SWIFT, the cooperative that connects 11,000+ financial institutions and facilitates over $150 trillion in annual cross-border payments, completed the design phase of its blockchain-based shared ledger on March 30, 2026. The minimum viable product (MVP), built on Hyperledger Besu with EVM-compatib...
"We're beyond experiments now. The question is how to scale—regardless of whether the instrument is a tokenized deposit, a CBDC, a stablecoin, or a tokenized fund." — Tom Zschach, Chief Innovation Officer, SWIFT
SWIFT, the cooperative that connects 11,000+ financial institutions and facilitates over $150 trillion in annual cross-border payments, completed the design phase of its blockchain-based shared ledger on March 30, 2026. The minimum viable product (MVP), built on Hyperledger Besu with EVM-compatible architecture, is scheduled to process live tokenized deposit payments before year-end 2026. More than 40 banks—including JPMorgan, HSBC, Deutsche Bank, Bank of America, Wells Fargo, and Standard Chartered—participated in the design. Over 25 institutions are expected to go live by mid-2026 across payment corridors spanning 11 countries.
The initiative represents a direct institutional response to the $312 billion stablecoin market, offering banks an on-chain settlement mechanism that preserves existing compliance infrastructure, deposit insurance protections, and correspondent banking relationships. Rather than integrating public blockchains or stablecoins, SWIFT chose a permissioned ledger that records and validates interbank payment commitments using tokenized commercial bank deposits—a model that keeps value creation inside the banking system.
The shared ledger MVP runs on Hyperledger Besu, an open-source, enterprise-grade Ethereum client maintained under the Linux Foundation's Decentralized Trust project. The architecture is EVM-compatible but permissioned—no native cryptocurrency, no public chain exposure, no gas fees paid in tokens. SWIFT developed the system in collaboration with ConsenSys, the firm behind MetaMask, Infura, and the Linea L2.
The ledger functions as a shared digital orchestration layer, not a full replacement for existing payment rails. Its operating model follows a two-step process:
Four enforceable standards bind every participating bank: fee certainty, full-value delivery, instant settlement where possible, and end-to-end traceability. The ledger builds on existing SWIFT messaging standards, including ISO 20022 compliance, and integrates with banks' current core systems rather than requiring wholesale infrastructure replacement.
This design choice is deliberate. SWIFT is not asking its 11,000-member network to adopt a new blockchain from scratch. Instead, it is extending its existing trusted orchestration layer with on-chain capabilities—a strategy that minimizes integration risk and leverages the cooperative's incumbent position.
The design cohort expanded from 30 institutions at the September 2025 Sibos announcement in Frankfurt to over 40 by the time the design phase concluded in March 2026. Named participants include:
The initial live corridors will cover payments to and from Australia, Bangladesh, Canada, China, Germany, India, Pakistan, Spain, Thailand, the United Kingdom, and the United States. This selection targets high-volume remittance and trade finance routes where the gap between legacy processing times (1–3 business days) and the ledger's near-instant settlement capability is widest.
According to SWIFT's own data, the cooperative processed an average of 49.8 million FIN messages per day in 2025, a 4.8% increase from 2024. The blockchain ledger's initial scope is cross-border payment settlement—a subset of total messaging volume, but one that accounts for the highest-value flows.
SWIFT's choice of tokenized deposits as the on-chain settlement asset—rather than integrating existing stablecoins like USDC or USDT—reflects a broader institutional preference emerging in 2026.
According to a March 2026 staff report from the Federal Reserve Bank of New York, tokenized deposits and stablecoins differ structurally in ways that matter to regulators: tokenized deposits remain liabilities of regulated banks, backed by deposit insurance (up to $250,000 per depositor in the U.S.) and subject to the full supervisory framework governing commercial banks. Stablecoins, by contrast, function more like narrow-bank instruments backed by reserves, operating outside the traditional deposit-taking and credit-creation system.
JPMorgan's Kinexys platform illustrates the scale already achieved through tokenized deposits. According to JPMorgan's 2026 milestones disclosure, Kinexys processes more than $5 billion daily and has surpassed $3 trillion in cumulative transaction volume since inception. The platform targets a $10 billion daily run-rate within the next 12 months.
Citigroup has taken a dual approach, expanding its Citi Token Services for tokenized deposits while simultaneously partnering with Coinbase in early 2026 to build stablecoin payment capabilities. According to Citi, corporate clients will demand access to both instruments.
The Brookings Institution noted in its analysis that corporations prefer tokenized deposits for large-value, low-margin institutional flows, in part because they avoid the reputational and regulatory baggage associated with stablecoins' use in illicit finance. However, stablecoins retain their utility in retail, remittance, and DeFi applications where 24/7 availability and permissionless access are priorities.
As of January 2026, the total stablecoin supply exceeded $310 billion. SWIFT's shared ledger does not directly compete with this market—it competes for the far larger pool of institutional cross-border settlement volume, where tokenized deposits are positioned as the programmable instrument of choice.
SWIFT's move intensifies competition across three fronts:
Ripple/XRP: Ripple's On-Demand Liquidity (ODL) service and XRP Ledger offer 3–5 second settlement times, with 93% of XRP-based cross-border payments settling in under 10 seconds, according to Ripple's data. However, Ripple's cumulative institutional payment volume stands at approximately $95 billion—orders of magnitude smaller than SWIFT's daily throughput. Notably, at least 30 of the 50+ banks in SWIFT's new framework maintain existing Ripple connections, including Santander, HSBC, Deutsche Bank, and Standard Chartered. Banks are operating in both ecosystems simultaneously rather than choosing between them.
Fireblocks: Fireblocks CEO Michael Shaulov publicly criticized SWIFT's approach, noting on LinkedIn that SWIFT has been conducting blockchain experiments since 2018 with limited production results. "You can technically browse the internet using a fax machine—but I haven't seen anyone doing it," Shaulov wrote. Fireblocks has processed over $10 trillion in digital asset transactions and counts BNY, Revolut, and Worldpay among its clients. The firm is building its own payments network targeting both fintechs and traditional financial institutions.
Public blockchain stablecoin rails: Mastercard's June 2026 announcement of stablecoin settlement across eight chains, and the broader growth of stablecoin payment infrastructure, represent an alternative path where value moves on public, permissionless networks. SWIFT's permissioned model offers compliance certainty and institutional control but sacrifices the composability and global accessibility of public chains.
JPMorgan Kinexys: While JPMorgan participates in SWIFT's shared ledger design, its proprietary Kinexys platform operates independently, processing $5 billion daily. This creates an open question about whether SWIFT's cooperative model can coexist with proprietary bank-operated blockchain platforms or whether the two will converge.
The economic architecture of SWIFT's shared ledger follows a pattern consistent with how value distributes in traditional financial infrastructure, with one significant difference: the orchestration layer moves on-chain.
Under the current correspondent banking model, cross-border payment fees typically range from $25 to $50 per transaction for retail flows, with corporate and institutional transactions negotiated individually. An IMF working paper from June 2025 valued the global cross-border payments market at approximately $1 quadrillion annually, with SWIFT-enabled institutions moving roughly $150 trillion of that volume.
SWIFT's ledger introduces tokenized deposits as the settlement instrument, but initial settlement still routes through conventional RTGS systems and correspondent relationships. This means the fee structure—and the economic value captured by intermediary banks—may not change materially in the MVP phase. The potential for disintermediation exists if on-chain settlement assets eventually replace conventional settlement, but that remains a roadmap item, not a near-term reality.
Banks that invested over $100 billion collectively in blockchain infrastructure between 2020 and 2024, according to FinTech Weekly's analysis, now face a question about returns. SWIFT's cooperative model distributes costs across members but also distributes the efficiency gains, potentially compressing margins rather than concentrating profits.
For the broader Web3 ecosystem, SWIFT's adoption of Hyperledger Besu—an EVM-compatible technology—represents an indirect validation of Ethereum's virtual machine as the standard execution environment for institutional finance. However, value accrual flows to the permissioned implementation, not to ETH holders or the Ethereum mainnet.
MVP scope is narrow. The initial deployment supports only tokenized deposit payments with conventional settlement. On-chain settlement, CBDC integration, and tokenized fund capabilities are roadmap items without committed timelines.
Permissioned architecture limits network effects. Unlike public blockchains where any participant can build and compose, SWIFT's ledger requires membership and compliance with cooperative standards. This ensures regulatory alignment but constrains the innovation surface.
Incumbent inertia risk. SWIFT has conducted blockchain experiments since 2018. Fireblocks' Shaulov and other critics point to a pattern of prolonged experimentation without production-scale deployment. The MVP's success depends on whether 2026 represents a genuine transition from pilot to production.
Settlement remains off-chain initially. The ledger records and validates payment commitments, but actual money movement still depends on existing rails. Until on-chain settlement is implemented, the ledger functions more as a coordination tool than a settlement system.
Competitive fragmentation. With JPMorgan running Kinexys independently, Mastercard integrating stablecoin settlement on public chains, and Ripple maintaining its own network, banks face a multi-rail environment with uncertain standardization.
SWIFT's shared ledger represents the largest coordinated effort by the traditional banking system to bring blockchain technology into production for cross-border payments. The scale—40+ banks, 11 country corridors, backed by a cooperative serving 11,000+ institutions across 200+ countries—dwarfs any existing blockchain payment initiative in institutional reach.
The strategic calculus is straightforward: rather than adopting public blockchains or stablecoins that would redirect value outside the banking system, SWIFT is building a permissioned layer that tokenizes the instruments banks already control—commercial deposits. This keeps the economic value chain intact while capturing the operational benefits of shared, real-time ledger technology.
Whether the MVP delivers on its 2026 production timeline will determine whether SWIFT's approach becomes the default institutional standard or joins the long list of bank blockchain pilots that never scaled. The 40+ banks that participated in the design now face the harder question: committing operational resources to live deployment. The data from that deployment will settle the debate that experiments cannot.