Digital assets
Stablecoin settlement has become a corridor business, and banks are finally pricing an answer
Adjusted monthly settlement volume has close to doubled in a year, concentrated in a handful of payment corridors where correspondent banking is slow and expensive. The incumbent response has stopped being dismissal.
Illustrative. Bank Season is an editorial prototype. This story, its sources, the issuers named in it and every figure it quotes are invented to demonstrate the publication. Nothing here is reported fact or investment advice. Read the disclosure.
Raw stablecoin transfer volume is a useless number. It counts an automated market maker rebalancing itself forty times an hour the same way it counts an importer in Lagos paying a supplier in Guangzhou. Filtering out the bot traffic, the internal exchange movements and the round-trip arbitrage leaves a much smaller figure, and that figure is the one that has been growing in a straight line for two years.
$1.18trn
Adjusted stablecoin settlement volume, August 2026
Ravensworth Research methodology, which strips intra-exchange transfers, automated market maker activity and transfers between wallets controlled by the same entity.
Ravensworth puts the August figure at $1.18trn against $610bn in the same month of 2025, growth of 93%. The composition matters more than the headline. Roughly 44% of the adjusted total is now business-to-business payment, up from about 29% two years ago, with the remainder split between retail remittance, treasury movement at digital-native firms and collateral flows in derivatives markets.
Adjusted monthly stablecoin settlement volume
Billions of US dollars, September 2025 to August 2026. Ravensworth Research adjusted methodology.
There is no month in that series where growth stalled, including the two months in the spring when digital asset prices fell hard and trading volumes dropped with them. That decoupling is the strongest available evidence that the settlement business has separated from the speculation business. Payment flows do not care what bitcoin did last week.
Where the volume actually is
Settlement concentrates where the incumbent rails are worst. That is not a slogan, it is what the corridor data shows. The routes carrying the most business volume are the ones where a correspondent banking chain runs through three intermediaries, costs more than three percent all in, and settles in two to five business days.
Estimated business-to-business stablecoin settlement by corridor, August 2026
| Corridor | Share of B2B volume | Correspondent cost | Correspondent settlement time | Stablecoin all-in cost |
|---|---|---|---|---|
| Nigeria and Kenya to China | 21% | 3.4% to 5.1% | 2 to 4 business days | 0.7% to 1.2% |
| Argentina and Brazil to United States | 16% | 2.8% to 4.2% | 1 to 3 business days | 0.6% to 1.0% |
| Turkey and Gulf to South Asia | 12% | 2.2% to 3.6% | 2 to 5 business days | 0.5% to 0.9% |
| Southeast Asia intra-regional | 9% | 1.9% to 3.1% | 1 to 2 business days | 0.5% to 0.8% |
| United States to Western Europe | 4% | 0.4% to 0.9% | Same day | 0.4% to 0.7% |
Ravensworth Research estimates from on-chain attribution and payment processor disclosures. All-in stablecoin cost includes both fiat conversion legs and network fees.
The bottom row is the control case. On the US to Western Europe corridor, where correspondent banking is cheap and fast, stablecoins have captured almost nothing. The technology is identical in both directions. What differs is whether the incumbent leaves room.
We stopped arguing about the technology two years ago. The corridors we were losing were corridors where our own cost to serve was four percent and our settlement was Thursday. You do not need a philosophical position on blockchains to lose that business, you just need to keep charging four percent.
Adaeze Okpara, head of transaction banking at Northgate Bancorp
Corridor share is also unstable in a way the totals hide. Two of the four leading corridors were negligible three years ago and grew because a local regulator tightened access to dollar accounts, which pushed importers toward any settlement route that still functioned. Policy creates these corridors and policy can close them, so a payment firm underwriting a corridor at current volumes is underwriting a regulatory posture rather than a customer relationship.
What the banks are building
The institutional response has converged on tokenised deposits rather than on issuing stablecoins directly. The distinction is legal rather than technical. A tokenised deposit is a claim on the issuing bank recorded on a shared ledger, covered by deposit insurance within the applicable limits, and it never leaves the regulated perimeter. A stablecoin is a bearer claim on a reserve pool held by a non-bank issuer.
Bank settlement products compared with stablecoin rails
| Attribute | Correspondent banking | Tokenised deposit | Fiat-backed stablecoin |
|---|---|---|---|
| Settlement finality | 1 to 5 business days | Seconds within the network | Seconds to minutes |
| Operating hours | Banking hours, both ends | Continuous | Continuous |
| Counterparty exposure | Each correspondent in the chain | Issuing bank, insured to limits | Issuer and reserve manager |
| Reach outside the network | Universal | Members only | Anyone with a wallet |
| Compliance screening | At each hop | At issuance and redemption | At the on and off ramps only |
| Typical all-in cost | 0.4% to 5.1% by corridor | 0.15% to 0.45% | 0.4% to 1.2% |
Fee ranges are indicative for business payments above $50,000 and exclude foreign exchange spread, which is frequently the largest single component in all three columns.
Tokenised deposits win on cost inside the member network and lose badly outside it. A Nigerian importer whose bank is not a member gains nothing. That is the structural limit on the bank response, and it is the reason the corridors in the table above are unlikely to be won back quickly. Network membership takes years to build and the existing networks are concentrated among large banks in developed markets, which is precisely where the problem was already solved.
0.15%
Low end of tokenised deposit settlement cost for large business payments
Compared against 0.4% at the cheapest end of the stablecoin range once both fiat conversion legs are included.
That point is increasingly well understood by the payment firms themselves, which is why the newer entrants are building local currency liquidity pools rather than competing on transfer fees. The margin in the business has moved one step upstream.
Nobody in Lagos wakes up wanting a stablecoin. They want dollars at a rate close to the one on the screen, today. The token is the plumbing. If we ever start selling the plumbing instead of the outcome, we deserve to lose to a bank.
Felix Aumann, founder of Ostrander Payments
Regulatory treatment remains the variable that could reorder all of this. Reserve composition rules, redemption guarantees and the question of whether a non-bank issuer may pay yield to holders are all unsettled in the major jurisdictions, and each has the capacity to change the economics substantially. The settlement volume has grown regardless, through three separate regulatory regimes and two significant price drawdowns, which suggests the demand underneath it is not speculative.
The number to watch through the end of the year is the business-to-business share rather than the total. If it keeps rising toward half of adjusted volume, this is a payments industry story that happens to use digital asset infrastructure. If it flattens while the total keeps climbing, the growth has gone back to being about trading.
.jpeg){kind=link}