Jamie Dimon spent the spring savaging stablecoins on television, warning that banks “will not accept” the new rules¹ and conceding, in the same breath, that they might pass anyway. It made for good clips. It also obscured what JPMorgan was doing while he talked: running more than $5 billion a day through Kinexys², its own tokenised deposit platform, and shipping a yield-bearing token on a public blockchain. Over the same year, stablecoins moved roughly $390 billion in genuine payments, up 733%³. The digital dollar has split into two rails, and the loudest banker in the room is quietly betting on both. 

A year ago, when we last looked at stablecoins in cross-border payments, the question was whether they would displace banks. The answer turned out to be neither yes nor no. The market split. Stablecoins took the long tail: remittances, SMEs, emerging-market corridors, merchant acceptance. Tokenised deposits won institutional flow. And the most interesting fight isn’t happening at the issuance layer where the headlines sit. It’s one layer up, where Stripe, Mastercard, Visa, and even SWIFT are converging from different directions to control the connective tissue between the two rails. 

Rather than replacing each other, stablecoins and tokenised deposits are evolving into two separate rails. Stablecoins are finding their place in retail and cross-border use cases, while tokenised deposits are increasingly serving institutional flows. Payments executives still framing this as a binary (banks versus crypto, SWIFT versus stablecoins) are reading the wrong map. 

The twelve-month scoreboard 

Several developments since July 2025 hardened the picture. 

Regulation set the rails. The GENIUS Act was signed into law on 18 July 2025, with five federal rulemakings from the OCC, FDIC, NCUA, Treasury, and FinCEN/OFAC following in early 2026. MiCA has applied in full since December 2024, with the transitional window for existing crypto-asset service providers closing on 1 July 2026. The UK joined the regulating jurisdictions: the Bank of England’s consultation on sterling systemic stablecoins closed in February 2026, proposing a 40% central-bank-reserve and 60% gilts backing split with no holder interest, and final rules due in the second half of 2026. Hong Kong awarded its first two stablecoin licences on 10 April 2026, to HSBC and to a Standard Chartered/Animoca/HKT joint venture. Brazil banned the use of stablecoins to settle cross-border eFX from October 2026, a useful reminder that “regulatory clarity” cuts both ways. 

The yield question hasn’t gone away. The CLARITY Act, which governs the broader digital asset market structure, cleared the Senate Banking Committee in May 2026 carrying a compromise on the one issue that matters most here: it bars yield that is economically equivalent to a bank deposit while permitting rewards tied to genuine activity. That forces a shift from a buy-and-hold reward model to a buy-and-use one, and it is the closest thing yet to a federal answer on whether the wrapped-yield workarounds (Aave, Morpho, Pendle, tokenised money market funds) survive. For payments executives, the practical question is whether corporate treasury clients will be able to earn yield on stablecoin balances by 2027, and on what rails. The answer reshapes the competitive position of tokenised deposits, which can pay interest natively. 

The banks are fighting it, and floor passage is still a coin toss. But watch what they do, not what they say: even as the lobbying intensifies, the largest US bank is running billions a day through its own tokenised deposit rail. 

Banks moved decisively on tokenised deposits. JPMD went live on Base in mid-2025. Citi Token Services is running real-value transfers between New York, London, and Hong Kong. Hong Kong’s Project Ensemble went live in November 2025 with seven banks. Société Générale-FORGE issued the first major-bank USD stablecoin, USDCV, with BNY as custodian. 

PSPs and networks bought orchestration. Stripe completed its $1.1 billion acquisition of Bridge in February 2025. Mastercard agreed to acquire BVNK for up to $1.8 billion in March 2026, the largest stablecoin deal to date. Visa’s stablecoin settlement hit a $3.5 billion annualised run-rate by November 2025⁴. Western Union announced its own stablecoin USDPT for May 2026 launch. 

The Treasury Secretary framed the politics. On the day the GENIUS Act was signed, Scott Bessent said stablecoins would “buttress the dollar’s status as the global reserve currency, expand access to the dollar economy for billions across the globe, and lead to a surge in demand for US Treasuries.” That isn’t a crypto policy statement. It’s a dollar policy statement, and it’s the most important thing anyone in the US government has said about cross-border payments in a decade. 

Figure 1: Quarterly developments in stablecoins and tokenised deposits, Q1 2025 to Q2 2026

Two rails, plainly 

Stablecoins are bearer instruments on public blockchains. The holder owns the token; the issuer holds the reserves, mostly in US Treasuries. Settlement is final, 24/7, and programmable. There’s no deposit insurance, and under the GENIUS Act, no yield. The product fits people and businesses whose problem is their local currency or their lack of access to good banking, which is why adoption skews heavily towards emerging markets. Tether and Circle together hold more than 85% of the market⁵. Tether alone is now the 17th-largest holder of US debt⁶. 

Tokenised deposits are bank liabilities recorded on permissioned ledgers. The holder has a claim on a specific bank, deposit insurance applies, yield is permitted, and settlement runs bank-to-bank on consortium infrastructure like Kinexys, Citi Token Services, or Project Ensemble. The product fits regulated institutions moving large value, which is why adoption skews towards developed markets. 

These aren’t competing products. They’re complementary rails serving different customers. The BIS made the analytical case directly in its June 2025 Annual Economic Report: stablecoins fail the “singleness of money” test that tokenised deposits pass⁷. In an EM remittance corridor where the local currency is the actual problem, that failure is a feature. In institutional treasury where same-day finality at par with central bank money is the requirement, it’s a deal-breaker. 

Think of it as retail dollar and wholesale dollar, each going digital on its own timeline

Four layers, one prize 

The cross-border stack now divides cleanly into four layers, and the economics aren’t evenly distributed across them. 

Figure 2: Four-layer cross-border payments stack

Issuance is becoming a low-margin utility. Tether and Circle dominate, with combined supply north of $260 billion⁸. Bank-issued stablecoins are a regulated flank rather than a frontal assault: SocGen’s USDCV, HSBC and Standard Chartered’s Hong Kong tokens, MUFG/SMBC/Mizuho’s joint Japanese exploration, and the WSJ-reported consortium of JPMorgan, Bank of America, Citi, and Wells Fargo. Europe’s answer arrived in the form of Qivalis, an Amsterdam consortium of 37 banks across 15 countries (ABN AMRO, BNP Paribas, ING, UniCredit, and CaixaBank among them) building a MiCA-compliant euro stablecoin for a second-half 2026 launch. JPMorgan’s Umar Farooq put it crisply earlier this year: proliferating stablecoins “are going to look more like loyalty points than money.”⁹ Reserve requirements under GENIUS and MiCA squeeze the economics further. Issuance is unlikely to be where the value accrues. 

Settlement chains are commoditising. Ethereum, Solana, Tron, and Base host the bulk of public stablecoin flow. Tron dominates USDT volume globally and remains the workhorse for EM remittances. Permissioned bank chains host tokenised deposits. Circle’s Arc Layer-1 and SWIFT’s blockchain ledger (announced at Sibos Frankfurt in September 2025 with ConsenSys and more than 30 banks) are the two strategic moves that matter. Both are attempts to claim the institutional settlement substrate. But chains are infrastructure. The value is unlikely to accrue here either. 

Orchestration is the prize, and this is where the recent M&A premiums have landed. Stripe paid $1.1 billion for Bridge. Mastercard paid $1.8 billion for BVNK, outbidding Coinbase, which Fortune reported came close to a roughly $2 billion deal of its own. Visa is building stablecoin settlement infrastructure across multiple chains and stablecoins, and is a design partner for Circle’s Arc. Fireblocks, Zerohash, and Circle’s CPN are positioning here. Even SWIFT is repositioning itself as an orchestration layer rather than a messaging-only utility. 

Mastercard’s Chief Product Officer Jorn Lambert framed the rationale for BVNK plainly: most financial institutions and fintechs will provide digital currency services “be it with stablecoins or tokenised deposits,” and Mastercard wants to support them with “a best in class, highly compliant, interoperable offering.” The operative phrase is “stablecoins or tokenised deposits.” Mastercard didn’t pay $1.8 billion to issue a stablecoin. They paid to be the connective tissue between the two rails. 

Why orchestration takes the margin: it’s where FX gets executed, where compliance gets bundled, where programmability gets monetised, and where routing decisions between rails happen. It’s also where serving EM corridors (Bitso, Yellow Card, Bridge) and serving DM institutional flow (BVNK, Fireblocks, Fnality) are starting to specialise. The acquirers are buying both. 

Distribution is where customer relationships still live. Stripe is rolling stablecoin acceptance to Shopify merchants in 34 countries. PayPal extends PYUSD through Xoom and Yellow Card. Banks distribute tokenised deposits to corporate treasury clients through existing relationships. Bitso, Nubank, and M-Pesa distribute stablecoins to EM consumers. Distribution matters, but it isn’t sufficient on its own. Western Union and MoneyGram had distribution. What they lacked was a position in the layers above. 

Four provocative reads 

This is a US dollar story, not a crypto story. Roughly 99% of stablecoin supply is USD-denominated. The euro stablecoin market sits below $700 million against more than $310 billion for USD coins¹⁰, even after 37 European banks lined up behind Qivalis to build a credible euro alternative. Tether’s $135 billion in US Treasuries¹¹ makes it a larger holder than most sovereigns. Bessent’s framing makes the policy intent explicit. The bank consortia forming on both sides of the Atlantic are competing to issue the digital dollar and the digital euro, but the demand is overwhelmingly for the dollar. Treating this as digital-asset strategy rather than dollar-ecosystem strategy risks being a category error. 

Banks didn’t lose. They split the playbook. Tokenised deposits for institutional cross-border, bank-issued stablecoins for the regulated public-chain flank. JPMorgan is the tell: it fights yield-bearing stablecoins in Washington while shipping JPMD, a yield-bearing deposit token, on Coinbase’s Base chain. That isn’t a contradiction. It’s the split playbook in miniature, defending the deposit base in one venue and building the tokenised-money flank in another, keeping both options open until the rules settle. Banks don’t appear to be the main losers. The pressure seems to be shifting instead to the standalone middleware in the correspondent banking stack and the remittance specialists who controlled distribution but never owned orchestration. 

The disruption narrative was always wrong. The substitution narrative is right. Stablecoins aren’t replacing correspondent banking, they’re replacing its float. Not replacing card networks, replacing the settlement leg. Not replacing remittance brands, replacing the FX margin. The economics under threat are specific: nostro float, settlement spreads, FX markup. If your P&L depends on those, the next 24 months will be uncomfortable. 

Tokenised deposits look like the quiet institutional winner. The bullish $2 to 4 trillion stablecoin forecasts from Citi and Standard Chartered¹² overstate the institutional opportunity because most large-value cross-border flow will move on tokenised deposit rails. JPMorgan’s more conservative $500 to 600 billion projection is the better directional anchor for stablecoin stock. 

What this means for Payments Executives 

Tier-1 banks. The choice isn’t whether to engage but which lane to commit to, and there are now three visible templates. The first is bank-issued stablecoin: Société Générale-FORGE issued USDCV on public chains with BNY Mellon holding the reserves, while HSBC and Standard Chartered took licences in Hong Kong to issue HKD-pegged tokens. This route puts your brand on the public-chain rail and reaches the customers who use it, namely EM corporates, fintechs, merchant ecosystems. The second template is the tokenised deposit, the JPMorgan and Citi approach: keep the liability on your own balance sheet, settle on a permissioned ledger, and serve institutional clients who need insured, interest-bearing, par-value digital cash. The third is white-label infrastructure: BNY Mellon custodies reserves for SocGen’s USDCV without issuing anything itself, taking fees from the orchestration layer without taking issuer risk. Drift (waiting another year to decide) is now the most expensive option, because each of these positions is being claimed by your competitors as you read this. 

Card networks and PSPs. Mastercard paid $1.8 billion for BVNK. Stripe paid $1.1 billion for Bridge. Coinbase reportedly came within a final round of a $2 billion BVNK deal before losing it. The orchestration land grab is real and the price is rising. If you’re not already a buyer or a partner of a stablecoin orchestration platform, you’re behind, and the available targets are thinning. 

Corporate treasurers. Two things in the next twelve months. Build dual-rail readiness: at minimum, the ability to send and receive in USDC or USDT on at least one corridor where it materially beats your current bank rails, and the ability to settle large-value flows on a tokenised deposit network if your bank offers one. Then pressure-test your cross-border FX margin against a scenario where 25% of your B2B flow migrates to stablecoin rails. The answer changes how you negotiate your next banking RFP. 

Remittance and FX specialists. Western Union and MoneyGram have set the floor: either issue your own stablecoin (Western Union’s USDPT, launching May 2026) or integrate one (MoneyGram with Circle and Stellar). The ceiling is owning a corridor end-to-end (distribution plus orchestration plus a stake in the issuer), which is the position Bitso has built in LATAM and Yellow Card in Africa. Anything between those two points risks being a strategic dead zone. 

EM fintechs. The corridor expertise built over the last two years is now genuinely valuable, and the acquirers are visible: card networks, PSPs, and increasingly Tier-1 banks looking for instant geographic reach. Decide now whether you’re building to be acquired or to scale independently. The answer determines how you raise, how you partner, and how you price your services to enterprise customers. 

The open question for 2027 isn’t who issues the most stablecoins or whose tokenised deposit network clears the most institutional volume. It’s who controls the orchestration layer that connects them. The answer determines who keeps the cross-border revenue pool. The bidders aren’t all in yet.

***

Author: Martin Wallraff, Director, London, Payments Consulting Network   

Martin brings over 18 years of experience in financial services strategy consulting, both with external consultants and internal strategy teams. For the past decade, he has specialized in payments, fintech, and transaction banking, advising clients on strategy, market entry, product development, and cost management. His extensive international expertise includes leading projects worldwide and residing in various locations across Europe and Southeast Asia. 

***

To get notified of our latest posts, follow the Payments Consulting company LinkedIn page, and click on the bell icon at the top right section of our company profile.