A trade is only a promise until it settles. The United States has already cut the standard wait for most securities trades from two business days to one, and tokenization promises to cut it to a single, simultaneous exchange. Faster settlement genuinely reduces one kind of risk — and just as genuinely creates costs and leaves other risks exactly where they were.

A trade is a promise. Settlement is keeping it.

When an investor buys a share, two different things happen, and everyday language collapses them into one.

The first is the trade: buyer and seller agree a price, and the order is executed. The second is settlement: the money and the security actually change hands, and the record of ownership is updated. Until settlement happens, what the buyer holds is not yet the share itself, settled and final. It is a well-organized, heavily regulated claim that the share is coming.

That gap sounds like a technicality on a quiet day. It is not. Think of it the way you would think of selling a car privately: agreeing the price over the phone is not the same moment as handing over the keys and receiving cleared payment. Between agreement and completion, either side can fail to perform — the seller may not have what was sold, the buyer may not have the money, or something in between can break. Markets run that same gap at enormous scale, for a fixed, standardized length of time.

That length of time is the settlement cycle, written as T plus a number: the trade date, plus the number of business days until settlement is due.

What T+1 actually changed

On February 15, 2023, the U.S. Securities and Exchange Commission (SEC) adopted rule changes shortening the standard settlement cycle for most broker-dealer transactions in securities from two business days after the trade date (T+2) to one business day (T+1). The compliance date was May 28, 2024.

The Commission’s stated purpose was specific, and it is worth keeping it specific. In the SEC’s own words, the final rule was designed to benefit investors and reduce the credit, market and liquidity risks in securities transactions faced by market participants. The then-Chair described the same idea in plainer terms: reduce latency, lower risk, and promote efficiency and greater liquidity, and make the market plumbing more resilient, timely, orderly and efficient.

Note what the reform did not claim to do. It did not make any security a better investment. It did not change what a company must disclose, what a share entitles its holder to, or whether a price was fair. It shortened the interval during which a completed trade was still an uncompleted promise.

The same package of rules also tightened the work that happens inside that interval, requiring broker-dealers to put arrangements in place so that institutional allocations, confirmations and affirmations are completed as soon as technologically practicable and no later than the end of trade date. In other words, T+1 was not only a shorter deadline. It was a demand that the paperwork that used to drift into the second day be finished on the first.

Why even one day still matters

If one day can be removed, a fair question follows: why is there a day left at all?

Because conventional settlement is not one step. As the SEC’s Investor Advisory Committee described it in its March 12, 2026 recommendation on the tokenization of equity securities, settlement involves multiple counterparties — including brokers, transfer agents, the Depository Trust Company and the National Securities Clearing Corporation — that must update separate ledgers and facilitate the exchange. During the one-day delay, investors and brokers are exposed to the risk that settlement will fail, for reasons the Committee lists plainly: the seller lacks the shares to deliver, the buyer lacks the cash to pay, or there are other unexpected operational and technical failures.

That description does two useful things. First, it explains why the remaining day is not dead time: separate records are being reconciled and the exchange is being prepared so that money and securities can move against each other reliably. Second, it names the exact risk faster settlement is meant to attack — counterparty and operational risk in the gap — rather than investment risk in the security itself.

Atomic settlement: the tokenization promise, stated precisely

Tokenization’s settlement claim goes one step further than T+1. It does not propose a shorter gap. It proposes, in some designs, no gap at all.

The Investor Advisory Committee puts it this way: with a tokenized equity security, the delivery of the tokenized security and the payment can happen as a single transaction, with ownership records embedded directly into a single blockchain. This is often referred to as “atomic settlement,” and, the Committee says, it can eliminate settlement risk due to an inability to deliver shares or cash payments.

Read that claim carefully, because it is both real and narrow. If delivery and payment are truly simultaneous — both happen, or neither does — then one side cannot be left having paid for shares that never arrive, or having delivered shares that were never paid for. That is a genuine improvement in the mechanics of exchange, aimed at exactly the failure modes listed above.

It is also, at present, a design that sits outside the ordinary rulebook the Committee was examining. Allowing atomic settlement of tokenized equity securities, the Committee notes, requires exemptive relief or reforms to the SEC’s existing T+1 settlement rules. Instant settlement is not simply T+1 done faster by a more enthusiastic firm. It is a different settlement model, and it has to be authorized and structured as one.

What instant settlement takes away

Here is the part the sales pitch usually leaves out, and the Committee does not.

The current system’s delay is not pure waste. Part of what happens in that interval is netting: an investor’s — more typically a firm’s — many trades over the course of a day can be offset against each other, so that only the net amount of cash and securities needs to move. Atomic settlement, the Committee warns, prevents multilateral netting of that kind, and the elimination of multilateral netting can result in higher capital demands that increase costs for both broker-dealers and large institutional investors compared with the existing T+1 process.

Translated into a household analogy: netting is settling the whole week’s shared expenses in one transfer on Friday. Atomic settlement is paying every single item, in full, at the moment it happens. The second method is cleaner, item by item. It also requires everyone to have the full amount ready, every time, instead of only the balance at the end.

The same trade-off appears with mistakes. A system with an interval has room, inside that interval, to identify and correct a trading error before the exchange becomes final. A system designed to settle in the same instant has, by design, less of that room. Speed does not only remove waiting. It removes the slack in which waiting was doing quiet, useful work.

None of this is an argument against faster settlement. It is an argument for pricing it honestly: less counterparty risk in the gap, in exchange for more capital committed up front and less room to fix errors after the fact. Different markets, and different sizes of participant, will weigh that exchange differently — which is precisely why the choice should be debated as market structure, not repeated as a slogan.

What faster settlement does not fix

Settlement speed, at T+1 or at zero, leaves a longer list of things untouched:

It does not fix a bad investment. A share that settles in a minute can still fall in value the next minute. Settlement risk and investment risk are different risks, and removing the first says nothing about the second.

It does not fix disclosure. Whether investors were told the material facts still depends on the offering documents and the issuer’s reporting — not on how quickly the trade completed afterwards.

It does not create liquidity. A token that can settle instantly still needs a willing buyer at a fair price when its holder wants to sell. Instant delivery of something nobody wants to buy is still instant, and still a problem.

It does not settle the rights question. As this site explored in “A Share on a Blockchain Is Still a Share,” whether a token carries the dividend, the vote and the claim on residual assets depends on its legal structure and custody arrangements. Atomic settlement moves whatever rights the token actually carries, faster. If the rights were never attached, speed delivers the shortfall sooner, not the share.

It does not remove the need for controls. Identity checks, sanctions screening, record-keeping and error handling do not disappear when a ledger updates instantly. They have to happen earlier — before the trade — which is a harder operational problem, not an easier one.

The question to ask about any faster-settlement claim

When a platform, a venue or an issuer presents speed as the headline benefit, four questions separate the plumbing improvement from the pitch:

1. Faster than what, and under which rules? Is this T+1 under the current standard cycle, or an atomic model operating under specific exemptive relief — and if so, which relief, with which conditions?

2. What happens to netting? Are trades netted before settlement, or must each trade be funded in full, trade by trade? Who bears that capital cost, and is it shown in the fees?

3. What happens when something goes wrong? If an instruction is wrong, a wallet is compromised, or a counterparty fails mid-process, what is the correction and reversal procedure — and who decides?

4. What, besides speed, is actually being promised? If the answer drifts from settlement mechanics toward returns, liquidity or safety of the investment itself, the claim has left the ground settlement can support.

Faster settlement is a real direction of travel: the United States moved from T+2 to T+1 because a shorter gap is, on balance, a less risky gap. The next step, to simultaneous settlement, may prove just as worthwhile in the settings where its costs are worth paying. But settlement has never been the whole of a market. It is the part that keeps a promise made in a trade — and a market still needs the promise itself to have been honest, disclosed and fairly priced in the first place.

This article is education, not legal or investment advice. Settlement rules and any exemptive relief for tokenized securities turn on the specific instrument, venue and structure involved. Investors and issuers should take their own arrangements to qualified securities counsel in every jurisdiction involved — in Canada, that includes provincial regulators such as Quebec’s Autorité des marchés financiers (AMF) and the Ontario Securities Commission alongside U.S. federal law where U.S. securities or investors are involved.

Jean-Francois Amyot is an independent journalist and author of The Art of War Remastered.

By Jean-Francois Amyot

Tokenization changes how ownership of a share is recorded and transferred. It does not change what a share is. The practical question for any investor offered a tokenized stock is not how fast it trades — it is whether the rights that make a share a share actually travel with the token: the dividend, the vote, and the claim on what is left if the company is wound up.

The format changes. The law does not.

On January 28, 2026, the staff of three divisions of the U.S. Securities and Exchange Commission (SEC) — Corporation Finance, Investment Management, and Trading and Markets — published a Statement on Tokenized Securities. Its definition is worth reading slowly.

A tokenized security, in the staff’s words, is a financial instrument that is already a security under the federal securities laws, formatted as or represented by a crypto asset, where the record of ownership is maintained in whole or in part on one or more crypto networks.

Two consequences follow from that definition, and the statement makes both explicit.

First, format does not change legal status. The statement says the format in which a security is issued, or the method by which holders are recorded — onchain or offchain — does not affect the application of the federal securities laws. Every offer and sale must still be registered with the SEC unless an exemption is available, and stock is an equity security under the Securities Act and the Exchange Act regardless of its format.

Second, tokenization is, at its core, a record-keeping change. In the issuer-sponsored model the statement describes, an issuer or its agent integrates distributed ledger technology into the master securityholder file — the official record of who owns the security — so that a transfer of the crypto asset results in a transfer on that file. The only difference from a traditionally issued security, the staff says, is that the master file is maintained on a crypto network instead of in a conventional offchain database.

That is a useful demystification. A tokenized share is not a new species of investment. It is a share whose register has moved.

Note, too, what kind of document this is. The statement is a staff statement: it records the views of the three divisions, it is not a rule or regulation of the Commission itself, and it has no legal force or effect and creates no new obligations. It is guidance on how existing law applies — not a new regime, and not a permission slip.

Not every token that names a stock is the stock

If the story ended with issuer-sponsored shares, it would be simple. It does not, because the same statement describes a second category: securities tokenized by third parties unaffiliated with the issuer. Here the statement is careful, and readers should be too.

The rights, obligations and benefits attached to a third-party token, the staff says, may or may not be materially different from those of the underlying security. The token may or may not represent an ownership interest in the issuer of the underlying security, and it may or may not confer any rights as a holder of that underlying security. A holder can also face risks in relation to the third party itself — the statement gives bankruptcy as an example — that a direct holder of the underlying share would not necessarily face.

The statement separates two third-party models:

Custodial tokenized securities. The underlying share is held in custody by a third party, and the token evidences the holder’s ownership interest in it — typically as a security entitlement, an indirect interest recorded on the third party’s own records. What the holder owns is real, but it is mediated: dividends, votes and other entitlements reach the holder through the custodian’s arrangements, on the custodian’s terms.

Synthetic tokenized securities. The third party issues its own security that provides exposure to the referenced share — a linked security or a security-based swap. The statement is blunt about the linked security: it is not an obligation of the issuer of the referenced security and confers no rights or benefits from that issuer. Of security-based swaps it says they typically do not convey any equity, voting, information or other rights in the referenced security at all. The price may follow the share. The shareholding does not follow the price.

This is why “tokenized stock” is a label to interrogate, not a description to accept. Three products can display the same company name and the same chart on a screen, and be, in law, three different things: the share itself with a blockchain register, an entitlement to a share held by someone else, and a contract whose value references a share its holder does not own.

The SEC’s later Innovation Exemption for tokenized listed stocks, covered on this site when it was issued in September 2026, applies the same instinct in a trading-venue setting: a venue must verify that a tokenized stock made available for trading there provides the same rights and privileges as the traditional stock — the same dividends, the same voting rights, and the same share of residual assets on liquidation. A product that cannot pass that test is not, for that exemption’s purposes, the share in tokenized form.

A share of the same class — or a different class altogether?

The staff statement adds a refinement that matters to issuers and holders alike. A single class of securities can be issued in multiple formats, including tokenized format, and holders may be permitted to convert between formats. But an issuer can also issue a tokenized security as a different class, with different rights and features.

Whether a tokenized security counts as the same class as a traditionally issued one, for certain purposes under the federal securities laws, turns on substance: if it is of substantially similar character and its holders enjoy substantially similar rights and privileges, it may be treated as the same class. If the rights differ materially — a different dividend, a restricted or absent vote, a different place in line on liquidation — then calling it by the same name does not make it the same share.

For founders and boards, the lesson runs in both directions. If you tokenize your own shares, decide deliberately, and disclose clearly, whether holders are getting the same class in a new format or a new class with its own terms. Ambiguity here is not a marketing asset. It is a diligence finding waiting to happen.

Five questions to ask before buying — or issuing — a tokenized share

None of this requires a law degree. It requires reading the offering documents with five questions in hand:

1. Who issued this token — the company, or a third party? If it is a third party, everything else depends on that party’s documents, custody arrangements and solvency, not only on the company whose name is on the screen.
2. Where is the official record of ownership? Is the crypto network part of the issuer’s master securityholder file, part of a custodian’s records of entitlements, or neither? The answer tells you what a transfer of the token actually transfers.
3. How do dividends reach me? Directly as a holder of the share, or passed through an intermediary — in cash, reinvested, net of fees and withholding? “Same dividend” should be verifiable in the terms, not assumed from the product name.
4. Can I vote, and how? Is voting exercisable by the token holder, passed through by a custodian, or absent altogether? A token with no vote and no pass-through is not carrying the same rights as the share.
5. What do I hold if things go wrong? On a liquidation of the company, do you rank as a shareholder? On a failure of the intermediary, is the underlying share segregated for you, or are you a creditor of the platform? The staff statement’s bankruptcy warning belongs in every reader’s checklist.

If the documents answer those five questions clearly, tokenization is what it should be: better plumbing around a familiar instrument. If they do not, the speed of settlement and the convenience of an app are being offered as substitutes for ownership — and they are not the same thing.

This article is education, not legal or investment advice. Whether a particular token is a share, an entitlement, or a synthetic exposure turns on its specific terms and structure. Investors and issuers should take their own documents to qualified securities counsel in every jurisdiction involved — in Canada, that includes provincial regulators such as Quebec’s Autorité des marchés financiers (AMF) and the Ontario Securities Commission alongside U.S. federal law where U.S. securities or investors are involved.

A share on a blockchain is still a share — when the rights travel with it. Checking that they do is the whole job.

Jean-Francois Amyot is an independent journalist and author of The Art of War Remastered.