SWIFT launched a shared ledger for tokenized deposits in July 2026, connecting 17 global banks across six continents including ANZ, BNP Paribas, BNY, Citi, DBS, First Abu Dhabi Bank, FirstRand, HSBC, Itaú Unibanco, Lloyds, Mashreq, MUFG, OCBC, Standard Chartered, UBS, UOB and Wells Fargo. The network has successfully completed its design phase and announced a live scheme by June 2026 based on the first minimum viable product (MVP) of the ledger. According to SWIFT, the ledger, built on Hyperledger Besu in nine months, validates and coordinates tokenized-deposit movements between member banks around the clock with final settlement through existing rails. The system represents a significant milestone as the banking industry's response to the $320 billion stablecoin sector, offering a digital dollar that maintains all the properties of traditional bank deposits while providing blockchain-based settlement capabilities.
JPMorgan's Kinexys platform is already settling institutional payments with tokenized deposits, having extended onto public infrastructure including Base and the Canton network. As reported by crypto.news, the single-bank model proves the concept at the only scale that matters - real money, real clients. However, the structural limit of this approach is reach: one bank's token moves one bank's money, and every large bank running its own rail recreates the fragmentation problem that correspondent banking exists to solve.
SWIFT's new cross-border retail payments framework launched in June 2026 with over 25 participating banks including Lloyds Bank, NatWest, NAB, Societe Generale, ANZ, HSBC, JPMorgan, Deutsche Bank, Standard Chartered, and Bank of America. The framework enables instant settlement, 24/7 transaction processing, and interoperability between banks' tokenized deposits across Australia, Bangladesh, Canada, China, Germany, India, Pakistan, Spain, Thailand, the UK and the US. According to SWIFT, participating banks can process cross-border payments in under a minute for inbound transfers from countries like Australia, Brazil and Turkey. Projections indicate that approximately 50 banks from more countries are expected to join by year-end, creating a rapidly expanding network for enhanced retail transfers.
The SWIFT ledger runs an Ethereum Virtual Machine (EVM)-compatible architecture built on Hyperledger Besu, with cross-chain interoperability handled by Chainlink's Cross-Chain Interoperability Protocol (CCIP), which Swift moved from pilot to production in November 2025. According to Payment Expert, choosing EVM compatibility over a bespoke permissioned design is a concession to where developer tooling and institutional custody integrations already exist. The architectural choice matters significantly: this is an orchestration layer, not a settlement network, where banks issue tokenized deposits on their own ledgers while SWIFT coordinates movement between them, leaving deposit liability exactly where it started. This design approach is crucial for institutional buy-in, as it avoids the counterparty risk associated with settlement networks while maintaining the practical value of extended trust and stability.
As reported by crypto.news, three architectures now compete in the tokenized deposit space: the single-bank model (JPMorgan's Kinexys), the shared-network model (SWIFT's ledger), and the public-facing frontier. A parallel American effort through The Clearing House, backed by JPMorgan, Bank of America, Barclays, and BNY, targets a 2027 launch, meaning even the shared-network lane already has competing networks. The interoperability trap represents a significant risk, as every architecture mints tokens that work within its own perimeter, with history showing that merchants and users refuse to hold seventeen incompatible instruments. According to crypto.news, the tokenized deposit-versus-stablecoin contest is a fight over the two-tier monetary system, where central banks serve banks and banks serve everyone else. The likeliest outcome is partition, with banks holding the regulated core and stablecoins holding the open edge, as the boundary is contested by regulation as much as preference.
