For a decade, tokenised money was mostly a proof of concept. That has changed. Central banks have settled real value on shared ledgers, large banks run deposit-token platforms processing billions of dollars a day, and the United States has a federal statute for payment stablecoins. Three forms of digital money now compete and, increasingly, need to interoperate: tokenised commercial bank deposits, private stablecoins and central bank digital currencies (CBDCs).
The Bank for International Settlements has been explicit about its preference. In its 2025 Annual Economic Report it concluded that stablecoins "perform poorly" against the three tests for serving as the mainstay of the monetary system (singleness, elasticity and integrity), and argued that a unified ledger bringing together tokenised central bank reserves, commercial bank money and financial assets can harness tokenisation's full benefits1.
Wholesale settlement is where tokenisation is proving itself
Project Agorá, led by the BIS with seven central banks and more than 40 private financial firms, reported in May 2026 that its prototype showed atomic, all-or-nothing settlement of cross-border payments using tokenised central bank reserves and tokenised commercial bank deposits is achievable securely, and that work would advance to real-value testing2. By July, 22 financial institutions and five central banks had completed 30 real-value transactions worth about CHF 800,000 across six currencies, with an average of about 80 seconds from initiation to settlement, though the platform was not yet integrated with central bank or core banking systems3.
Commercial deployments are further along. One large US bank reports that its blockchain platform has processed more than $3 trillion since inception and averages more than $5 billion a day, and that its US dollar deposit token is available to institutional clients on a public Ethereum layer-2 network4. Another global US bank's tokenised deposit service is now live in seven markets, most recently Japan and the UAE, offering clients 24/7 cross-border liquidity5. The use cases are practical: intraday and out-of-hours treasury transfers, collateral mobility and conditional, event-driven payments.
Our view: the first killer app of programmable money is not retail payments. It is a corporate treasurer moving liquidity across entities at 2 a.m. on a Sunday, with the controls attached to the money.
The distinction between forms of money matters more than the technology. A tokenised deposit is a claim on a regulated bank, sits on its balance sheet, can fund lending and carries the same protections as the underlying deposit. A payment stablecoin is a claim on a narrow issuer backed by segregated reserves; it cannot fund lending, and its value to holders depends on the quality and accessibility of those reserves. A CBDC is a direct claim on the central bank. Each has different implications for bank funding, which is why the choice of which money moves on-chain is ultimately a question about where deposits sit.
Stablecoins: regulated, concentrated and yield-free
Stablecoins are the fastest-growing competitor. The total market was about $292 billion at the end of September 2026, with the two largest coins, USDT at about $184 billion and USDC at about $74 billion, making up nearly nine-tenths6. The US GENIUS Act, signed on 18 July 2025, requires one-to-one reserves in cash, short-term Treasuries and similar assets, prohibits issuers from paying interest or yield, and lets issuers with up to $10 billion outstanding opt for state regulation. It takes effect on the earlier of 18 months after enactment (18 January 2027) or 120 days after regulators issue final rules7. The OCC published its proposed implementing rule in March 20268.
A concentrated market
Stablecoin market capitalisation by coin, end September 2026, $ billion ($bn)
Note: 'All other' is the total market capitalisation ($292.0bn) less USDT and USDC.
Source: CoinGecko, “Top stablecoins by market capitalisation” (2026)
For banks, the threat is not the size of the stablecoin market today, which is small relative to bank deposits, but the direction: dollar tokens that move 24/7 across public networks, embedded in wallets, exchanges and, increasingly, commerce platforms. The yield prohibition blunts direct competition for savings; it does not stop payments and working balances migrating.
CBDCs: the digital euro edges closer
In October 2025 the ECB's Governing Council moved the digital euro into its next phase, targeting a possible pilot in 2027 and first issuance during 2029 if legislation is adopted in 2026; it estimated development costs at about €1.3 billion to first issuance and running costs of about €320 million a year9. The European Parliament voted on 9 July 2026, by 416 to 169, to open negotiations with the Council, with the aim of completing the legislative process by the end of 202610.
Three forms of on-chain money
Status and scale indicators, September 2026
| Form of money | Issuer | Status | Scale indicator |
|---|---|---|---|
| Tokenised deposits | Commercial banks | Live; deposit tokens on private and public networks | >$5bn a day on one large bank's platform |
| Payment stablecoins | Licensed non-bank or bank subsidiaries | US GENIUS Act effective by 18 Jan 2027 | ~$292bn market capitalisation |
| Wholesale tokenised reserves | Central banks | Project Agorá real-value testing | 30 transactions, ~CHF 800,000, six currencies |
| Retail CBDC (digital euro) | Eurosystem | Legislation in trilogue; possible issuance 2029 | ~€1.3bn development cost to issuance |
Note: Compiled from the sources cited in the text: bank disclosures, CoinGecko, Latham & Watkins, BIS, Ledger Insights and the ECB.
What banks should decide now
- Pick use cases, not ledgers. Start where tokenisation removes a real cost: intragroup liquidity, cross-border treasury, collateral and conditional payments.
- Design for interoperability. Deposit tokens must settle against central bank money and coexist with regulated stablecoins; single-chain bets will age badly.
- Decide your stablecoin role. Issuer, reserve custodian, distributor or none: each has a different capital, compliance and technology profile under GENIUS-style regimes.
- Prepare deposit analytics. Model how much operational balance could migrate to on-chain forms and how quickly, as part of ALM and liquidity stress testing.
The next two years will set the architecture. US stablecoin rules will be finalised, the digital euro legislation should complete or stall, and wholesale experiments such as Agorá will show whether tokenised correspondent banking can integrate with central bank and core banking systems at scale. Banks do not need to bet on a single outcome. They need the data foundations, treasury processes and partnerships to operate in a world where all three forms of digital money circulate, and the discipline to move first where the business case is already proven.