DTCC, the custodian holding almost all US stocks and bonds, goes live with converting the securities themselves into 'tokens' on the blockchain

Last time, I wrote about how the EU AI Act made it a legal requirement to attach machine-readable labels to AI-generated content. It was an asymmetric story where only the labeling rules moved forward as planned, while regulations on performance and safety were postponed. I think that was interesting in its own right as a move by the rules to try and catch up with reality.

This time, I want to go in a completely different direction. It is a story about the deepest part of finance, a place no one usually looks.

What happened

On July 15, 2026, the DTCC—the American organization equivalent to the 'Japan Securities Depository Center'—announced that it had converted the actual securities it holds into tokens on a blockchain and processed actual transactions using them in a production environment. This was not a test or a proof of concept; it was the real thing.

The types of transactions processed included collateral pledging, securities lending, simultaneous delivery of US Treasuries and repo transaction funds and securities, simultaneous delivery of stocks, token transfers, and the exchange of margin for clearinghouses. More than 30 organizations participated, with traditional giants like BlackRock, Goldman Sachs, JPMorgan, Vanguard, and State Street standing on the same ledger alongside digital asset-side companies like Circle, Fireblocks, and BitGo, as well as exchanges like the New York Stock Exchange, Nasdaq, and CME. Two networks were used: Besu, which the DTCC owns itself, and the public network Canton. The scope initially covers about 1,000 issues, starting with Russell 1000 stocks, US Treasuries, large-scale index-linked ETFs, and some bonds.

The full-scale service launch is announced for October 2026.

What is the DTC in the first place?

I want to explain this carefully. It may be unfamiliar to Japanese readers, but the weight of this story is determined here.

When you buy stocks in the US, those stock certificates are not held in your name. Most listed stocks and government bonds are, in terms of title, held collectively under a single institution called the DTC. What investors hold is an 'equity interest' recorded in the books of their brokerage firm, while the underlying securities are concentrated at a single point: the DTC. Assets on the scale of tens of trillions of dollars are effectively moving within a single ledger.

You can think of it as having the title deeds for all the real estate in the country kept together in a single vault. Instead of carrying out the title deeds every time a sale occurs, they simply rewrite the name field in the vault's management ledger. That is the mechanism of the DTC.

Therefore, this event is different in kind from a story about 'a company issuing a new token.' It is a story about the vault's management ledger itself being replaced by a different technology.

The meaning of the term 'digital twin'

The approach the DTCC took was not to replace existing securities with tokens, but to create a copy of the securities held by the DTC as tokens. This is called a digital twin. Legal ownership, the right to receive dividends, and the framework for investor protection all remain exactly the same as the original securities.

This is subtle, but I think it is decisive. Many of the 'tokenized stocks' that have emerged from the crypto asset side so far have been in the form of some company holding real stocks and issuing depositary receipts on a blockchain. As long as they were depositary receipts, the question of what would happen if that company went bankrupt always remained. This time, the very entity holding the securities is issuing a copy of its own ledger. The issuer of the depositary receipt and the custodian of the securities are the same.

What made this possible was a no-action letter issued by the SEC to the DTC—an expression of the authorities' intent that 'we will not subject this activity to law enforcement.' It is not that a new law was created. Within the existing framework, the authorities acknowledged that 'these remain existing securities.' It is a form of opening a door by shifting the interpretation of the system by one step, rather than rebuilding the system itself.

What will change—the speed at which collateral moves is directly a matter of capital

This is likely where it will have an impact on business readers.

Current securities settlement is processed in batches during business hours. From trade execution to settlement, it takes one business day. Whether pledging collateral or posting margin, it takes time to move securities from right to left, and the market does not stop during that time. Therefore, financial institutions must keep excess cash and securities idle in preparation for 'times when they cannot be moved.' This idle portion is generated as a cost every day.

If you know that a highway will be closed at night, a trucking company has no choice but to stock inventory at each base. If it becomes possible to run 24 hours a day, that inventory becomes unnecessary. However, you cannot reduce inventory until you have verified that the road can truly be used 24 hours a day.

If collateral can be moved by the minute, and including weekends, this idle portion can theoretically be reduced. The reduced portion becomes capital that can be used elsewhere. That is what I mean when I say that the story of settlement infrastructure, which seems boring, is actually a story about capital efficiency.

And this is a direct continuation of the mandatory central clearing for US Treasuries that I wrote about previously. Back then, the question was, 'Who stands in the middle of the trade?' By centralizing counterparty risk, you increase the safety of the market as a whole, but in exchange, the need to post collateral increases. If the need to post collateral increases, then how quickly that collateral can move becomes critical. One side increases the amount of collateral, and the other side increases the speed of the collateral. Looking back now, I realize these two seemingly separate pieces of news were just two sides of the same coin.

Whose work will decrease, and whose work will be created?

To be honest, I don't think a massive amount of work will disappear overnight. However, the direction is clear.

What will decrease is the work of reconciliation and cleanup. A significant number of people are still tied up in tasks like: 'my company's ledger doesn't match the counterparty's,' 'the settlement didn't make the deadline,' or 'the collateral is in the wrong place and needs to be swapped.' If a state where everyone is looking at the same ledger becomes widespread, this type of work will, in principle, shrink.

What will be created is the work of writing rules on the ledger. People who can describe collateral swaps and margin calculations as programs rather than contracts, and who can determine what that means legally. People who design the operations for moving assets across multiple networks. And people who monitor 24/7 systems 24/7. I believe the value of talent that sits at the intersection of law and technology within financial institutions will continue to rise.

Regarding barriers to entry, I would like to point out that a counter-intuitive movement is occurring. For startups that were trying to provide tokenized stocks on their own, this is not a tailwind. It is because the most centralized incumbent institution has started doing the same thing, bringing its own authority with it. While the technology used is decentralized, the structure is becoming even more centralized. I think this irony is worth recording.

Things we don't know yet

I don't want to exaggerate, so I will write down the parts that are not yet certain.

First, it has not been proven whether there is demand. What was proven this time is that it 'can be done,' not that it 'will be used.' Many of the participating companies likely raised their hands just to see how things go. No one knows yet how much balance will actually move to the token side after the full-scale launch in October.

Second, the final settlement of trades still resides within the existing DTC system. It is accurate to understand that the blockchain has not replaced settlement, but rather that a new layer of transfer has been placed on top of existing settlement.

Third, the risk of concentration. A structure where everything is gathered on a single ledger is efficient, but there is no alternative if it stops. Even if the technology is new, this nature does not change. In fact, because the speed of movement increases, the speed at which errors propagate also increases when something goes wrong.

The turning point usually has a boring face

You know a new technology has truly taken root not when it becomes a hot topic, but when it is embedded in the foundation and no one talks about it anymore. Considering that the word 'blockchain' has spent over a decade oscillating between expectation and disappointment, this event, where the word quietly entered the ledger of a vault, may be more significant than any hype we have seen so far.

Events that are called 'the turning point' when looking back years later usually have a plain face on the day they are announced. I see this as that kind of event. Of course, there remains the possibility that nothing will happen after October. Including that, I want to keep a close eye on this for a while.