Seventeen of the largest U.S. and international commercial banks announced on June 5, 2026 a joint initiative to build a tokenized deposit network operated by The Clearing House (TCH). The platform — targeting a first-half 2027 launch — will convert commercial bank deposits into on-chain tokens c...
"A regulated market-infrastructure solution for clearing and settling tokenized deposits — built on the same proven principles as core payment settlement — is essential." — Max Neukirchen, Global Co-Head of J.P. Morgan Payments
Seventeen of the largest U.S. and international commercial banks announced on June 5, 2026 a joint initiative to build a tokenized deposit network operated by The Clearing House (TCH). The platform — targeting a first-half 2027 launch — will convert commercial bank deposits into on-chain tokens capable of 24/7 clearing and settlement between member institutions, bridged to existing fiat rails including TCH's RTP real-time payments network and the CHIPS high-value wire system.
The initiative is a direct institutional response to two converging pressures: a $287 billion stablecoin market that settled $8.8 trillion in the first half of 2026, and a Jefferies research estimate projecting 3%–5% core deposit runoff from U.S. banks over the next five years. Bank of America CEO Brian Moynihan warned in January 2026 that as much as $6 trillion — roughly 31% of the $19.1 trillion U.S. commercial deposit base — could migrate to stablecoins if interest-bearing features were permitted. The GENIUS Act's prohibition on stablecoin yield has contained that scenario, but the deposit defense strategy is proceeding regardless.
A parallel effort — the Cari Network, backed by five regional banks on ZKsync infrastructure — targets a Q4 2026 retail launch, while JPMorgan's Kinexys platform already processes $7 billion in daily blockchain-based settlement. The combined effect: regulated bank money is moving on-chain at scale, not to embrace crypto ideology, but to prevent capital leakage to stablecoin issuers.
The Clearing House's "bank-led on-chain money initiative," announced June 5, 2026, involves 17 financial institutions: Bank of America, BMO Financial Group, BNY, Citi, Citizens Financial Group, Fifth Third Bank, HSBC, Huntington National Bank, JPMorgan, KeyBank, PNC Bank, Regions Bank, Santander (via Getnet Platforms), TD Bank U.S., Truist, U.S. Bank, and Wells Fargo.
The platform has two core components:
On-chain clearing and settlement layer. Tokenized deposits move between participating banks on a blockchain-based ledger. Settlement is available 24/7, eliminating the constraints of business-day-only processing windows that define legacy wire and ACH systems.
Fiat bridge layer. A connectivity mechanism links on-chain token activity to TCH's existing infrastructure — specifically, the RTP network (which processes real-time payments up to $1 million) and CHIPS (which clears approximately $1.8 trillion in daily wholesale payments). This bridge allows tokenized deposits to enter and exit the blockchain layer without creating a closed-loop system.
TCH President and CEO David Watson stated: "The banking industry has long provided the trusted infrastructure that underpins the movement of money throughout the global economy."
A blockchain vendor has not been selected. This is a notable gap for a system targeting H1 2027 production, though the focus on interoperability with existing TCH rails suggests the chain itself may function more as a settlement database than a general-purpose programmable layer.
Initial use cases center on multinational corporate treasury operations: programmable payments, real-time interbank liquidity transfers, cross-border settlement, and — notably — digital asset settlement, suggesting awareness that tokenized deposits must interact with broader on-chain capital markets.
The stablecoin market stood at $287 billion as of August 2, 2026, according to market data aggregators. Tether (USDT) holds 63.9% market share at $183.3 billion; USDC holds 25.1% at $72 billion. Combined, the two issuers control 89% of total stablecoin supply.
The deposit defense thesis rests on three data points:
Jefferies March 2026 research note. Analysts led by David Chiaverini projected 3%–5% core deposit runoff over five years. The intermediate-term risk, they wrote, stems from "emerging activity-based yield opportunities and payments use cases." The report estimated this runoff could cut average bank earnings by approximately 3%, as funding costs rise and fee income erodes. The most exposed institutions named: Wintrust Financial, Flagstar Financial, Webster Financial, Eagle Bancorp, and Axos Financial — all mid-cap banks with concentrated deposit franchises.
Moynihan's $6 trillion scenario. On Bank of America's Q4 2025 earnings call (January 14, 2026), CEO Brian Moynihan cited Treasury Department studies suggesting $6 trillion in deposits could migrate to stablecoins if they were permitted to pay interest. That figure represents roughly 31% of the $19.1 trillion in total U.S. commercial bank deposits as of May 2026 (per Federal Reserve data). Moynihan noted: "If you take out deposits, they're either not going to be able to loan or they're going to have to get wholesale funding, and that wholesale funding will come at a cost."
Stablecoin settlement velocity. Stablecoins processed $8.8 trillion in payment volume during H1 2026, according to industry data compiled by webthreepedia's prior research. That volume — dominated by B2B cross-border and institutional settlement — represents real competition for bank wire infrastructure, not speculative trading.
The GENIUS Act's prohibition on passive yield for payment stablecoin holders partially neutralizes the most acute deposit flight scenario. But the Act does not prevent stablecoins from being used as payment rails, and the velocity data suggests banks are losing transactional share even without yield competition.
A July 14, 2026 Dallas Federal Reserve paper by Dunbar, Sagnanert, Saretto, and Venkatu examined the structural distinctions between tokenized deposits and stablecoins. The core difference is balance-sheet treatment.
| Feature | Tokenized Deposits | Payment Stablecoins (GENIUS Act) | |---|---|---| | Issuer | Licensed commercial bank | Nonbank with federal/state charter, or bank subsidiary | | Backing | Bank balance sheet (loans, securities) | 1:1 reserve in low-risk liquid assets (Treasuries, cash) | | FDIC insurance | Yes, up to statutory limits | No | | Interest/yield | Permitted (bank deposit rules apply) | Prohibited under GENIUS Act | | Credit creation | Yes — deposits fund lending | No — issuers cannot make loans | | Regulatory framework | Existing banking law | New GENIUS Act requirements |
The Dallas Fed research noted that tokenization accelerates front-end transactions while leaving interbank settlement anchored in existing infrastructure like Fedwire. This is the design philosophy TCH's initiative follows: the token is a wrapper for regulated bank money, not a new form of money.
The regulatory asymmetry creates a strategic advantage for banks. Tokenized deposits can pay interest, fund lending, and remain on bank balance sheets — none of which GENIUS-compliant stablecoins can do. However, stablecoins hold advantages in programmability, composability with DeFi protocols, and frictionless cross-border movement without correspondent banking relationships.
The Brookings Institution published an analysis noting that the GENIUS Act's definition of "payment stablecoin" expressly excludes deposits, meaning tokenized deposits are not subject to the Act's restrictions. This deliberate carve-out gives banks regulatory space to tokenize without triggering new compliance obligations.
JPMorgan's Kinexys platform provides the strongest evidence that bank-issued deposit tokens can operate at scale. Key metrics as of mid-2026:
JPM Coin (JPMD) launched on Coinbase's Base network in November 2025, marking the first deployment of a major bank's native payment product on a public blockchain. The proof of concept, completed in June 2025, included transactions with B2C2, Coinbase, and Mastercard.
Kinexys is expanding in two directions. First, integration with the Canton Network — a privacy-focused enterprise blockchain — began in phases throughout 2026 for institutional asset settlement. Second, JPMorgan plans to issue a euro-denominated token (JPME), pending regulatory approval.
The platform's trajectory — from $5 billion to $7 billion in daily volume within six months — demonstrates that institutional demand for on-chain bank settlement exists independent of crypto market conditions. Kinexys volume grew during a period when the broader crypto market declined, with Bitcoin ETFs experiencing $5.4 billion in outflows.
While TCH's initiative targets wholesale institutional use, a parallel consortium of five regional banks is building the Cari Network for retail-facing tokenized deposits.
Participants: Huntington Bancshares, First Horizon, KeyCorp, M&T Bank, and Old National.
Infrastructure: The network selected Matter Labs' Prividium — a privacy-focused variant of ZKsync — as its blockchain layer. This choice reflects the need for transaction privacy in retail banking, where consumer data protection requirements exceed those in institutional wholesale markets.
Timeline: Q3 2026 pilot, Q4 2026 customer-facing launch.
Market position: Cari addresses the segment of the deposit base most vulnerable to stablecoin alternatives — retail customers who might otherwise use stablecoin-based neobank products or on-chain savings applications. The network keeps funds inside the regulated banking system while offering the instant settlement that stablecoin users expect.
The two-track approach — TCH for wholesale, Cari for retail — covers the full deposit spectrum. Neither initiative has disclosed fee structures or specific throughput targets, which will ultimately determine competitive viability against stablecoin rails that operate at near-zero marginal cost.
Complicating the bank-versus-stablecoin narrative is Open USD (OUSD), a consortium stablecoin launched June 30, 2026 by Open Standard — an entity led by Zach Abrams, cofounder of Bridge (acquired by Stripe in late 2024).
Open USD's 140-plus partners include Stripe, Visa, Mastercard, Coinbase, BlackRock, BNY, DBS, Standard Chartered, Google, and Shopify. Its economic model differs from USDC and USDT in two key ways: minting and redemption carry zero fees, and reserve income is shared with distribution partners rather than accruing solely to the issuer.
Stripe has designated OUSD as its default stablecoin for merchant transactions. Coinbase confirmed OUSD integration on Base.
The timing is not coincidental. Coinbase's $908 million annual revenue-sharing arrangement with Circle — under which Coinbase receives 100% of USDC reserve interest on platform-held balances and 50% elsewhere — comes up for renewal in August 2026. Coinbase's endorsement of Open USD triggered a 17% decline in Circle's stock price in a single week.
For the banking industry, Open USD represents a more formidable competitor than USDC or USDT alone. Backed by the payments infrastructure of Stripe and Visa, and the asset management credibility of BlackRock, OUSD has distribution advantages that pure-play stablecoin issuers lack. The banks' tokenized deposit initiative must compete not just against existing stablecoins, but against a consortium stablecoin with equivalent institutional backing.
Credit creation effects. If deposits migrate to stablecoins backed by Treasuries, credit creation capacity diminishes — stablecoin issuers cannot make loans. The Dallas Fed paper highlights this as the core macroeconomic concern. Tokenized deposits preserve the lending multiplier; stablecoins do not. From a Federal Reserve monetary policy perspective, this distinction matters.
Interoperability questions. TCH's network, Cari, and Kinexys currently operate as separate systems. No unified interoperability standard exists for tokenized deposits across different bank consortia. If tokenized deposits fragment into incompatible networks, they lose the primary advantage stablecoins offer: universal fungibility across chains and platforms.
Vendor selection risk. TCH has not selected a blockchain vendor. With a target launch of H1 2027, the technology decision must be made in the coming months. The choice will determine whether the system can interoperate with public chains (where stablecoin activity occurs) or remains a permissioned, bank-only settlement layer.
Regulatory arbitrage window. The GENIUS Act's yield prohibition gives banks a structural advantage. Should future legislation permit stablecoin yield — a possibility given midterm election dynamics in November 2026 — the deposit defense thesis weakens considerably. Banks are building infrastructure now while the regulatory window favors them.
The U.S. banking system is responding to the stablecoin challenge with infrastructure rather than rhetoric. Three concurrent initiatives — TCH's 17-bank wholesale network, Cari's regional retail platform, and JPMorgan's already-operational Kinexys — represent a coordinated attempt to keep deposits on bank balance sheets while adopting the settlement speed that stablecoin rails provide.
The question is not whether banks will tokenize deposits — that outcome now appears certain. The questions that remain: whether fragmented bank networks can match the universal fungibility of stablecoins, whether TCH can select a blockchain vendor and launch within 12 months, and whether the GENIUS Act's yield prohibition — the banks' regulatory moat — will survive the next legislative cycle.
The $19.1 trillion U.S. deposit base is the asset under contention. The $287 billion stablecoin market is the challenger. The banks have regulatory advantages and institutional trust. The stablecoin ecosystem has speed, programmability, and increasingly, its own institutional backers. The outcome will be determined by execution, not by which side has the better pitch deck.