Every bank needs a wallet
Most banks will never issue a stablecoin. Almost every bank will need to receive one.
JPMorgan’s Kinexys platform has now processed more than $3 trillion in cumulative transactions and is averaging over $5 billion in daily volume. JPM Coin, the bank’s USD deposit token, is live on Base for institutional clients, the first time a major US bank has issued a deposit token on a public permissionless chain. SoFi, a nationally chartered insured bank, launched SoFiUSD on Ethereum in December and announced a Mastercard settlement partnership in March. Nine European banks including ING, UniCredit, and CaixaBank are building a MiCA-compliant euro stablecoin for issuance in the second half of 2026. Six Swiss banks led by UBS launched a CHF stablecoin sandbox last month. Fidelity launched FIDD, its dollar stablecoin, on Ethereum in February. Stablecoin transaction volumes hit $33 trillion in 2025, up 72% year over year. The market cap is now over $321 billion.
The instruments are arriving at industrial scale. The GENIUS Act, signed in July 2025, created the first federal framework for payment stablecoin issuers. The OCC published proposed implementing rules in February. The OCC conditionally approved national trust bank charters for Circle, Paxos, BitGo, Ripple, and Fidelity Digital Assets in December. A consortium of ten global banks, Citi, Deutsche Bank, Goldman, UBS, BofA, MUFG, Barclays, TD, Santander, and BNP, is exploring jointly issued stablecoins pegged to G7 currencies.
And it’s not just stablecoins. Treasury funds passed $5 billion in market value in March; BlackRock’s BUIDL alone is over $2.8 billion. Franklin Templeton’s Benji crossed $1 billion. Citi Token Services and Partior are running tokenized deposits in production for institutional clients. In April, Securitize and Computershare opened the path for US-listed equities to be issued in tokenized form, sitting alongside the certificates Computershare already keeps for more than half the S&P 500. Tokenized real world asset value crossed $26 billion in March, four times the year before. The wallet that catches a stablecoin is the same wallet that catches all of this. Stablecoins are the leading edge, they are the tip of the spear as I’m fond of saying.
Sound money is the prerequisite, and you need the instruments before you need anything else.
But most banks are not going to be issuers. They are going to be receivers, and the vast majority have no way to receive a stablecoin when one arrives. At a panel at the Chicago Fed last October, a room full of senior bankers was asked how many had ever used a crypto wallet. Not a single hand went up. These are the people running the institutions where the stablecoins are going to land.
The issuance problem is being solved, by JPMorgan, SoFi, Circle, Tether, and a growing list of bank consortia. How those instruments actually reach and move through the other 3,900 banks in the US, the hundreds of banks in Europe, the thousands across Asia and Latin America, that part is still missing.
What credit cards already taught us
I’ve been thinking about how stablecoins are replaying the early credit card market. In the early 1960s, credit cards were American, dollar denominated, and concentrated among a handful of issuers. BankAmericard was a Bank of America product. Diners Club, American Express, and Carte Blanche were each single firms. You could have made every argument about credit cards in 1965 that people make about stablecoins today: it will stay an oligopoly, it will stay dollar dominated, network effects will entrench the early movers.
The prediction was right about the network layer and wrong about everything underneath it. Visa and Mastercard remain a duopoly. Visa has 4.9 billion payment credentials in circulation as of mid-2025, Mastercard has more than 3 billion. Underneath those two networks, thousands of financial institutions issue cards across every currency on earth.
What changed was where the concentration sat. Bank of America launched BankAmericard in Fresno, California in 1958 and began licensing it to other banks in 1966. A consortium of California banks, Wells Fargo, Crocker, United California Bank, and Bank of California, formed the Interbank Card Association the same year, which became Mastercard. In 1970, Bank of America gave up direct control of BankAmericard, transferring it to a cooperative owned by the issuing banks (renamed Visa in 1976). The structural move was to separate the act of issuing from the act of accepting. The networks consolidated. The issuers fragmented. When you walk into a department store in Tokyo and pay with an Indonesian debit card, the merchant doesn’t care about the issuer. The merchant cares about the network mark on the card. That’s where the trust sits. The acceptance network made the issuer interchangeable, and making the issuer interchangeable is what made thousands of issuers possible.
Stablecoins have no equivalent. The GENIUS Act regulates the issuer. The market argues over which token will win. Nobody has built the acceptance layer that makes the identity of the issuer irrelevant to the institution receiving the payment.
Today the right to redeem a stablecoin at par sits with a tiny set of authorized participants. Circle has 521 of them for USDC, Tether has 6 for USDT. Everyone else has to find a secondary market or a willing intermediary. That’s the acceptance gap, expressed in numbers.
The wallet is the acceptance layer
So what should a bank that isn’t JPMorgan or SoFi actually do?
Get a wallet. An address on one or more public blockchains, with the bank custodying the keys and the customer seeing the balance through familiar interfaces. The pragmatic move at this point is to treat it as a sidecar architecture, a parallel ledger that connects to the existing core through adapters. A multi chain wallet sits next to the bank’s legacy systems and lets it participate in tokenized money without forcing a full core replacement. It requires only the recognition that tokens on chains are going to be part of the financial plumbing, and that the institution needs a bridge between its existing core and that world. Getting connected matters more than being right about which chain or token wins. The financial outlay is modest compared to building a new payment rail or launching a stablecoin program, and the bank ends up covered against most of the plausible futures.
A bank joining a multi chain wallet network in 2026 is doing what a bank joining Visa in 1970 did. You didn’t need to be right about which cards would dominate. You needed to be inside the network when the volume came.
Barclays seems to have understood this. In January 2026, the bank invested in Ubyx, a clearing system for digital money designed to turn stablecoins and tokenized deposits into cash for receiving institutions. The framing from Barclays’ head of digital assets was the interoperability imperative. Infrastructure without interoperability is just an expensive silo. The right question is whether the institution can receive and process whatever arrives.
The ATM network economics
There’s a commercial case here that goes beyond hedging. Every foreign stablecoin that arrives at a bank’s wallet is an FX conversion opportunity. The bank catches the dollar stablecoin, converts it to local currency for its customer, and earns the spread.
The Visa and Mastercard networks didn’t just solve card acceptance at merchants. They also rewired global ATM economics. By the 1990s, every retail bank on earth had connected its ATMs to Visa Plus or Mastercard Cirrus. The reason was FX. Every foreign card that walked up to a local ATM produced an FX conversion the local bank captured, plus a network interchange fee. A US tourist pulling €200 out of a BBVA machine in Madrid generated FX margin and interchange that BBVA captured without underwriting the card or knowing anything about the issuer. The bank just had to be on the network.
That’s the wallet trade. The US is pumping dollar denominated stablecoins into every market on the planet. A bank in Singapore that catches a USDC transfer for its corporate client and converts it to Singapore dollars earns FX revenue on a flow that would previously have moved through correspondent banking rails where the margin was already compressed. Cross border payment flows exceeded $190 trillion in 2023 and are heading toward $290 trillion by 2030. Even a small fraction of that moving to stablecoin rails is real money for the institutions that can catch it.
The wallet pays for itself in FX revenue. Everything else is upside.
Receive before you issue
The instinct in banking is to own the liability, to be the issuer. That instinct is sound. Banks are balance sheets, and issuing is what balance sheets do. But for the roughly 3,900 US banks that aren’t JPMorgan, I think the right first move is to receive rather than issue.
A bank that issues a stablecoin today faces an immediate distribution problem. Where can the coin be accepted? Who can redeem it? What happens when it crosses a border? Without a mutualized acceptance network, the issuing bank has to build bilateral relationships with every receiving institution. A bank that implements a wallet and starts receiving whatever stablecoins arrive learns the operational reality first: custody, compliance screening, conversion, settlement. It builds the muscle. And it earns FX from day one. Issuance can come later, once the acceptance infrastructure exists to support it.
There's a structural reason the receiving bank wins. Stablecoins move the asset; they don't move the risk. Whoever catches the inbound stablecoin and provides the local currency deposit holds the redemption risk and provides regulated deposit insurance to the holder. Tokenization shifts which institution performs that function. The function remains. That's a banking function and it always has been.
There’s a deposit flight dimension too. The 820 million active crypto wallets that existed in 2025 are mostly unhosted, sitting outside the regulated financial system. If a bank’s customers want to hold stablecoins and the bank doesn’t offer a wallet, those customers will use MetaMask or Phantom or Coinbase. The deposits leave because the bank offered no alternative. The wallet is the minimum viable move to keep customers inside the perimeter.
Central bankers see the same thing from a different angle. If you’re the governor of Bank Negara Malaysia and foreign stablecoins are entering your economy, where do you want them to land? On unhosted wallets you can’t see, or on wallets provided by your regulated banks, where the flows are visible and conversion to local currency is automatic? Every central bank running that calculus arrives at the same answer.
Banks don’t need to pick a chain
Every ingredient already exists. The chains work, the instruments are being issued at scale, and the regulatory frameworks are forming. What’s missing is coordination.
The industry’s own go to market is making it worse. Every chain walks into a bank and says my chain is the best. Every issuer says accept my coin. Each pitch forces the bank to pick a winner. Banks hate picking winners in technology categories that are still moving. A chain evaluation can take three years inside a large institution. Multiply that by fifteen chains pitching fifteen different stories and the bank ends up doing nothing.
The better approach is to stop asking banks to choose and start asking them to connect. A multi chain wallet lets the competition between tokens and chains play out after the bank is connected, not before. Visa understood this in the 1960s. The smart move for a bank wasn’t to build its own proprietary card that only worked at merchants it had signed individually. The smart move was to join the network. The banks that became co-owners of Visa and Mastercard before the IPOs captured enormous value because they chose the collective infrastructure over the proprietary instrument.
The open question is whether the new tokenization consortia reach critical mass before fragmenting. Cari Network, Project Keystone, the nine-bank European group, the ten-bank G7 consortium are all placing bets, and not all of them will clear. The wallet move hedges that question. A bank with a multi chain wallet doesn’t need to predict which consortium wins. It just needs to be reachable from whichever does.
I think the same pattern is forming now. The institutions that move early to build acceptance infrastructure will own a piece of the plumbing through which trillions of dollars eventually flow. The ones still running chain evaluations will find the network was built around them while they were still in committee.
The instruments are here. The acceptance layer is the next move. The banks that get a wallet will be well positioned to catch what comes next.
References
Legislation and regulatory sources
GENIUS Act (S.1582), signed July 18, 2025
OCC proposed rulemaking implementing GENIUS Act, issued February 25, 2026
OCC conditionally approves five national trust bank charter applications — Circle, Paxos, BitGo, Ripple, Fidelity Digital Assets (December 12, 2025)
Market data and institutional activity
Stablecoin transactions rose to record $33 trillion in 2025, up 72% YoY (Artemis Analytics via Bloomberg)
Stablecoins reach $321B market cap as of May 3, 2026 (CoinSpectator); Stablecoin Market Cap Live Data (DefiLlama)
Cross-border payment flows and projections — exceeded $190 trillion in 2023, projected $290 trillion by 2030 (Visa)
Kinexys 2026 Milestones: Fund Flow, JPM Coin on Base, Leadership and More — $3 trillion cumulative, $5 billion daily average
JPMorgan launches JPM Coin (JPMD) on Base for institutional clients
SoFi launches SoFiUSD stablecoin (December 2025)
SoFi and Mastercard partner to enable SoFiUSD settlement across Mastercard’s global network (March 3, 2026)
Fidelity launches FIDD stablecoin on Ethereum (February 4, 2026)
Nine European banks building MiCA-compliant euro stablecoin — ING, UniCredit, KBC, Danske, DekaBank, Banca Sella, SEB, CaixaBank, Raiffeisen Bank International (September 25, 2025)
UBS, PostFinance, Sygnum, Raiffeisen, ZKB, BCV launch CHF stablecoin sandbox (April 8, 2026)
Ten global banks explore G7 stablecoin consortium — BNP Paribas, Santander, BofA, Barclays, Citi, Deutsche Bank, Goldman, MUFG, TD, UBS (October 2025)
Barclays invests in Ubyx to advance digital money connectivity (January 7, 2026)
Tokenized assets beyond stablecoins
Tokenized Real-World Asset Value Jumps Fourfold to $26 Billion (PYMNTS, March 2026)
BlackRock BUIDL Tokenized Treasury Fund — over $2.8B AUM (Securitize)
Securitize and Computershare announce agreement to enable tokenized shares for U.S. issuers (April 29, 2026)
Securitize, Computershare open path for $70 trillion in U.S. stocks to move onchain (CoinDesk)
Card network data
Visa Fiscal 2025 Annual Report — 4.9 billion payment credentials in circulation
Mastercard cards in circulation — over 3 billion active cards globally
Visa Inc. Wikipedia — BankAmericard 1958, NBI cooperative 1970, renamed Visa 1976
Mastercard Wikipedia — Interbank Card Association formed 1966 by Wells Fargo, Crocker, United California Bank, and Bank of California