Signal Scanner · FINANCIAL SERVICES & FUTURE OF MONEY

Deposits, Not Stablecoins: How Rival Bank Consortia Are Rebuilding the Correspondent Hierarchy On-Chain

As policymakers argue over stablecoins, the working US answer to on-chain money is the tokenised bank deposit. Two rival bank consortia and a hardening deposit-insurance perimeter are rebuilding banking's correspondent tiers on-chain, with community banks most exposed.

The story policymakers keep telling is about stablecoins: the GENIUS Act gave private dollar tokens a federal rulebook in 2025, and the debate since has been whether they will drain bank deposits. That misses what US banks and their regulators have been building. The working answer to on-chain money is the tokenised bank deposit, and the contest is no longer banks against stablecoins but banks against each other. Two rival consortia are forming, and a deposit-insurance perimeter is hardening around them. The inflection runs through late 2026 and 2027: who sets the rails the rest of banking must plug into?

Signal Identification

This is a structural shift with a regulatory pivot at its core, not a product launch. It is the moment the US chose deposits over stablecoins for programmable money and organised that choice into competing bank networks. The signal is not that tokenised deposits exist; it is that their governance is splitting along the same tiered lines as the correspondent-banking system they are meant to modernise.

Time horizon: 2–4 years (regional network go-live targeted end-2026; money-centre network targeted H1 2027; interoperability contest 2027–2029) Plausibility band: Medium–High Geographic / Jurisdictional Scope: United States primary (GENIUS / FDIC perimeter and US bank consortia); spillover to cross-border correspondent banking and jurisdictions benchmarking the US model (EU, UK, Hong Kong) Sectors exposed: regional and community banks; payments and clearing (CHIPS, RTP, correspondents, cards); corporate treasury; stablecoin issuers; core-banking and ledger technology providers

What's Changing

The perimeter came first. The FDIC's proposed GENIUS rule (Federal Register, 10/04/2026) treats any tokenised product that meets the statutory definition of a deposit as a deposit, insured like any other, while cash backing a stablecoin gets no pass-through insurance to holders. The Bank for International Settlements pushed the same way (BIS, 23/06/2026): stablecoins fail the test of "singleness", and tokenisation belongs inside the two-tier bank and central-bank system, not around it.

Then the market split. On 5 June the largest US banks — JPMorgan, Citi, Bank of America, Wells Fargo and thirteen others — said they would clear and settle tokenised deposits through The Clearing House, bridged to CHIPS and RTP, targeting a first-half-2027 launch (ABA Banking Journal, 05/06/2026). Months earlier, five regional lenders including Huntington, KeyCorp and M&T had teamed up through the Cari Network for a year-end 2026 rollout that would at first stay within their own customer base (Banking Exchange, 23/02/2026).

The scale gap explains the urgency. US stablecoins stood near $280bn at the end of 2025 and JPMorgan's deposit token already clears more than $7bn a day, yet interbank settlement of tokenised deposits does not yet exist (Brookings, 14/04/2026). New York Fed research finds tokenised deposits raise welfare wherever bank credit matters and risk-shifting is contained (New York Fed, 01/02/2026).

The 2026 pivot to tokenised deposits and the two-network split

Feb 2026 Cari formed Apr 2026 FDIC perimeter Jun 2026 TCH net + BIS End 2026 Cari go-live H1 2027 TCH go-live

Sources: Banking Exchange; Federal Register / FDIC; ABA Banking Journal; BIS.

Disruption Pathway

The pathway runs in three stages. Through 2026 the perimeter is set and the two consortia move from announcement to build: the Cari network pilots in the third quarter and aims to go live by December, while the money-centre group readies its Clearing House network for 2027. Across 2027 two production networks run in parallel, each mostly closed and moving deposits among its own members — the "cash islands" problem that has kept tokenised deposits intrabank so far (Brookings, 14/04/2026). The decisive stage is interoperability: whichever network bridges cleanly to the incumbent CHIPS and RTP rails becomes the hub the others connect through, inheriting the correspondent function on-chain.

Stress concentrates in three places: community and smaller regional banks, which Brookings notes have the least capacity to build their own token and so risk depending on a larger rival's rails; corporate treasurers facing deposit tokens that do not move freely between banks; and the fee pools tied to correspondent and card settlement as money moves around the clock. Three adaptations follow — vendor platforms such as FIS's Lyriq extending issuance to the long tail (PYMNTS, 05/06/2026), regulators pressing the consortia for common standards, and banks repricing treasury economics as continuous settlement erodes overnight float.

Why This Matters

For the boards this reaches — regional and community bank directors, and the corporate treasurers who bank with them — the question is no longer whether to hold a digital-money strategy but which network to sit inside. Building alone is out of reach for most; joining the money-centre network means accepting terms set by the largest banks; a regional consortium or vendor rail trades some independence for reach. The choice made over the next few quarters will shape deposit economics and competitive standing for years, because payment network effects are sticky and late entrants inherit the terms first movers set. Treasurers should press their banks now on interoperability and the insured-deposit status of any token offered.

Decision-action posture for this signal: Prepare — the networks are being built and the perimeter set this cycle, but commercial rollout and the decisive interoperability contest are still twelve to twenty-four months out.

Counter-Argument

The strongest objection is that this tiering never bites, because stablecoins keep the lead and the two forms coexist. Stablecoins run on open networks, already move real volume, and a Brookings-cited survey found more than half of corporate non-users expect to adopt them within a year (Brookings, 14/04/2026). PYMNTS reads the market as both-and rather than winner-take-all (PYMNTS, 05/06/2026), and a Fed "skinny" master account could give stablecoin issuers direct rail access that narrows the gap.

Even so, coexistence does not dissolve the hierarchy; it relocates it. Whoever controls the deposit network and its bridge to CHIPS and RTP intermediates the on-chain dollar however much stablecoin volume grows. The RTP-then-FedNow sequence — large banks in 2017, smaller ones only by 2023 — suggests that access gap is durable, not transient.

Implications

On the available evidence this looks like durable change, not a passing experiment. The perimeter is written into a federal rulemaking and two consortia have committed capital to dated rollouts; that combination rarely reverses. Oxford's Ross Buckley argues tokenised deposits will likely prove the most durable form of digital money because they carry the banking system's prudential weight (Oxford Business Law Blog, 26/03/2026). The inflection window is roughly eighteen months. Banks that set or interoperate with the winning standard, and the vendors bridging the long tail, gain; those renting a competitor's rails lose margin and control of their deposits.

Early Indicators to Monitor

Disconfirming Signals

Strategic Questions

Keywords

Tokenised deposits; stablecoins; GENIUS Act; FDIC deposit insurance; The Clearing House; CHIPS; RTP; Cari Network; correspondent banking; interbank interoperability; programmable money; community banks

Bibliography

Source tiers: Tier 1, governments, regulators and intergovernmental bodies. Tier 2, think-tanks, academic institutes, major consultancies and quality data providers. Tier 3, quality journalism and specialist trade press. Tier 4, vendor, company and practitioner sources, used only as directional corroboration.


Prepared by Shaping Tomorrow: 13 July 2026