Banks have been running blockchain pilots since roughly 2016. For most of that decade the output was consistent: a well-produced press release, a consortium, a proof of concept, and then silence.
Something changed in 2025 and 2026. Tokenized treasuries, money market funds, private credit, bank deposits and regulated settlement rails moved into production workflows at banks, asset managers, custodians and corporate treasury teams. Coinbase and EY-Parthenon found 67% of institutions prioritising asset tokenisation over the next two years.
Understanding why the second attempt is working requires being precise about what banks are actually doing differently — and honest about how early it still is.
First, the measurement problem
Market size figures for tokenisation vary by an order of magnitude depending on what is counted, so treat any single number with caution.
Figure | What it counts | Source basis |
|---|---|---|
Around $17 billion | Tokenized financial assets narrowly defined, up roughly 3x year on year. US T-bills, bonds and money market funds account for over 55%; gold and commodities around 34%. | DefiLlama data cited by Citi Institute, April 2026 |
Above $36 billion | The broader real-world asset tokenisation market, reported as growing around 380% from roughly $5 billion in 2022. | Industry research, 2026 |
Above $340 billion | The tokenized asset market once cash-like instruments and regulated stablecoin rails are included. | Blockchain Council research, early 2026 |
The honest summary is that the narrow number is small and the broad number is large because it includes stablecoins. Citi’s own assessment is unusually candid: tokenisation sits at roughly one or two on a ten-point adoption curve, though the technology development curve is considerably further along. Its 2030 forecast is $4–5 trillion of tokenized financial assets.
The technology is nearly ready. The market is not. Anyone telling you otherwise is selling something.
What actually went into production
That said, the production use cases are real, and several carry volumes large enough to be unambiguous.
Deployment | What it does | Evidence of scale |
|---|---|---|
Broadridge DLR | Repo and short-term funding settled on distributed ledger with enterprise controls and supervision. | Reported average daily volumes of $339 billion in September 2025 and $385 billion in October. |
J.P. Morgan tokenized collateral network | Live transactions in which tokenized money market fund shares served as collateral with a global bank. | Production, not pilot — demonstrating that high-grade assets can secure obligations on-chain under institutional controls. |
Eurex / HQLAx | DLT-based collateral mobility for European participants. | Live for European market participants. |
MAS Project Guardian | Cross-border repo using tokenized assets executed by UBS, SBI and DBS. | Real-time collateral movement across regulated counterparties. |
Tokenized money market funds | Yield-bearing cash-equivalent instruments usable as collateral and for liquidity management. | BlackRock’s tokenized fund crossed $1 billion in March 2025 and later became eligible as off-exchange collateral at a major venue. Franklin Templeton expanded US on-chain funds and launched UCITS structures in Luxembourg. |
DTCC and Nasdaq | Tokenization pilots involving custodied stocks, ETFs and Treasuries; an SEC-approved framework for issuing, trading and settling certain tokenized stocks and ETFs through the DTC. | Regulatory clearance obtained; first trades anticipated during 2026. BlackRock, JPMorgan and Goldman Sachs jointly testing tokenized securities settlement through DTCC. |
Look at what dominates that list. Not tokenized art, not fractional real estate, not supply chain provenance. Collateral, repo and cash-equivalent funds — the least glamorous corner of capital markets.
Why collateral is the killer app
This is the central insight, and it explains why the second wave is working where the first failed.
Collateral management has a specific, expensive, universally acknowledged problem: the right asset is frequently in the wrong place at the wrong time. Moving it involves settlement cycles, cut-off times, multiple custodians and reconciliation. Firms hold buffers of high-quality liquid assets precisely because they cannot move what they have quickly enough.
That is a problem tokenisation genuinely solves. A tokenized money market fund share can be pledged, recalled and re-pledged without disturbing the underlying treasury position, in minutes rather than days, continuously rather than within business hours.
The efficiency gain is measurable in basis points on funding costs and in reduced buffer requirements — which means a treasurer can build a business case that survives contact with a CFO. Nothing about the earlier wave of blockchain experiments could do that.
Industry testing has shown how far the plumbing has come: tokenized money market fund units transferred across multiple heterogeneous ledgers including Ethereum, Canton, Polygon, Hedera, Stellar and Besu, as well as institutional cash networks, with subsequent tests connecting SWIFT messaging to tokenized collateral workflows and completing a full cycle from bilateral to triparty repo in under a minute.
The four things that changed
1. Regulation stopped being the blocker
MiCA created a unified rulebook across 27 states and became fully applicable during 2026. The GENIUS Act gave the US a federal framework for payment stablecoins. Hong Kong, Singapore and the UAE built their own regimes.
This matters more than it sounds. A bank cannot deploy production infrastructure against regulatory ambiguity, because the capital treatment, the audit position and the supervisory conversation are all unresolved. Once the rules exist, the internal approval path exists.
2. A settlement asset finally existed
The earlier experiments had a fatal gap: you could tokenize a bond, but you could not settle it in anything. Delivery versus payment requires payment, and there was no institutional-grade on-chain cash leg.
Tokenized deposits, regulated stablecoins and wholesale central bank money have filled that. Short-term sovereign and cash-equivalent instruments are expected to drive the largest on-chain volume growth precisely because institutions need trusted settlement assets before they can scale anything else.
3. Banks stopped trying to replace the market
This is the strategic shift that separates the two waves. Major market infrastructure providers are now testing tokenized settlement inside existing rails rather than trying to replace the entire securities system.
The 2018 pitch was that blockchain would disintermediate custodians, clearing houses and central securities depositories. Unsurprisingly, custodians, clearing houses and central securities depositories declined to fund their own disintermediation.
The 2026 approach routes tokenisation through the DTC, through The Clearing House, through Euroclear and Clearstream. The incumbents are the deployment channel rather than the target — which is slower and less romantic, and is why it is actually shipping.
4. Institutions asked for specific things, and got them
The most attractive tokenized asset classes for institutional investors are money market funds, corporate bonds and government bonds — and the bank products going live match that product set almost exactly.
That alignment is new. The earlier wave was supply-led: technologists building what was interesting to build. This wave is demand-led, which is a far better predictor of survival.
Where it is still hard
An honest account has to include the friction that remains.
Tokenized money market funds remain operationally and regulatorily more complex than direct tokenisation of treasuries, because they depend on fund structures and existing market infrastructure. Regulatory scrutiny following liquidity stresses in 2020 and 2023 could slow adoption further.
Broader adoption depends on liquidity and participation, not technology. A tokenized instrument with three counterparties is a demonstration, not a market.
Interoperability across ledgers is proven in sandboxes but not yet routine in production at scale.
Private-market assets — credit, real estate, infrastructure — are harder to price and harder to make liquid, which is why near-term adoption concentrates in high-quality liquid assets.
What this means depending on where you sit
If you are… | The relevant implication |
|---|---|
A bank or broker-dealer | Collateral and repo are where the demonstrable return is. Institutions expect to be managing live tokenized collateral, which makes execution speed a competitive variable rather than a research question. |
An asset manager | Tokenized cash-equivalent products are the entry point because the underlying assets are familiar, liquid and easy to price. Private credit follows once the settlement layer is routine. |
A corporate treasurer | The near-term benefit is collateral and liquidity management — posting or recalling without disrupting treasury operations — rather than any change to how you hold cash. |
A technology or infrastructure provider | The buyers now want compliant design, privacy controls, legacy integration and monitoring. Pilot-grade tooling no longer qualifies. |
An investor in the theme | Note the gap between a roughly $17 billion narrow market and a $4–5 trillion 2030 forecast. That is a long runway, and long runways contain plenty of failed intermediate steps. |
What to watch
First trades under the Nasdaq and DTC framework for tokenized stocks and ETFs. Equities settling through existing depository infrastructure would be the clearest signal yet.
Whether DLT repo volumes keep compounding. Daily volumes in the hundreds of billions are the strongest evidence in the sector and the easiest to track.
Collateral eligibility decisions. Every venue that accepts a tokenized fund as margin expands the addressable use case materially.
Cross-ledger interoperability moving from sandbox to production. Fragmentation across chains is the most plausible thing to stall this.
Whether private credit tokenisation gains traction, or stays a story told about the future while liquid assets do the volume.
The bottom line
Banks moved from experiments to production because they narrowed the ambition. The first wave tried to rebuild capital markets and produced pilots. The second wave picked one expensive, unglamorous problem — moving collateral — solved it inside existing market infrastructure, and produced volumes measured in hundreds of billions a day.
That is the lesson worth generalising. Tokenisation is not winning because the technology finally works; the technology largely worked in 2018. It is winning where there is a settlement asset, a regulator with a clear position, an incumbent willing to host it, and a treasurer who can put a number on the saving.
Everywhere those four conditions are absent, tokenisation is still a pilot. Which is why the market is genuinely at one or two out of ten — and why the parts that have shipped have shipped so decisively.
Important: This article is general information and market analysis. It is not investment, financial or legal advice. Market size figures vary substantially between sources because of differing definitions and are not directly comparable; figures cited reflect the dates stated. Forecasts referenced are published projections by their authors, not outcomes. Several initiatives described are at pilot or early production stage and may change or not proceed.
Sources: Citi Institute Tokenization 2030 report (June 2026), DefiLlama data, Coinbase and EY-Parthenon institutional survey, IMF notes on tokenized finance, Broadridge and J.P. Morgan disclosures, MAS Project Guardian materials, Global Digital Finance sandbox results, plus reporting from Blockchain Council, Zeeve, Crypto.com and B2Broker.
Bitnxt tracks RWA platforms, tokenization infrastructure and licensed service providers across the US, UK, EU and UAE. Explore the directory at bitnxt.io.
.jpg)
