablecoins are increasingly a settlement rail inside payment products that still look familiar at either end.

The distinction matters because the numbers are easy to abuse. BCG and Allium Labs counted more than $62 trillion of stablecoin transfers on public blockchains in 2025. After removing bots, internal movements, intermediary routing and other activity that did not represent a distinct economic transaction, they were left with about $4.2 trillion. Their narrower estimate for identifiable payments for goods and services was $350 billion to $550 billion. BCG and Allium describe the payment figure as a conservative lower bound because some off-chain activity, including card payments, is not visible in their methodology.

Key point

A transfer is not necessarily a payment. A blockchain records that a token moved. It does not tell you, by itself, whether somebody paid an invoice, an exchange moved inventory, a market maker rebalanced, or a protocol routed the same money through several addresses.

Start with the payment, not the stablecoin

Suppose a U.S. importer owes a European supplier €100,000. The importer holds dollars. The supplier wants euros. There are several ways to execute that payment, and the stablecoin version does not require either company to become a crypto trader.

In a conventional cross-border payment, the importer's bank or payment provider takes the instruction, prices the foreign exchange, moves funds through its own accounts or correspondent banks, and hands the payment to a local euro rail. Depending on the provider and corridor, money may be prefunded at the destination so payouts can happen before the corresponding cross-border settlement is complete.

A stablecoin-enabled provider can rearrange the middle. It can take dollars from the importer, acquire or mint a dollar stablecoin, transfer it on-chain, sell that stablecoin for euros or a euro stablecoin, and pay the supplier through a local bank rail. The two companies can still see dollars and euros.

Payment flow

USD bank account → payment provider → USDC/USDT → blockchain → liquidity + USD/EUR FX → euro off-ramp → SEPA/local rail → supplier

There is no requirement that every transaction follow that exact route. A provider may already hold stablecoin liquidity. It may net flows internally. FX can occur before the blockchain transfer, after it, or between stablecoins denominated in different currencies. What changes is the settlement inventory available to the payment company: it can use a token that moves continuously rather than relying only on bank money moving through bank operating windows.

Where the economics actually sit

The network fee is the easiest number to see and often the least useful number for comparing payment systems. A cheap blockchain transfer does not mean a cheap cross-border payment.

Cost or constraint — Conventional route — Stablecoin-enabled route

Funding — Bank transfer or provider balance — Usually still bank money unless payer already holds stablecoins

Settlement — Internal ledgers, correspondents, payment networks — Stablecoin on a supported blockchain

FX — Bank/PSP/FX-provider spread — Still required whenever source and destination currencies differ

Liquidity — Provider balances and correspondent liquidity — Stablecoin/fiat or stablecoin/stablecoin liquidity

Prefunding — Often used for fast local payout — Can be reduced in some models because settlement inventory moves 24/7

Network fee — Usually embedded — Visible blockchain fee, often small relative to FX/payout economics

Last mile — ACH, SEPA, Pix, RTP or local equivalent — Usually the same local rails after off-ramp

Operating hours — Depends on banks and payment systems — Blockchain leg is continuous; fiat endpoints may not be

Take the €100,000 supplier payment. Assume, purely for illustration, that the interbank market is $1.1700 per euro. The reference dollar value is $117,000.

Worked example

Reference: €100,000 × $1.1700 = $117,000

Illustrative all-in FX spread of 40 basis points: $117,000 × 0.40% = $468

Funding requirement before any fixed provider or payout charge ≈ $117,468.

If the blockchain transfer costs $0.20, that does not make this a $0.20 payment. In the illustration, the FX economics are more than two thousand times larger than the network fee. If liquidity in the required pair is poor, the spread can widen further. If the provider has strong two-way flow and local liquidity, it can tighten.

The same discipline applies to speed. A stablecoin may reach a destination wallet in seconds while the euro payout waits for a bank, compliance review or local rail. The relevant metric is time from funds available at the payer to funds usable by the beneficiary, not blockchain finality alone.

Why 24/7 settlement matters to payment companies

The more interesting advantage is often balance-sheet efficiency. Cross-border payment providers commonly keep money in destination markets so they can pay beneficiaries quickly. That prefunding ties up capital. If settlement between participants can happen continuously, a provider can in some cases replenish or rebalance positions more frequently rather than waiting for banking windows.

Circle Payments Network markets stablecoin settlement partly around reducing prefunding and connecting banks, PSPs and other financial institutions through one network. Circle said in its second-quarter results that CPN reached a $14.7 billion annualized transaction-volume run rate for the trailing 30 days at quarter-end, with 175 financial institutions enrolled. Those are Circle's figures, not a measure of the entire stablecoin payments market.

The Circle-Nium partnership makes the architecture easier to see. A financial institution can route a payment through CPN using USDC for settlement, while Nium handles the local payout through its network covering more than 190 countries and 100 currencies. Stablecoin settlement and local fiat delivery are separate jobs joined into one payment.

Visa shows why “paying with stablecoins” is the wrong test

Visa's stablecoin activity is useful precisely because the cardholder does not need to know it exists.

Visa said in April that its stablecoin settlement pilot had reached a $7 billion annualized run rate, up 50% quarter over quarter, and expanded to nine blockchains. Issuers and acquirers can use stablecoins to settle with Visa while consumers continue paying with cards and merchants continue receiving the products their acquirers provide.

At its 2026 Payments Forum, Visa said more than 160 stablecoin-linked card programs were live or in development. Those cards are a different use case from network settlement: one lets a holder spend a stablecoin-backed balance through existing merchant acceptance; the other uses stablecoins between financial institutions behind the card transaction.

Activity — What moves — Who notices the stablecoin? — Example

Wallet payment — Stablecoin payer → recipient — Usually both sides — USDC invoice paid to a wallet

Cross-border rail — Stablecoin in the middle; fiat can sit at both ends — Possibly neither endpoint — USD funding → stablecoin → local bank payout

Card funding — Stablecoin-backed balance converted for spend — Cardholder/provider — Stablecoin-linked Visa card

Network settlement — Stablecoin between institution and network — Institutions — Visa issuer/acquirer settlement

Treasury movement — Corporate liquidity — Treasury/provider — Weekend liquidity rebalance

Stripe is building the whole middle

If Visa shows stablecoins disappearing inside an established network, Stripe shows what happens when the infrastructure is assembled into software.

Its 2026 roadmap spans stablecoin acceptance, payouts, wallets, treasury accounts, cards, on- and off-ramps, cross-chain movement and foreign exchange. Through Bridge, Stripe says fixed-output FX is already generally available: the recipient amount can be specified first and the system determines how much source currency is required. Bridge also supports SWIFT and FedNow for parts of its fiat connectivity. Through Privy, Stripe is building wallet and custody infrastructure into the same stack.

This matters because payments are bought as outcomes. A marketplace does not want “a blockchain transaction.” It wants to collect $100 from a buyer, keep $8, pay $92 to a seller in another country, satisfy compliance requirements, and know what the transaction will cost before it commits.

Mastercard's BVNK acquisition makes the same point from another direction

Mastercard completed its acquisition of BVNK on Aug. 3. BVNK provides infrastructure to send, receive, store, spend and convert stablecoins and fiat while abstracting wallets, liquidity providers, blockchain connections, banking relationships and compliance.

That is a revealing asset for a global payment network to buy. The strategic value is the translation layer between forms of money. Visa, Stripe, Circle and Mastercard approach the market from different starting points, but the direction is similar: stablecoins are being folded into payment infrastructure rather than built as a parallel consumer payment universe.

Where stablecoins are actually being used

The strongest evidence remains cross-border and business-oriented. BCG and Allium estimate that identifiable B2B flows made up about 40% of real-economy stablecoin payment value in 2025 and were growing about 65% annually. Consumer-to-consumer transfers, led partly by remittances, accounted for roughly a quarter. Consumer-to-business payments represented another quarter, with digital services among the important categories.

That distribution is logical. Stablecoins are most useful where the existing payment is awkward: cross-border settlement, expensive or opaque FX, weekend liquidity, fragmented payouts, limited access to dollar accounts, or businesses operating across several banking systems.

They have a weaker proposition where domestic rails already work exceptionally well. Replacing Pix in Brazil or PayNow in Singapore simply because a blockchain is available solves a different problem from connecting those systems internationally.

The chain matters, but not in the way crypto markets usually discuss it

For a payment company, the “best” blockchain is not necessarily the chain with the loudest ecosystem or highest token valuation. It needs the stablecoin the provider wants to settle in, sufficient liquidity, predictable transaction costs, acceptable finality, reliable infrastructure, compliance tooling and support from institutions on either side.

BCG and Allium estimate Tron handled 60% to 80% of identifiable real-economy stablecoin payment value in 2025, reflecting low fees and entrenched USDT distribution. They also found incremental institutional activity spreading across Ethereum, Solana, BNB Smart Chain and other networks. Visa's expansion to nine blockchains points in the same direction: payment infrastructure is becoming multi-rail rather than choosing one universal chain.

Stablecoins do not eliminate foreign exchange

A dollar stablecoin is still a dollar-denominated asset. A supplier that prices in euros has not stopped needing EUR because the payer can send USDC.

The FX can move to a different venue. It can be bundled into an orchestration quote. It can increasingly occur between stablecoins denominated in different currencies. But somebody still bears the currency conversion and liquidity cost.

This is where non-dollar stablecoins become strategically interesting. Deeper EUR, BRL, SGD or other local-currency stablecoin liquidity would allow more of the currency conversion to happen on-chain. Until then, many stablecoin corridors remain dollar bridges with conventional FX at one or both edges.

How to read stablecoin payment statistics without getting fooled

Metric — What it measures — What it does not prove

Gross transfer value — Value moved between blockchain addresses — That the value paid for goods or services

Adjusted economic activity — Transfers remaining after behavioral filtering — That every remaining transfer is a payment

Payment volume — Transfers classified as economic payments — Complete market size when off-chain/card flows are missing

Settlement run rate — Annualized activity in a particular settlement program — Consumer merchant-spend volume

Trading volume — Stablecoins exchanged in markets — Commerce

Mint/redemption — Primary-market creation/destruction of tokens — End-user payment demand

This is why the $62 trillion figure and the $350 billion to $550 billion estimate can both be correct. They answer different questions. BCG and Allium's $4.2 trillion adjusted figure is about organic economic activity after filtering. Their smaller payment estimate attempts to isolate bilateral payments for goods and services. It deliberately leaves uncertain activity out rather than guessing.

What stablecoins change — and what they do not

Stablecoins can make a settlement asset programmable, globally transferable and continuously available. That can reduce dependence on banking windows, make some treasury movements easier, and give payment companies another way to manage cross-border liquidity.

They do not remove the need for regulated access to bank accounts. They do not remove KYC, sanctions screening or transaction monitoring. They do not create deep FX liquidity by themselves. They do not make a local payout rail operate on Sunday if that rail is closed. And they do not make every cross-border payment cheaper simply because the on-chain transaction fee is low.

The practical advantage is architectural: a payment provider can combine stablecoins with existing rails and choose where each is strongest.

The most important stablecoin payment may be the one nobody sees

The consumer version of stablecoin adoption is easy to picture: hold USDC, scan a code, pay a merchant. That will remain part of the market. It may not be the part that matters most.

A more consequential version is already emerging. A company funds in fiat. An infrastructure provider moves settlement value through stablecoins. FX happens where liquidity is best. A local partner pays the beneficiary. Or a card issuer settles with Visa seven days a week while the cardholder notices nothing different at checkout.

In that world, the stablecoin is not the payment experience. It is one of the assets the payment system uses to keep money moving.

That is also why the market should not be judged by gross blockchain transfer volume or by the number of merchants displaying a crypto logo. The better questions are how much genuine payment activity exists, where stablecoins improve the economics, how much capital they free from prefunding, how deep the FX and redemption liquidity is, and whether providers can connect on-chain settlement to the bank accounts and payment systems people already use.

Stablecoins do not need to replace the financial system to become important. They need to become useful enough that the financial system starts using them.