The SEC Will Let a Blockchain Be the Share Register, as Long as One Firm Controls It

In eleven days, two regulators said in writing that a shared ledger can be the legally authoritative record of who owns what.

The SEC Will Let a Blockchain Be the Share Register, as Long as One Firm Controls It

Summary: The SEC's proposed rewrite of the transfer agent rules, in the Federal Register would let a blockchain serve as the master securityholder file, the official list of who owns a company's stock, for the first time since the rules were written in the late 1970s.

On September 10, Canada's bank supervisor said in one page that "tokenized deposits are, for example, not legally distinct from traditional deposits."

The same day, India's securities regulator announced that three companies had issued about $107 million of ordinary corporate bonds as tokens, settled against wholesale digital rupees so bond and cash moved as one transaction.

On Saturday, September 5, DBS and Citi moved U.S. dollars between Singapore and New York over Swift's shared ledger in minutes. U.S. Bank paid itself across the Atlantic with a dollar token on Stellar and used the trip to test minting, redemption, freezing, and clawback.


Thesis: For fifty years the official record of who owns a share, a bond or a deposit has been a private database inside one institution, and every other copy of that record has been legally a copy. This month the regulators started saying the shared ledger can be the original.

The claim of this issue, in one sentence: the shared ledger is being accepted as the official record of ownership, and the acceptance carries a condition, which is that a licensed institution keeps the power to rewrite it.

India's depositories hold the keys and remain the statutory record. Swift's ledger matches and nets, and the cash still settles in the old systems. U.S. Bank tested the undo button on purpose. And the SEC's version, the most permissive on paper, requires that one registered firm have exclusive control of the ledger it uses. On a public chain, exclusive control of the record means control of the token contract, which means the admin key I wrote about two weeks ago has just been promoted from a compliance choice to a recordkeeping requirement.

In the terms this newsletter uses to score every week, the record itself moved. Anyone can now verify it, software can act on it, and other systems can build on it, which is three of the five properties of open money arriving at once for deposits, bonds and share registers.

The two that didn't arrive are the ability to use the rail without asking and to hold the asset without an intermediary, and what's new is that the rule text treats their absence as a condition rather than a gap to close later. So the verdict this week is a hybrid with the gate written into statute: the record is open, the authority over it is closed, and for the first time the closure is a legal requirement rather than a design choice.


The book has been a private database since 1973

A share of stock used to be a piece of paper. By the late 1960s Wall Street was drowning in it: exchanges closed on Wednesdays to catch up on paperwork, and firms failed under the weight of unmatched trades.

The fix was the Depository Trust Company, opened in 1973, which took the certificates in, locked them in a vault, and replaced them with entries in a ledger. Ownership became a row in DTC's database. The paper became optional and then, for most securities, extinct.

That move sped up settlement, cut the fraud that came with lost certificates, and put the legally authoritative record of American equity inside a small club of intermediaries. A transfer agent keeps the issuer's official list of registered holders. A depository keeps its own list of participants. Your broker keeps a list of you.

What changed this month is the answer to a question that hasn't needed asking since 1973: does the official book have to be private?

Two regulators said the ledger can be the original

Canada answered first and shortest. The Office of the Superintendent of Financial Institutions, which supervises banks holding C$3.2 trillion in Canadian-dollar deposits and another C$3.7 trillion in foreign currency, published a statement on September 10 that runs to a few paragraphs. "The underlying technology of a financial product or service does not determine its legal nature."

So a tokenized deposit is a deposit. The bank that issues one owes the same money under the same law.

The SEC's answer runs to a couple of hundred pages and is more interesting. The transfer agent rules were "first adopted in the late 1970s and early 1980s," in the Commission's own words, and "have not been substantively updated since." The proposal, announced August 28 and printed in the Federal Register on September 4, rewrites the definition of the master securityholder file.

Proposed Rule 17Ad-9(b): "The master securityholder file shall be maintained in electronic form and may consist of multiple linked files or systems. The specific technology, systems, or files that compose the master securityholder file are within the transfer agent's discretion, provided the transfer agent maintains at all times exclusive control over the master securityholder file." The release says plainly that this "would permit a transfer agent to utilize a blockchain or other distributed ledger technology as its master securityholder file, or a component thereof, but it would not mandate it."

The practical effect is that a firm running tokenized shares can stop keeping a second, offchain register and reconciling the two after every transfer. The onchain file can be the file.

Commissioner Hester Peirce's September 1 statement shows where this is headed: she asks whether transfer agents should still collect names and physical addresses "or should the rule allow other identifiers, such as email and digital wallet addresses, to be collected instead," and the release contemplates a future "in which many shares will be tokenized."

India put real bonds on a ledger and called the ledger a copy

India reached the same place from the opposite legal direction, which makes it interesting to study to try and understand where this is all headed.

Three companies issued about $107 million of ordinary corporate bonds as tokens this week, on a ledger owned by India's two depositories, and buyers paid with the Reserve Bank's wholesale digital rupee so that "the bond and the money move instantaneously," in the Securities and Exchange Board of India's (SEBI) words.

The mechanism's legal wrapper is the part to read. SEBI's FAQ calls the network "a private, permissioned DLT network owned by the Depositories," says the depositories "will hold and manage the private keys on behalf of investors," and then draws the line: "The depository remains the authoritative record of beneficial ownership. The DLT ledger is the form in which the record is maintained for purposes of the pilot; it does not displace the depository's statutory role under the Depositories Act, 1996."

So the SEC would let the blockchain be the official record, provided one firm controls it. India put the bonds on a blockchain and declared the blockchain a format, with the statutory intermediary still the official record.

Swift moved weekend money and settled it somewhere else

The interbank version played out in Singapore over eight days, and it's the cleanest example of a shared ledger that changes the timing of money without changing who owes it.

On Saturday, September 5, DBS and Citi completed a live U.S. dollar payment between Singapore and the United States using tokenized deposits on Swift's Digital Ledger. DBS says it "took minutes to complete."

Citi's Mridula Iyer put the point the way a bank would: "Processing a live transaction over a weekend demonstrates that always-on cross-border payments are already a reality." A correspondent payment on a Friday night in Singapore would normally land in New York the following business day, or the one after.

On September 11, DBS, OCBC and UOB completed Singapore's first live domestic Singapore-dollar interbank transactions on the same ledger. The reporting describes the mechanics precisely: obligations were recorded on each bank's tokenized deposit infrastructure, then "matched and netted for final settlement through existing systems." That last clause is the whole design. The deposits stay on the banks' own books. Swift runs a permissioned matching layer where members can see and commit to obligations at any hour. Final settlement, the moment central-bank money moves, still happens in the domestic payment system when it opens.

What exclusive control means on a public chain

The SEC's phrase converts a design preference into a legal requirement, and that's the most consequential thing that happened this month.

Two weeks ago this newsletter looked at Coinbase's tokenized stocks and found the pattern: open rail, controlled asset, freeze and seize powers written into the token contract.

The SEC proposal requires that anyone using a blockchain as the official register maintain "exclusive control" of the master securityholder file "at all times," keep audit trails, and produce legible copies without third-party intervention. On a permissioned ledger that's a matter of who runs the nodes. On a public chain, the only way one firm can have exclusive control of a record that thousands of nodes replicate is to control the token contract: the mint function, the transfer policy, the ability to correct an entry a court says is wrong. In plain terms, the admin key. A transfer agent that puts the share register on Ethereum without a way to rewrite it has failed the rule.

The verification layer is opening: an Indian bond, a Singapore-dollar obligation, a U.S. Bank transfer, and soon an American share register can all sit on a record that outside software can read and compose with.

The authority layer is closing around the same institutions that held it in 1973, with one difference. In 1973 the record was private and so was the power to change it. Now the record can be public while the power to change it stays with one licensed firm, and that power is where the money is.

The honest case against this read

The strongest objection is scale, and it's fair. Almost nothing that moved this month is a market. Pilots can also ossify as press releases and never materialize as actual change.

The second objection is that I'm reading "exclusive control" too hard. The proposal is a proposal, comments run to November 3, and the phrase is probably aimed at something mundane, like making sure a transfer agent can't outsource its recordkeeping to a vendor and lose the ability to produce the file.

My reply is that intent doesn't change effect. Whatever the clause was written to catch, a transfer agent that satisfies it on a public chain will do so by holding administrative powers over the contract, and the rule will have made that structure the compliant one.

The third is jurisdictional. If multi-country issuance spreads without matching redemption terms, an open token layer becomes a cross-border liquidity problem. I'd treat that as the next policy fight, and as a reason the gates in this week's designs will get more elaborate from here.

What to watch

Whether "exclusive control" survives the comment period intact, and whether anyone at the SEC says what it means on a public chain. Watch for letters from Computershare, from the blockchain-native transfer agents the release mentions, and from anyone arguing the clause can be satisfied by something other than an admin key.

Whether India's pilot reaches secondary trading and retail, which SEBI says later phases will add. The first secondary trade of a tokenized bond between two investors, settled in digital rupees, is the moment the ledger stops being a primary-issuance convenience and starts being a market.

Whether Swift's ledger ever settles anything with finality on its own. Every live transaction so far has netted on the ledger and settled off it.

Strategic implications

If you're building. The buyers this month were depositories, transfer agents, and correspondent banks and all of them want the official book to be programmable, reconciled once, and plugged into the custody, AML and redemption law they already run.

Build atomic delivery-versus-payment against commercial-bank or central-bank money. Build the freeze, restriction and correction tools as explicit, documented policy modules, because the SEC has just told you what the compliant version looks like and it's the version with the admin key.

If you're allocating capital. Separate rail exposure from token exposure. Another dollar token adds nothing scarce. The scarce thing is software and licenses that sit at the point where the official record moves.

If you write policy. Canada showed that one page of technology-neutral language unblocks product work that years of novel-instrument debate froze. The SEC showed the other half: a ledger can be the file if someone is accountable for it.

What neither has done is say which kind of gate is allowed for which use. Identity checked once at issuance with open movement afterward produces one kind of system. A token the issuer can freeze in any wallet produces another. Both are about to exist under the same statutes, and the statutes should say which is which rather than leaving it to whoever drafts the contract.


x402 Volume Numbers Are Wrong: The Counter Everyone Quotes Hasn’t Moved Since March (2026)
Forty of the world’s largest payment and technology companies joined a foundation this summer for a market whose only public number comes from an unstamped counter that has not moved since March.

Last week's issue.


Sources

[1] U.S. Securities and Exchange Commission. "Transfer Agent Rules," proposed rule, Release No. 34-106246. Federal Register, September 4, 2026. Comments due November 3, 2026. Proposed Rule 17Ad-9(b) text and the statement that the amended definition "would permit a transfer agent to utilize a blockchain or other distributed ledger technology as its master securityholder file." https://www.federalregister.gov/documents/2026/09/04/2026-18190/transfer-agent-rules

[2] U.S. Securities and Exchange Commission. Fact sheet, "Proposed Transfer Agent Rule Modernization," announced August 28, 2026. https://www.sec.gov/files/34-106246-fact-sheet.pdf

[3] U.S. Securities and Exchange Commission. Commissioner Hester M. Peirce, "Time Transfer: Statement on Proposed Transfer Agent Rules," September 1, 2026. https://www.sec.gov/newsroom/speeches-statements/peirce-transfer-agent-rules-090126-time-transfer-statement-proposed-transfer-agent-rules

[4] Jones Day. "Recordkeeping in the Blockchain Era: SEC Proposes Overhaul to the Transfer Agent Rules," September 2026. Used for the exclusive-control conditions and the offchain identity, onchain position split. https://www.jonesday.com/en/insights/2026/09/recordkeeping-in-the-blockchain-era-sec-proposes-overhaul-to-the-transfer-agent-rules

[5] Office of the Superintendent of Financial Institutions. "Statement on tokenized and other digitally represented deposits," September 10, 2026. https://www.osfi-bsif.gc.ca/en/news/statement-tokenized-other-digitally-represented-deposits

[6] Bank of Canada. Chartered banks, classification of deposit liabilities (formerly K12), 2026 Q2: total Canadian-dollar deposits C$3,199,124 million, total foreign-currency deposits C$3,664,949 million. https://www.bankofcanada.ca/rates/banking-and-financial-statistics/chartered-banks-classification-of-deposit-liabilities-formerly-k12/

[7] Securities and Exchange Board of India. "Successful launch of Demat 2.0 pilot project for tokenised corporate bonds," press release and FAQs, September 10, 2026. Issuers, amounts, wholesale CBDC settlement via UMI, depository-owned permissioned ledger, key custody, and the Depositories Act language. https://www.sebi.gov.in/media-and-notifications/press-releases/sep-2026/successful-launch-of-demat-2-0-pilot-project-for-tokenised-corporate-bonds_104418.html

[8] The Block. "India's SEBI Demat 2.0 pilot debuts with over $100 million in tokenized bonds," September 11, 2026. Dollar conversion of ₹1,025 crore. https://www.theblock.co/news/regulation/2026-09-11-indias-sebi-demat-2-0-pilot-debuts-with-over-100-million-in-tokenized-bonds-414252

[9] IndiaBonds, citing CCIL and SEBI. Outstanding corporate bonds of ₹58,01,825 crore (US$644.9 billion); the page carries no as-of date. https://www.indiabonds.com/bonduni/blogs/size-of-the-indian-bond-market/

[10] DBS. "DBS and Citi partner to enable instant 24/7 cross-border USD payments with tokenised deposits," September 7, 2026, describing the September 5 transaction. https://www.dbs.com/newsroom/DBS_and_Citi_partner_to_enable_instant_247_cross_border_USD_payments_with_tokenised_deposits

[11] Fintech News Singapore. "DBS, OCBC and UOB complete Singapore's first live SGD tokenised deposit transactions on Swift's ledger," September 11, 2026. Source of the matched, netted and settled-through-existing-systems description. https://fintechnews.sg/137069/blockchain/dbs-ocbc-uob-singapore-tokenised-deposits/

[12] U.S. Bank. "U.S. Bank launches USBDC stablecoin," September 9, 2026. Live pilot between North American and European entities on Stellar; minting, redemption, freezing and clawback evaluated. https://ir.usbank.com/news-events/news/news-details/2026/U-S--Bank-Launches-USBDC-Stablecoin/default.aspx

[15] Bruegel. Lucrezia Reichlin, Bo Sangers and Jeromin Zettelmeyer, "A new strategy to contain stablecoin risks in the European Union," policy brief, May 20, 2026. Section 3.2.1 on multi-issuance and redemption-fee asymmetry. https://www.bruegel.org/policy-brief/new-strategy-contain-stablecoin-risks-european-union