The token stays in-house: what Wells Fargo’s tokenised deposit can do this autumn, and what does not arrive until 2027

On 4 August 2026 Wells Fargo & Company announced that it will introduce tokenised deposits this autumn: a representation of commercial bank money on the bank’s own blockchain that, in the words of the release, lets corporate clients move, programme and settle funds “24/7/365” without leaving the regulated, insured banking system. The launch is limited to selected corporate and commercial clients and to a single corridor, the exchange of US dollars into sterling; a broader rollout to more clients, countries and currencies is, according to Wells Fargo, planned for the course of 2027. Chief financial officer Mike Santomassimo calls the move, in the same release, “an important step forward as we expand payment options, including on-chain solutions”.

The announcement reads like a product launch. For a treasurer it is, first of all, a timetable, and the timetable has two halves. Wells Fargo delivers the first half alone, and this autumn: a token that moves between accounts, subsidiaries and counterparties, provided all of them bank with Wells Fargo. The second half is settlement with another bank. On 5 June the bank handed that task, together with 16 other institutions, to The Clearing House, operator of the CHIPS and RTP networks. That release names no date; the Wall Street Journal reported the same day that the banks were aiming for the first half of 2027. Between the two halves lies at least one winter in which one bank’s token is worth nothing to a supplier who banks elsewhere. Stablecoins live in precisely that space, and since 1 September Wells Fargo has been on the participant list there as well, as one of 21 banks announcing a dollar stablecoin of their own for the first half of 2027. What follows is a description of that starting position, drawn from the participants’ own releases; it is not a forecast of how it ends.

This autumn the token moves only between accounts at the same bank

Wells Fargo’s release promises three things. First, settlement that does not wait for batch runs and wire windows: clients are to move funds “between accounts, subsidiaries, or counterparties 24/7, including weekends and holidays”. Second, programmability, in the bank’s words “conditional payments with Wells Fargo smart contracts, which will allow the release of funds based on predefined logic”. Third, the same protections as before: the tokenised deposit is to carry “the same regulatory protections and deposit insurance eligibility” as any other deposit product of the bank. Nothing changes in how clients interact with the bank, the release says; the capability is integrated into the existing offering and routes payments through tokenised deposits on its own whenever that improves speed or timing.

The sentence that matters comes further down. The capability rests on “its leading proprietary blockchain platform, which can support in-house custodial wallets and the bank’s inter-chain connectivity technology in future offerings”. Connectivity to other chains is announced, then, and not delivered. The release contains no name for the platform, no technology partner and no statement on whether the network is permissioned, and the financial press has added none since. This autumn the token moves between Wells Fargo accounts. Everything beyond the bank belongs to the second half of the timetable.

The undertaking is not new for the bank. In September 2019 Wells Fargo announced “Digital Cash”, an internal settlement network on the enterprise version of Corda, the ledger built by R3, which a spokesman told CoinDesk was “not connected to any other digital cash solutions”; in December 2021 a bilateral blockchain settlement for foreign exchange trades with HSBC followed, in dollars, sterling, euros and Canadian dollars, which by November 2022 had settled more than USD 200 billion according to CoinDesk. Dollars against sterling is therefore a corridor the bank has run on a ledger with a partner bank for years; what is new in 2026 is the client who holds the token. Whether the platform is built on the old one, no source says.

The choice of corridor is plausible, then, but it is also modest. Dollars against sterling is one of the most densely served payment routes in the world: two liquid major currencies, a tight correspondent banking network, Swift gpi with credit within hours. The advantage Wells Fargo promises lies at the weekend, on the public holiday and in the condition that triggers a payment, and not in the underlying speed on a Tuesday morning. For a treasurer managing liquidity across time zones that carries weight; it is simply a different advantage from the one the headline suggests.

How much demand there is, the chief executive has assessed himself. Charlie Scharf told CNBC on the day of the announcement, as American Banker reports, that clients were asking about tokenisation “because they’re curious about it” and not because they thought “there’s a huge benefit to them yet”. James Wester, co-head of payments at Javelin Strategy & Research, put to the same publication the question that runs through every announcement of this kind: “We know what we’re being told can be done with deposit tokens and even stablecoins, for that matter. But is anyone using them?” Wells Fargo had to enter the market to stay competitive in liquidity and cash management, he said; without interoperability between banks, deposit tokens would have “relatively limited utility”. A product that its provider justifies by its clients’ curiosity is rarely a bad product, and almost always an early one.

Seventeen banks are building the bridge, and their release names no date

The second half of the timetable dates from 5 June 2026. That day The Clearing House in New York announced an initiative to enable “clearing and settlement of tokenized commercial bank money at scale”. The release names two components: on-chain clearing and settlement of tokenised deposits between banks within the established banking framework, with settlement around the clock, and a connectivity layer to the existing RTP and CHIPS networks so that money can move between digital and conventional form. Seventeen banks appear with a quote, from Bank of America through HSBC, J.P. Morgan and PNC to Wells Fargo; the operator is The Clearing House itself, which by its own account is owned by 25 of the largest US financial institutions and whose networks clear and settle more than USD 2 trillion a day.

The release contains no start date, and no blockchain or technology partner either. The first half of 2027 comes from the Wall Street Journal, which reported the same day that the banks were planning a launch in that window, called the project “the bridge” or “the chain” internally and had not yet chosen a blockchain provider. Set the two halves of the timetable side by side and this is the picture: a token that lives from autumn 2026, and a network that carries it to the next bank in spring 2027 at the earliest, if the newspaper’s timetable holds.

Nellie Liang, now a senior fellow at the Brookings Institution and previously Under Secretary for Domestic Finance at the US Treasury, set out in one sentence in April 2026 why the network is what counts: “Interbank settlement of tokenized bank deposits offered on private blockchains does not exist.” The interoperability of conventional deposits is carried by the Federal Reserve’s payment rails, though neither in real time nor around the clock. For tokens on a single bank’s chain, that route does not yet exist. Forbes contributor Zennon Kapron wrote on 28 July: “Each of these banks has spent years tokenizing money alone. The new fact is the word shared.” An interbank token, he argued, is the first product with which banks meet stablecoins on their own ground, and it brings a problem stablecoins never had to solve: “getting four of the most competitive institutions on earth to run one ledger”.

The record of bank consortia that have tried exactly that is known, and Zennon Kapron lists it with sources: the trade finance platform we.trade, backed among others by HSBC, Deutsche Bank and Santander, entered insolvency in 2022; Marco Polo, backed by BNY and Commerzbank, in February 2023; Contour, built by nine banks for letters of credit, shut down in late 2023 while processing a few dozen transactions a month. Against that stands one success, and it belongs to the same houses: Zelle, run by Early Warning Services and owned by seven large banks including Wells Fargo, moved USD 1.2 trillion in 2025 according to its operator, a fifth more than the year before. The pattern, in Zennon Kapron’s words: “consortia die when the pain is asymmetric and the operator is new”; Zelle worked because the threat was existential and shared and the operator already existed. The Clearing House network has both conditions. What it lacks, he writes, is time.

The members themselves speak more soberly than their release. Mark Monaco, head of global payments solutions at Bank of America, said the day before the announcement, according to PYMNTS, that clients were not beating down the door for tokenised deposits, and that “with any sort of new adoption, it takes time”. Demand is expected rather than measured, and the network is being built against a threat that every participant considers real without a single client having commissioned it so far.

The same banks sit in the stablecoin camp at the same time

On 1 September 2026 a third release appeared in London and New York, this time with no operator in the title. In it, 21 financial institutions commit to establishing a company in the second half of 2026 “to support the issuance of a stablecoin solution”, initially in US dollars and later in further G7 currencies with the euro as a priority; market launch “in the first half of 2027”. The stablecoin is to run on public blockchains as “a 1:1 reserve-backed form of digital money” and, by its own account, to meet the requirements of the GENIUS Act and MiCA. The list reads like June’s: Bank of America, Citi, PNC, TD, Santander and Wells Fargo appear on both. They are joined by Goldman Sachs, Capital One, Fidelity, Deutsche Bank, Commerzbank, UBS, BBVA, Crédit Agricole, Lloyds, Rabobank, Scotiabank, WisdomTree, MUFG, Sirius International and Standard Bank. J.P. Morgan, whose own blockchain Kinexys settles USD 7 billion a day by the bank’s account, is absent.

The camp is well populated. On 30 June, according to Fortune, a consortium of more than 140 companies around Visa, Mastercard, Stripe and Coinbase, joined by BlackRock, BNY and Standard Chartered, had announced Open USD, a jointly governed dollar stablecoin, for a launch “later in 2026” without a date; shares in the issuer Circle fell by 15 to 17 per cent on the day, according to several reports. On 9 July Swift announced a shared ledger for tokenised deposits that 17 banks from six continents intend to pilot, among them, according to CoinDesk, Wells Fargo again. The consultancy Flagship Advisory Partners counted on 17 August: 52 per cent of the 50 largest US banks were evaluating or building tokenised deposits, up from 18 per cent at the end of 2025, and “dozens of schemes are being built in parallel, and almost none of them talk to each other”. The risk, the authors wrote, was “a solution searching for a problem”.

Wells Fargo is therefore running at least three tracks rather than two, and they contradict one another only at first sight. The bank’s own platform keeps the client in-house, the Clearing House network connects the houses, the stablecoin leaves them. As recently as the earnings call in October 2025, Jane Fraser, chief executive of Citigroup, still set out a clear ranking: “There’s an overfocus on stablecoin at the moment. Most of this is going to get solved by tokenized deposit capabilities.” Eleven months later her bank appears on all three lists. Only the lists taken together give the picture: whoever does not know which instrument will prevail builds all three and lets the client decide. Zennon Kapron describes the same stance with another example: two of the four founding banks of the Clearing House network were simultaneously funding the rival settlement consortium Fnality, “which tells you how much conviction any single design commands”.

Deposit or stablecoin: for the treasurer the balance sheet decides, not the blockchain

The three tracks differ less in technology than in the question of whose balance sheet carries the money. The GENIUS Act, in force since 18 July 2025, defines the payment stablecoin in Sec. 2(22) and in subparagraph (B) expressly excludes anything that “is a deposit (as defined in section 3 of the Federal Deposit Insurance Act), including a deposit recorded using distributed ledger technology”. Wells Fargo’s token is thus, in law, what it is in economics: a deposit kept on a different ledger. The Federal Deposit Insurance Corporation (FDIC) confirmed as much on 7 April 2026 in its proposed rule implementing the GENIUS Act: tokenised deposits that meet the statutory definition “would be treated no differently under the Federal Deposit Insurance Act than any other types of deposits”. The stablecoin, by contrast, may not pay its holders “any form of interest or yield” for merely holding it, under Sec. 4(a)(11), and it is not a deposit, so it is not insured. What the supervisors have not yet answered, the Conference of State Bank Supervisors listed in its letter of 4 November 2025 to the FDIC, the Federal Reserve and the Office of the Comptroller of the Currency (OCC): deposit insurance and recordkeeping, anti-money-laundering checks, liquidity risk under “always-on, 24/7 redemption”, smart contracts. The legal framework is clear; supervisory practice is not yet.

In Europe the line runs in the same place. Art. 2(4)(b) MiCA excludes “deposits, including structured deposits” from its scope; the European Banking Authority’s (EBA) report on tokenised deposits of December 2024 draws the conclusion in paragraph 14: “when in the form of a crypto-asset, deposits as referred to in the DGSD are not regulated under MiCAR. Instead, the activity of accepting deposits from the public continues to be regulated by the CRD.” DGSD there stands for the EU’s deposit guarantee schemes directive, CRD for its capital requirements directive. In the same report’s comparison table the tokenised deposit carries deposit guarantee cover “under the same conditions as traditional deposits” and may bear interest; the e-money token under MiCA carries neither. For a treasurer with European subsidiaries that means a bank’s token remains a deposit with everything attached to it: limit, guarantee, interest. The stablecoin is a claim on an issuer and on that issuer’s reserve.

Pablo Hernández de Cos, General Manager of the Bank for International Settlements (BIS), explained the economic difference with an example at Jackson Hole on 28 August 2026. Ben holds USDT from Tether, Marie accepts only USDC from Circle; Ben has to sell his tokens in the secondary market and buy the others, and because deviations from par are, in his words, “the norm” and “sizeable under stress”, the payment may not go through at par: “There is no mechanism that enforces singleness.” Tokenised deposits, on the other hand, are account-based bank liabilities settled through central bank accounts in the background, and that settlement “preserves singleness”. Nellie Liang describes the same difference from the holder’s side: the stablecoin is a bearer instrument that travels from wallet to wallet without the issuer’s involvement; the tokenised deposit sits “on a ledger that a bank controls”. One instrument reaches everywhere and is guaranteed at par nowhere. The other is guaranteed at par and reaches as far as the bank’s boundary.

That contrast has the merit of clarity, but it reassures less than it should. The usefulness of the deposit hangs on a network that does not yet exist, and the reach of the stablecoin is precisely what makes it interesting to a head of payments. Whoever today wants to pay a supplier in Manchester who banks with Barclays gets nowhere with Wells Fargo’s token. A stablecoin gets there, at the price that the claim runs against an issuer’s reserve rather than a bank, and that someone at the end has to convert the tokens back into sterling. The comparison of the instruments therefore runs through the balance sheet: a claim that is insured and stays in-house, against one that travels anywhere and whose safety depends on a reserve.

Europe builds the same track with a different anchor

The German banking industry has been working on the same instrument for years under the name Commercial Bank Money Token, carried, according to Ledger Insights, by Commerzbank, Deutsche Bank, DZ Bank, Helaba and UniCredit, joined in 2026 by BNP Paribas and ABN AMRO in the shared sandbox and by industrial clients such as Siemens, BASF, Evonik and Mercedes-Benz in the pilots; the trade publication reported the first live transactions between Siemens and Evonik via DZ Bank and Commerzbank in May 2026. The difference from the American route lies in the anchor. With its trigger solution the Bundesbank has built a bridge between market platforms and the Eurosystem’s payment system; on 7 November 2024 Deutsche Bank and UBS used it to settle cross-bank payments based on tokenised deposits. Pontes, the Eurosystem platform meant to replace that bridge, is described in the analysis of Appia and Pontes on this site. In Europe settlement between banks is conceived in central bank money from the outset. The Clearing House connects its network to RTP and CHIPS, that is, to its own private systems; the release leaves open the question of when one bank’s claim on another is settled in central bank money. Zennon Kapron puts the question this way: when a Bank of America token sits in a Citi client’s wallet overnight, “who is exposed to whom, and when does the claim settle in central bank money?”

The stablecoin camp has its European counterpart too. Qivalis, the joint venture of 37 European banks for a euro stablecoin under MiCA, aims to launch in the second half of 2026, with DZ Bank as its only German member so far; Deutsche Bank is building a euro stablecoin of its own with AllUnity. Commerzbank and Deutsche Bank also sit in the dollar consortium of 21. Isabel Schnabel, member of the Executive Board of the European Central Bank, named the contest between the instruments openly in Seoul on 1 June 2026: “It remains to be seen whether, in such an environment, stablecoins can find their place in the financial system just as money market funds did 50 years ago, or whether other innovations, like tokenised deposits, will prove to be the more promising alternative.” That the question is put that way is news for European treasurers. The central bank considers both outcomes possible and is building the infrastructure that favours one of them.

The price of immediacy sits with the bank, not the client

What is an advantage for the treasurer, settlement on a Sunday and a change of bank at the push of a button, is, for the bank, a change to its liabilities. Rosie Levy and Srini Ramaswamy, economists at the Federal Reserve Bank of Dallas, quantified it on 25 August 2026: “Instant settlement would allow deposit holders who prioritize yield to switch banks almost instantaneously.” A 10 per cent reduction in the weighted average life of deposits would shrink the banking system’s capacity for maturity transformation by about USD 580 billion in ten-year equivalents; a 10 per cent increase in the rate sensitivity of deposits would reduce banks’ appetite for duration risk by about USD 700 billion. “Tokenization of deposits risks moving both these characteristics in adverse ways.” As evidence the authors cite Brazil, where heavier use of the Pix instant payment system increased banks’ demand for liquid assets and reduced credit intermediation.

That also explains why the banks are building the network regardless. The Treasury Borrowing Advisory Committee, the advisory body to the US Treasury, on 30 April 2025 classified USD 5.7 trillion of USD 8.3 trillion in US deposits as “at risk”, deposit types that could migrate to stablecoins given the right design. Brian Moynihan, chief executive of Bank of America, spoke on 15 January 2026, according to CoinDesk, of “the possibility of $6 trillion in deposits” flowing into stablecoins. Pablo Hernández de Cos adds the mechanism: if stablecoin reserves were held predominantly as wholesale deposits, retail funding would give way to “concentrated, more rate-sensitive wholesale liabilities”. A tokenised deposit keeps the funding in-house. A stablecoin takes it along. With the token, the banks are defending the liability side of their balance sheet.

Perhaps the gap closes faster than the timetables suggest; perhaps it stays open longer, because seventeen banks have to agree on one ledger and the consortium record argues against it. For the treasurer that changes little in the assessment. What arrives this autumn is an account on a different ledger and without opening hours. What may arrive in 2027 is a route to the next bank. And what the same banks are offering in parallel is an instrument that already travels that route today, at the price of not being a deposit. The contest between the three tracks will be decided in clients’ payment runs, not in San Francisco.

Recommendations

Glossary