Twenty-one banks commit to a jointly owned stablecoin company

Twenty-one financial institutions have committed to establish a company in the second half of 2026 to issue a jointly owned stablecoin, the group announced this week. The first product will be a dollar token, targeted for market in the first half of 2027, with stablecoins in other G7 currencies to follow and a euro token named as the next priority.

For a decade the question hanging over stablecoins was when the banks would arrive. The answer turns out to be: together, all at once, through a company none of them controls alone.

Who signed

The membership list rewards a slow read, because the geography is the strategy.

From North America: Bank of America, Capital One, Citi, Fidelity Investments, Goldman Sachs, PNC Financial Services, Scotiabank, TD Bank Group, Wells Fargo and WisdomTree. That is four of the five largest US retail banks, two of the largest custodians of American wealth, and Canada's two most international lenders.

From Europe: Banco Santander, BBVA, Commerzbank, Crédit Agricole, Deutsche Bank, Lloyds Banking Group, Rabobank and UBS. Spain, Germany, France, Britain, the Netherlands and Switzerland are each represented by at least one institution of systemic size.

The remaining three cover the gaps: MUFG, Japan's largest bank; Standard Bank, Africa's largest by assets; and Sirius International Holding from the Middle East. Twenty-one names, four continents, and almost every major payments corridor between them.

Notably absent: JPMorgan, which has run its own deposit token programme for wholesale clients for years and has little reason to share, and every large card network, which as of this summer are building a competing consortium of their own.

What was actually committed

The commitment is to establish a company, not yet to launch a product. The venture is due to be stood up in the second half of 2026, with the dollar stablecoin targeted for the first half of 2027. Boston Consulting Group and Brunswick Group are advising, and the announcement is careful to note that the advisers cannot bind the participants, which is the sort of sentence lawyers insert when the participants themselves are not yet fully bound either.

The stated design goals read like a checklist assembled by a risk committee, because they almost certainly were: bank-grade compliance, strong governance, institutional risk management, and regulatory alignment with the GENIUS Act in the United States and MiCA in Europe as applicable. The token is meant to serve three segments at once: wholesale, institutional and retail, with cross-border payments and the settlement of digital assets named as the first applications.

What the announcement does not contain matters as much as what it does. There is no company name. No jurisdiction of incorporation. No chosen blockchain, no reserve manager, no custodian, no chief executive. Those are not details; they are the decisions that will determine whether this venture ships in eighteen months or becomes a working group with a press release.

Why banks waited this long

The caution has history behind it. When Facebook unveiled Libra in 2019, several banks looked hard at the consortium model and concluded the regulatory cost of association was unpayable; Libra was rebranded, shrunk, and finally sold for parts in 2022 without ever launching at scale. The lesson institutions drew was not that shared digital currencies were a bad idea, but that launching one without a legal regime to put it in was.

That regime now exists on both sides of the Atlantic. MiCA's stablecoin provisions have applied in the European Union since mid-2024, giving e-money tokens a licensing path and reserve rules. The United States followed in July 2025 with the GENIUS Act, which for the first time defined a federal category of payment stablecoin, with licensing through federal or state channels and full-reserve requirements. The banks are not early. They waited, precisely and deliberately, until issuing a stablecoin became a regulated activity with a rulebook, because a rulebook is the terrain on which large banks win.

There is also a defensive motive, and it is the one the announcement politely omits. Stablecoins moved roughly 300 billion dollars of supply into circulation while banks watched, and every dollar that settles on token rails is a dollar of deposits, float and fee income that settles somewhere other than a bank ledger. Tether alone holds a reserve portfolio larger than the balance sheets of most mid-sized banks, and it built it substantially out of flows the banking system declined to serve. A jointly owned issuer is the banking industry's attempt to bring that activity back inside the perimeter, on infrastructure it owns.

The market they would enter

Scale the incumbency properly. Total US dollar stablecoin supply stands at roughly 309 billion dollars. Tether's USDT accounts for about 183 billion of it, a share near 59 percent; Circle's USDC holds around 75 billion, or 24 percent. Between them, two issuers control five of every six dollars in the category the banks now propose to enter, and both have spent years wiring themselves into exchanges, wallets, payment processors and corporate treasuries.

The economics of that supply explain the appetite. A payment stablecoin's reserves sit in cash and short-dated government paper, and at recent short-term rates the float income on a nine-figure supply is measured in billions of dollars a year. Tether's profitability, disclosed in its attestations, has for several years exceeded that of most banks on the consortium list. The banks are not only defending deposits; they are pursuing one of the best fee-free revenue models in finance, one that happens to be built out of the asset they already manage better than anyone: short-term dollar liquidity.

There is a definitional subtlety the venture will have to navigate in public. A stablecoin is a bearer instrument that moves between parties who need not bank with the issuer. A tokenised deposit, the model JPMorgan chose for its wholesale token years ago, stays inside the issuing bank's ledger and its customer relationships. The consortium's stated ambition, wholesale, institutional and retail, only makes sense as a true stablecoin, which means the banks are choosing to issue an instrument that can circulate away from them. That is a genuine strategic concession, and it is what makes this announcement different from every bank token programme before it.

The consortium problem

The structural risk is the number twenty-one itself.

Bank consortiums have built durable infrastructure before: the US real-time network run through The Clearing House, the Zelle transfer system owned by a group of large American banks, and Fnality, the London-based wholesale settlement venture owned by a syndicate of global institutions, which has been methodically bringing tokenised central-bank-money settlement to market since 2019. Each of those succeeded, where it has, by picking a narrow job and giving the operating company real independence.

The failure pattern is just as well documented. Shared ventures stall when every shareholder holds a veto, when the product committee has twenty-one seats, and when no single member's revenue depends on shipping. Eighteen months from commitment to launch is a demanding schedule for one bank. For twenty-one, each with its own risk, legal and brand review, it is heroic.

And the clock is running against incumbents who ship weekly. Tether and Circle will not pause while the venture incorporates. Nor will the card networks' Open Standard consortium, announced in June with more than 140 members and a no-fee mint-and-redeem model aimed squarely at the same institutional flows. By the first half of 2027, the corridors this token wants to serve will each have an incumbent with years of settled volume.

The nearest precedent deserves a paragraph of its own. Fnality, the London venture that grew out of the Utility Settlement Coin research project, was funded by a syndicate of global banks in 2019 to build settlement in tokenised central bank money, and has been edging into live operation ever since, methodically and slowly, with its shareholder banks as its first users. It proves the model can work. It also proves the timescale: Fnality took the better part of a decade from concept to meaningful operation, in a wholesale niche, with regulators actively supportive. The new venture is promising a retail-capable product across multiple jurisdictions in eighteen months.

What a bank-issued token changes

Assume the venture ships. What would actually be different?

Distribution, first. A stablecoin issued by a company owned by Santander, BBVA and Standard Bank does not need to win listings; it arrives pre-installed in the treasury relationships of a meaningful share of world trade. Corporate treasurers who cannot hold an asset issued by an offshore trust can hold one issued by a consortium of their own lenders, inside existing credit and custody agreements.

Redemption, second. The deepest anxiety about existing stablecoins has always been the redemption queue in a stressed weekend. A token redeemable at par across twenty-one banks in a dozen jurisdictions is a different promise, and closer to the correspondent-banking model treasurers already understand.

And regulatory treatment, third. GENIUS-licensed, MiCA-aligned issuance through regulated institutions would make the token holdable by counterparties whose compliance departments currently forbid the category entirely. The addressable market for a bank-grade stablecoin is not crypto users. It is the far larger population of institutions that have so far been unable to touch the asset class at all.

Inside the plumbing decision

Of the unmade decisions, the chain choice will say the most about who won the internal argument. A public network maximises reach and composability: the token can sit in any wallet, plug into any exchange, settle against any tokenised asset, which is what the retail and cross-border ambitions require. A permissioned ledger maximises control: known validators, reversibility in extremis, and a comfort blanket for twenty-one risk committees. Every bank token effort of the last decade chose control and paid for it in irrelevance; the stablecoins that matter all live on public rails. If the venture announces a permissioned-only design, read it as the wholesale faction winning and the retail ambition quietly dying.

Custody and interoperability follow the same logic. A token that moves across several public chains needs native issuance and burn-and-mint transfer on each, the model Circle standardised for USDC, plus custodians the member banks will accept. None of this is research; it is procurement. Which is the point: for the first time, a bank stablecoin is an integration project rather than an invention project.

What a treasurer actually gets

Cut through to the customer. Picture the treasury desk of a European exporter selling into Latin America, banking with two consortium members. Today its dollars move by correspondent chain: cut-off times, two-day tails on some corridors, pre-funded accounts in three cities earning nothing, and a weekend during which money is simply unavailable. With a bank-issued token held at its own lender, the same desk holds one balance, redeemable at par, movable in minutes at any hour, acceptable as settlement by any counterparty that banks with any of the twenty-one.

That final clause is the network effect the venture is really selling. Every additional member bank is not one more distribution channel; it is a multiplication of counterparty pairs that can settle without leaving the club. Twenty-one banks is roughly two hundred and ten pairs. That, and not the press release, is the product.

The euro question

The announcement names a euro stablecoin as the second product and the stated priority after the dollar, and that choice is more pointed than it looks. Euro stablecoins barely exist: the entire category circulates in the low single-digit billions, a rounding error against dollar supply, despite MiCA having provided a clear e-money token framework since 2024. The gap is structural. Negative-to-low euro rates for most of the last decade made float economics unattractive, and European corporate treasury has had less reason to hold tokenised cash than emerging-market users fleeing weak currencies.

A consortium that includes Santander, BBVA, Deutsche Bank, Crédit Agricole, Rabobank, Commerzbank and Lloyds is the first plausible issuer of a euro token at scale, and it would enter a lane where the European Central Bank is simultaneously building a digital euro with an explicitly defensive mandate. How Frankfurt receives a private euro stablecoin issued by the continent's own systemic banks, as a complement to the digital euro or as competition for it, will be one of the more delicate conversations of 2027.

What to watch

Four decisions will tell you whether this is real, and they should all land within months.

The name and charter path: whether the company seeks a US licence federally or through a state regime, and where in Europe it sits under MiCA. The chain: a public network would signal ambition; a permissioned one would signal that the wholesale segment won the internal argument. The reserve manager: twenty-one banks each want the deposits, and how they split the float is the venture's real shareholder agreement. And the leadership: a chief executive hired from a payments company would mean a product; one seconded from a member bank would mean a committee.

The stablecoin market has spent a decade being built by outsiders because insiders would not touch it. The most important sentence in this announcement is the implicit one: that era is over.