The Plumbing Goes Live: DTCC Opens Its Tokenization Service, and the Prize Is Collateral

What is actually launching

The Depository Trust & Clearing Corporation is the post-trade infrastructure through which the overwhelming majority of US securities transactions pass. Its subsidiary, the Depository Trust Company, custodies securities worth more than US$114 trillion, roughly 3.8 times US GDP. When an institution of that size decides to put security entitlements on a distributed ledger, it is not a pilot announcement. It is a change to the operating system of a market.

The timeline has three confirmed milestones. In December 2025 the SEC’s Division of Trading and Markets issued DTC a no-action letter authorising a three-year controlled environment for tokenizing assets held in DTC custody. On 15 July 2026, DTCC completed initial production trades of tokenized US securities with major institutions, testing collateral pledge, securities lending and Treasury workflows. The full commercial launch of the DTC tokenization service is targeted for October 2026.

The eligible asset scope at launch is deliberately narrow and deliberately mundane: constituents of the Russell 1000 index, exchange-traded funds tracking major indices such as the S&P 500 and Nasdaq-100, and US Treasury bills, notes and bonds. These are the most liquid, most standardised and most heavily documented instruments in the market. That is the point.

The design decision that matters most is easy to miss. DTCC is not issuing new digital instruments. It is recording onchain the entitlements to securities that already sit in custody, so that each token is a digital twin of a book-entry position, carrying identical investor rights and protections. Nothing about the underlying credit exposure changes. What changes is where the record lives and how fast a claim against it can move.

Participation is broad, and the roster tells you where the centre of gravity now sits. The pilot involved roughly 40 firms, with named participants including BlackRock, Vanguard, JPMorgan and Goldman Sachs. The wider industry working group that co-designed the service ran to more than 50 organisations spanning custodians, asset managers, brokers, trading venues and crypto-native infrastructure providers — a list reported to include Citi, Morgan Stanley, Nasdaq, the NYSE, Circle, Ondo and Fireblocks. The question has shifted from whether tokenization will be adopted to how quickly it can be standardised inside existing plumbing.

The prize is not a new asset class — it is collateral mobility

The financial case here is not about issuance fees. DTCC data shows roughly US$300 trillion in global high-quality liquid assets, of which only about 10 to 11 percent is currently being used as collateral. The binding constraint is operational: collateral cannot move fast enough across time zones, settlement windows and separate books. Digital Asset has estimated that tokenized workflows could drive a 30 to 50 percent increase in balance-sheet efficiency by enabling near-real-time collateral mobility.

That is why the July 2026 production trades concentrated on collateral pledge, securities lending and Treasury workflows rather than on issuing anything new. In the framing DTCC managing director Nadine Chakar used at Sibos 2026 — the industry conference held in Miami Beach from 28 September to 1 October under the theme “Digital finance for AI-driven economies” — the priority is infrastructure that streamlines processes regardless of asset type, rather than the specific method used to move an asset between parties. Robust systems matter more than the wrapper.

The next piece on the roadmap is the DTCC Collateral AppChain, which incorporates Chainlink’s Cross-Chain Interoperability Protocol and is projected for the first quarter of 2027. Its pitch is near-real-time collateral management across multiple markets and multiple blockchain environments. When that goes live, interoperability stops being a nice-to-have and becomes a requirement: a collateral token that cannot leave its own chain cannot improve a balance sheet.

Adjacent moves point the same way. Securitize Markets received FINRA approval in the second quarter of 2026 to custody tokenized securities, which enables atomic settlement between tokenized securities and stablecoins and lets it participate in underwriting and selling groups. The strategic signal is that issuance alone is no longer a moat. The value is migrating to the trading, custody and settlement layers.

Value is not liquidity: the IMF says it plainly

The same month DTCC went live, the International Monetary Fund published a chapter of its Global Financial Stability Report — “Scaling Tokenization” — that reads as a deliberate counterweight to the infrastructure enthusiasm. Its findings: tokenized markets are fragmented, illiquid and more volatile than their traditional counterparts; tokenized equities trade at roughly 1.5 times the realised volatility of traditional equities; and deeper links to traditional finance could amplify liquidity crunches, fire sales and contagion as the market scales. It was the IMF’s third such warning in 2026.

The distinction the IMF supplies is that tokenized value and tokenized liquidity are different things, and the second does not follow from the first. A token can carry large outstanding value and near-zero liquidity at the same time. Distribution can be broad while trading stays thin. Redemption liquidity — an issuer’s ability to buy its own token back — is not secondary-market liquidity either.

The market-sizing makes the point uncomfortably well, because the numbers do not even agree on what is being measured. The IMF put outstanding tokenized real-world asset value at roughly US$65 billion as of July 2026. On 15 October 2026, rwa.xyz’s global dashboard showed a distributed asset value of US$39.18 billion, alongside a represented asset value of US$352.65 billion on the same page, and total holders up 16.4 percent over 30 days to roughly 5.26 million. Distributed, represented and outstanding are three different lenses on the same market. Any claim that tokenization is either enormous or trivial can be sustained by picking the right number.

That is why the underlying trade deserves more attention than the headline. DTCC is not trying to grow the tokenized market by a headline percentage. It is trying to make an existing, regulated market settle and collateralise faster. If it works, the effect shows up in balance-sheet efficiency and funding costs, not in a leaderboard.

Why Treasuries work and almost everything else stalls

Tokenized Treasuries have become the one category that plainly works, and it is worth being precise about why. They arrived with two of the hardest layers of the stack already solved: a predictable yield and a clear legal structure. Once those exist, the token can function as both a yield instrument and a collateral instrument, which is exactly what an onchain treasury operation needs.

The categories that stall — most real estate, most private credit — are precisely those where the legal, compliance and venue layers are still being assembled. Minting a token on an illiquid building does not create buyers for it.

There is a structural choice underneath this that is often treated as cosmetic and is not. An on-chain registry model reduces administrative overhead, but unless the legal framework recognises onchain records as binding, institutional participants discount those tokens on secondary markets. A bankruptcy-remote special purpose vehicle, by contrast, attracts institutional depth precisely because its enforceability is understood. The decision made before minting sets the liquidity ceiling.

Then there is compliance, which is where active markets die quietly. Every secondary-market trade has to be checked against the token’s transfer restrictions, the buyer’s eligibility, jurisdictional rules and holding-period requirements. Manual compliance review does not scale. If eligibility checks are not automated into the token and the venue, continuous trading is structurally impossible no matter how much demand exists — a design constraint, not a marketing problem.

That is also the discipline to apply to market-data headlines. Ondo Finance announced in mid-October that its tokenized US Treasury products had crossed US$3 billion combined. The claim to first place on the rwa.xyz Treasuries leaderboard rests on a three-product lineup, not on any single token outranking all rivals — and OUSG, one of those three, itself invests in BlackRock’s BUIDL, so part of the total ultimately flows through BlackRock’s fund. Layering of this kind is normal in the sector, and it is exactly why the basis of a total matters as much as the total.

What to watch from here

Three concrete things. First, whether the DTC tokenization service launches on schedule in October 2026 and which institutions actually transact through it — a participant list is not a volume. Second, the Collateral AppChain’s progress toward its first-quarter 2027 target, which is the test of whether cross-chain collateral can handle real institutional size. Third, whether round-the-clock trading of tokenized equity becomes a working feature rather than merely a permission, because that is the one capability that changes behaviour rather than describing it.

The link back to energy assets is direct, even though none of the instruments above are energy. If the collateral layer works, the practical question for a solar farm, a battery portfolio or a green bond stops being “can this be tokenized” and becomes “can this be pledged, financed and settled inside the same system as everything else.” The first question has been answered repeatedly. The second one is what October 2026 is actually testing.