TechFlow Logo
Login/ Sign up
ETH Gas
Gwei
Fear
gas
Stablecoins Force Banks to Adapt as 24/7 Cross-Border Payments Arrive

Stablecoins Force Banks to Adapt as 24/7 Cross-Border Payments Arrive

2026.09.18
Share
SWIFT is finally living up to its name.

Written by: Simon Taylor

Translated by: Chopper, Foresight News

Banks are leveraging tokenized deposits to enable 24/7 movement of bank account balances without replacing their underlying, decades-old legacy core systems. With hundreds of millions of dollars in payment revenue at stake, these banks must defend this market, and they are already taking action.

On Saturday, September 5, DBS Bank and Citibank completed a US dollar cross-border transfer from Singapore to New York in minutes using tokenized deposits and a brand-new ledger for the Society for Worldwide Interbank Financial Telecommunication (SWIFT).

Yes, it is indeed SWIFT. A transfer completed on a Saturday, utilizing blockchain.

I used to find the name SWIFT somewhat ironic (since Swift originally means fast), but today it truly lives up to its name. Fairly speaking, bottlenecks usually lie with the banks themselves, not the SWIFT network; remember: SWIFT does not handle fund transfers, which are executed by individual banks.

These initiatives are currently still in pilot stages. But dismissing them simply as “just swapping out a database” misunderstands the real problems banks are trying to solve.

A single bank can upgrade its own database. But getting institutions like Citi, HSBC, and DBS to uniformly confirm claims and liabilities across different systems, countries, and time zones is an entirely different matter.

And blockchain happens to excel at solving exactly this kind of problem.

Corporate clients may not necessarily want to leave their existing banks. CFOs highly value the credit lines, FX conversion services, and guaranteed support during outages provided by their partner banks. They simply want their funds to be transferable over the weekend.

Tokenized deposits perfectly meet banks' need to deliver this capability. Understanding the logic behind it reveals what competition stablecoins face, and which scenarios remain uniquely advantageous for stablecoins.

Why Are Weekend Transfers So Difficult?

Bank deposits are essentially liabilities owed to you by the bank, recorded in a database. Cross-bank transfers mean sending instructions and synchronously updating records across multiple institutions. But these database updates cannot be completed simultaneously.

In practice, accounting operations are mostly processed only during business hours on weekdays. With differing time zones across countries, this makes weekend settlement impossible.

It can be compared to sending an email: the message arrives instantly, but staff only check it during working hours. In this analogy, a bank’s internal mainframe system acts as the person checking the email.

If the paying bank and receiving bank lack a direct relationship, intermediary correspondent banks are needed. This is correspondent banking: banks open accounts with other banks to execute payments on behalf of clients. Large correspondent banks also concurrently provide liquidity, FX conversion, and compliance screening services.

Below is a simplified US dollar payment flow where the paying and receiving banks share the same correspondent bank.

Each step has a cut-off processing time. Submitting even one minute late means the payment waits for the next batch. If Friday’s deadline is missed, the transfer rolls over to Monday.

Moreover, each step operates in a different time zone; 4 PM for one bank might be 4 AM for another.

Imagine if the transaction involves multiple currencies, several local payment systems, and more than one correspondent bank. Layered with each bank’s independent risk control rules, processing windows, and liquidity requirements, the complexity skyrockets.

SWIFT messages can arrive instantly, but once the system responsible for execution goes offline, transfers must queue. The same bank operating across different countries might run 10, 20, or even 30 mutually independent core banking systems.

Any solution must be compatible with these legacy systems. No one can force Jakarta’s mainframes to process transactions on a Sunday. Many will simply suggest: “Just upgrade the database, right?”

That’s what many assume. Banks themselves actually want to upgrade. But it’s far from that simple.

I once spoke with the chief architect of a major global bank. Roughly what he said was: We operate in over 60 countries, each deploying independent mainframe systems. The engineers who wrote most of the underlying code have long retired. For decades, this system has continuously adapted to local regulatory requirements and integrated with tens of thousands of peripheral systems. We can’t even fully map out the dependencies between modules.

Replacing a core banking system is extremely difficult. If you haven’t experienced it, search for real-world cases of failed bank core system migrations. This conversation made me truly grasp the sheer scale of this undertaking.

More than sixty independent systems. And the upgrades cannot interrupt ongoing banking operations.

His approach was to gradually reduce the bank’s reliance on the aging mainframe systems, decreasing the workload they bear. When introducing new features, there’s no need to trigger a global core system overhaul every time.

So the strategy isn’t to upgrade the core system, but to bypass it and build a parallel pathway.

Tokenized deposits generate corresponding bank deposit vouchers on a blockchain. They remain fundamentally bank liabilities. This service gives deposits an alternative transfer method: unconstrained by business hours, with programmable payment logic.

The new service won’t replace the original core banking system; instead, it runs in parallel, which industry insiders call a “sidecore.”

There are multiple technical approaches to tokenized deposits. Simply put: the bank holds client funds in segregated custody, generates corresponding tokenized balances. Clients can transfer this balance via the new service, while the bank ensures ledger consistency and prevents double-spending.

JPMorgan’s Kinexys, Citi’s tokenization service, and HSBC’s tokenized deposit solution all fall into this category. Corporate clients can move funds around the clock within supported markets and currencies.

What Clients Gain: Payments Completed on Saturdays

Picture this scenario: a supplier ships goods only upon receipt of payment. If the transfer cannot clear over the weekend, the company must either pay early tying up capital, borrow to bridge the funding gap, or simply wait. Delivery of goods is delayed accordingly. Corporate finance teams must plan liquidity based on settlement deadlines varying by currency and country.

Ant Group (its overseas Alipay business) has implemented this most deeply. It connected its treasury management platform, Whale, to HSBC, Standard Chartered, DBS, JPMorgan Chase, and at least five other banks’ tokenized deposit services. Ant can transfer US dollars, euros, Singapore dollars, Hong Kong dollars, and offshore RMB between Ant entities across these banks on any date and at any time.

This architecture integrates nine banks, meaning nine separate system integrations. Ant possesses sufficient engineering resources to handle development, but the vast majority of corporate finance departments do not have this capability.

Even for Ant, transfers can only occur between different Ant entities within the same bank. If Ant’s entity at HSBC needs to pay a DBS supplier on a Saturday, HSBC’s sidecore system must interoperate with DBS’s sidecore system.

Whose database should both parties use for accounting?

Tokenization Enables Synchronized Multi-Bank Reconciliation

A single bank’s sidecore can be built using a PostgreSQL database. Tokenization itself isn’t the only way to achieve weekend settlement; banks could completely use traditional databases to build additional payment services.

The real challenge lies in how this service collaborates with other banks’ systems.

Once Citi, HSBC, and DBS each launch this service, clients will naturally expect cross-bank mutual payments. Banks must confirm which payments are funded, which transactions are complete, and the mutual claims and liabilities between them.

If Citi hosts the ledger, HSBC must trust Citi’s database administrators; if SWIFT builds a traditional database ledger, all institutions must trust SWIFT. SWIFT would instantly become a bank, and arguably the world’s most important one.

The value of the SWIFT ledger is providing a shared record trusted by all banks and owned by none.

This is exactly what blockchain excels at. Blockchain itself doesn’t transfer funds; rather, it keeps ledgers synchronized among institutions that cannot fully trust each other or, for commercial reasons, cannot centrally host data.

Even if you trust a centralized ledger, blockchain’s characteristics remain highly valuable. Each bank holds cryptographic keys, requiring signatures for all instructions. No one can forge Citi’s payment commitments. It offers programmability: payments can be viewed as state machines with statuses like pending initiation, funded, and transaction complete. Smart contracts hardcode the rules, states only update when conditions are met, and each bank can independently verify them.

There is only one global ledger record. All banks see exactly the same data simultaneously, and no one can secretly tamper with it afterward.

The SWIFT ledger provides this shared record and status transition rule set. Cryptographic signing + state machine + multi-party consensus; these are precisely blockchain’s strength domains.

Example of the operational flow: You hold an account at Bank A, your supplier at Bank B, and it’s a Saturday.

You initiate a $100 payment instruction to Bank A. Bank A verifies your balance and authorizes the payment. It then sends a signed message to the shared ledger: Bank A owes Bank B $100 for payment to the supplier. The ledger verifies conditions and records this liability. Bank B sees the record in real-time and immediately credits the supplier’s account.

The supplier receives usable funds instantly, taking only minutes, without waiting until Monday.

The actual funds do not undergo cross-chain transfers. No tokens flow from Bank A to Bank B. Your deposit remains recorded at Bank A; the supplier’s new deposit is recorded at Bank B. What circulates on the ledger on Saturday are signed IOUs between the two banks. HSBC and Standard Chartered’s previously completed tokenized deposit transactions follow the exact same logic: claims and liabilities are recorded within each bank’s own systems, while SWIFT handles trade matching and net settlement.

Final Fund Settlement Occurs Only on Monday

Bank B credits the supplier before Bank A’s actual fund delivery. Bank B must either pre-allocate reserve funds (capital sitting idle awaiting potential weekend transactions) or extend short-term credit to Bank A, assuming the corresponding risk.

Interestingly, this aligns with the current SWIFT system’s operational logic. It’s just that even if banks wanted to, they couldn’t record this fund movement faster for clients.

By Monday, banks complete final fund delivery relying on existing traditional payment channels. If Bank A owes Bank B $100 and Bank B owes Bank A $80, both sides only need to settle the difference of $20. This is net settlement, a method banks greatly prefer, offering far higher efficiency compared to moving full amounts back and forth.

The ledger is responsible for recording fulfillment commitments and verifying transaction conditions, but does not provide the actual funds. The risk hasn’t disappeared; it’s merely recorded, cryptographically confirmed, and supports net processing.

This effectively upgrades correspondent banking with enhanced ledger capabilities. Not meant derogatorily, but stated objectively.

SWIFT is not the only institution positioning itself around this type of infrastructure. Clearing House, jointly controlled by JPMorgan Chase, Bank of America, Citigroup, and Wells Fargo, plans to launch a similar network in the first half of 2027. Some banks refer to it as a “bridge.”

Leveraging this architecture, banks’ internal 24/7 payment services can interconnect with other institutional systems, forming a clean, programmable, end-to-end workflow.

This is also banks’ tactic to counter the impact of stablecoins.

Why Not Use Stablecoins Directly?

Large enterprise clients maintain high acceptance of their partner banks. Not because they value the banks’ innovation or sleek interfaces, but because they rely on their reliability and ability to solve complex problems.

Consider a multinational corporation: its partner bank provides credit lines, cross-border treasury management, FX quoting, and dedicated personnel for issue resolution. The company likely doesn’t want to sever this relationship entirely, only improve the service experience.

This business carries massive revenue potential: Citigroup’s Treasury & Trade Solutions segment generated $21.3 billion in revenue in fiscal year 2025, accounting for roughly a quarter of Citigroup’s total revenue. This segment contributed $1.6 billion in new revenue, driving over a third of the group’s revenue growth.

JPMorgan Chase reported $19.3 billion in revenue for this segment in 2025, representing about one-tenth of the group’s managed revenue, an increase of $1.2 billion year-over-year, accounting for roughly a quarter of the group’s total revenue growth.

Fintechs like Airwallex and Nium, as well as global platforms such as Airbnb and Spotify, are prime clients that prioritize innovation. They generate massive cross-market payment flows, are highly sensitive to APIs and pricing, and cannot accept responses like “please come back on Monday.”

This has spurred startups like Augustus. It builds a 24/7 correspondent banking platform, expanding USD and EUR accounts into Latin America, Southeast Asia, and Africa.

Banks need to protect huge revenue streams; companies like Augustus target the profit margins within them. Augustus’s claim to build a native stablecoin core banking system is one competitive strategy.

I believe better technology and superior unit economics could steal a portion of customers, shifting them toward neobanks or fintech platforms like Airwallex.

Established banks’ counter-strategy is to enhance the value of existing bank accounts while continuing to provide the credit, FX, and ancillary services clients require. The speed at which banks are pushing forward implementation sufficiently demonstrates how seriously they take this competition.

Of course, the real world is highly complex, rarely presenting strict binary choices.

I often say that everyone in finance is both a competitor and a partner. The same client can utilize both solutions simultaneously.

Stablecoins offer unique value: as long as the recipient accepts tokens, you can pay individuals or businesses outside the traditional banking system. Their value is particularly pronounced in regions with inadequate bank coverage and high costs. However, adopting stablecoins does not replace the full suite of services enterprises obtain from banks.

The core difference lies in who holds the liability you possess.

Global platforms will most likely use both in tandem: keeping operating funds in banks and applying for loans, while simultaneously paying overseas suppliers with stablecoins. Suppliers can choose to hold stablecoins or convert them into bank deposits.

Funds No Longer Constrained by Bank Holidays

Your bank deposits will eventually run 24/7 and be available around the clock. It begins with internal connectivity within a single bank, then expands to multiple interconnected banks. Funds will circulate across institutions, currencies, and networks between bank deposits and stablecoins.

Funds need to move faster, cost less, and reach further than ever before.

Tokenization enables 24/7 fund movement and programmability, while stablecoins create an open payment ecosystem.

SWIFT finally lives up to its name. The remaining battlefield is competing for payment traffic.