In the summer of 2015, a reporter sat in a demo room at One Liberty Plaza in Manhattan and watched a stranger fill in a form: total amount, coupon, yield, maturity. One click, and the bond existed on a shared digital ledger. Issuing a security had been handled like an ordinary online checkout, and it had worked.
The reporter was Paul Vigna, and the demonstration came from Securitize, then an unfamiliar firm. Eleven years later, in August 2026, Neuberger Berman launched the first tokenized high-yield bond fund on Securitize's platform, its junk bonds and leveraged loans issued on four blockchains at once. Vigna, who spent five years predicting the financial system would be rewired along these lines, wrote in his American Banker column that it finally looked as though that was happening for real.
What deserves more attention than the launch is the decade that separated the demo from the fund. The technology did everything it was supposed to do in 2015, and it has in every pilot since. The missing piece was never the machinery. It was everything around it: who legally owns the token, what happens to it in a bankruptcy, whether a bank can hold it without wrecking its balance sheet, and whether two institutions can settle against each other on terms both accept. Those are not technical questions, and a ledger cannot answer any of them.
The 2015 demo already did the job
The value proposition was proven before most banks had heard of it. Issuance, record-keeping, reconciliation, transfer: the demo covered the whole pitch, and the benefits the industry describes today are the ones it showed then, easier record-keeping, lower costs, lower friction. Nothing about the core mechanism needed to be invented afterward.
What the demo could not show was what a token means off the ledger. If the platform operator fails, who holds the authoritative record of ownership? If the issuer goes bankrupt, does the claim stand? If two banks run different ledgers, how does a payment that starts on one end legally end on the other? The pilot decade was the industry answering those questions, one lawyer, one regulator, and one working group at a time. The pilots all worked, which is why their failure to scale was never a technology story.
The data is consistent with that reading. Trackers put the value of tokenized real-world assets at roughly $5.5 billion at the start of 2025 and above $30 billion by spring 2026, about a fivefold increase in fifteen months, with tokenized Treasuries the largest segment. Methodologies vary, but the direction does not, and the inflection coincides with the legal scaffolding, not the code: full application of Europe's MiCA framework, repeal of the rule that made crypto custody uneconomic, and the first federal stablecoin law.
The ledger could not promise ownership on its own
The clearest evidence that the constraint was legal sits in the history of one accounting bulletin. In 2022, the SEC's Staff Accounting Bulletin No. 121 required banks holding customer crypto in custody to record its full value as a liability on their own balance sheets, making the business uneconomic. Congress voted to overturn it, the president vetoed the resolution, and the rule stood until January 2025, when the SEC replaced it with SAB 122, which measures a probable loss rather than the whole balance.
The technology was identical the entire time. What changed was a rule about how a promise appears on a balance sheet, and the custody market responded immediately: U.S. Bank resumed institutional crypto custody in September 2025 after pausing in 2022, PNC followed in December, and Citi announced institutional bitcoin custody this month. Custody is not a technical capability; it is a legal commitment about safekeeping and return. The law changed, and a business that had been impossible for three years became ordinary.
The same pattern shows up in the international report. IOSCO's November 2025 final report on tokenization found adoption still dominated by pilots and small-scale implementations, and it identified legal uncertainty over ownership, transferability, and investor entitlements, plus the lack of credible settlement assets, as the central obstacles. Those are institutional gaps, not technical ones.
The laws went first, and the products followed
The sequencing of the past year reads like a test of the thesis that law, not code, paces tokenization. The GENIUS Act, the first comprehensive federal framework for payment stablecoins, was signed in July 2025, requiring one-to-one reserve backing and a licensed issuer. This year its rules are being written: the OCC proposed regulations in February, the Federal Reserve proposed requiring stablecoin issuers to run customer identification programs, and the Treasury published its scope rule this week.
The deposit networks arrived in the same window. JPMorgan's Kinexys says it processes more than $5 billion a day in tokenized payments. HSBC brought its tokenized deposit service to the United States in April. And in June, JPMorgan, Citi, Bank of America, and Wells Fargo, joined by more than a dozen other banks, announced a shared tokenized deposit network operated by The Clearing House, targeting the first half of 2027, built to connect on-chain settlement to the RTP and CHIPS rails that clear over $2 trillion a day.
None of these needed a new invention. They needed an agreement among competitors about who operates the network and under whose rules. The banks themselves frame the project partly as a response to stablecoins, whose supply sits around $263 billion and which the Treasury has estimated could put up to $6.6 trillion in deposits at risk. Whatever the merits of that estimate, the spur it describes is institutional: the question was whether the settlement asset of the future lives inside the regulated banking system or outside it. That is a question about trust, not throughput.
The interoperability problem is a trust problem in a technical disguise
The industry's standard explanation for the slow decade is interoperability: too many ledgers, no common standard, tokens that cannot move cleanly between platforms. Financial institutions have adopted dozens of different ledgers, and IOSCO warns that fragmentation splits liquidity and weakens network effects. It sounds technical, and partly it is.
But the people running the infrastructure say the hard part is not the protocol. Nadine Chakar, who heads digital assets at DTCC, the market's central settlement utility, said at a conference last year: "Technology isn't the barrier anymore: market trust and legal enforceability are." She also conceded that interoperability is a word the industry throws around loosely, and that the systems do not, in practice, work together. That admission matters: different smart contracts are easy to reconcile by comparison with different legal assumptions. Two tokens that both call themselves a deposit may carry different promises about recourse, redemption, and finality depending on the issuer's jurisdiction. Aligning those promises is governance work, and it is the work the industry has been doing for a decade.
That is why the shared networks matter more than any single product. The Clearing House structure is a rulebook and an operator, modeled by its participants on Zelle, an institution whose whole function is to make one bank trust another's payment instruction. The tokenized deposit network is the same product for the next generation of settlement: not a better ledger, but a shared promise about what a token means, enforced by institutions that all have a stake in keeping it true. A token is a claim with no guarantor. The network is the guarantor.
The skeptics are not wrong
None of this means the optimists have won. IOSCO found that most jurisdictions have no live tokenization use cases, or very few. The new products, including the Neuberger fund, are limited to accredited investors, and tokenized equities remain a sliver of a percent of the market they imitate. Vigna's own caution: "What Neuberger is doing may be real, but it still feels kind of proof-of-concept." The operational burden is real too: Tara Edmonds, an enterprise payments strategy executive at SouthState Bank, told American Banker: "Having to reconcile three platforms is probably a newer concept for banks." Controls were built for a single system of record, and replacing a core banking system can itself take a decade.
The record of bank consortiums is also sobering: three trade-finance networks, we.trade, Marco Polo, and Contour, were shut down between 2022 and 2023. The Clearing House network has not yet chosen its technology vendor, and aligning two dozen competitors is the kind of project that slips. Eight firms warned in February that the European Union's pilot regime, with its caps and license limits, is deterring participation and turning a first-mover advantage into a handicap. Revenue from tokenized products lags the volumes they process, and the final GENIUS Act rules are not done. The trust machinery is still being assembled.
The optimist's answer, which deserves its place, is that every large infrastructure shift in finance follows the same shape: technology proven early, law catching up late, products clustering at the moment the two converge. On that reading, the clustering of funds, networks, and custody services in the eighteen months since the statute-writing began is not coincidence; it is the model.
The decade was the product development
So the honest way to read the eleven years between the demo and the fund is not as a delay before tokenization, and not as a hype cycle finally maturing. It is as the product development itself. The demo proved the ledger could issue and record. Everything after has been building the thing a ledger cannot supply: legal certainty about ownership, finality, custody, and redemption, and shared rules that let two institutions trust the same token. That work happens in statutes, accounting bulletins, custody charters, and consortium rulebooks, one jurisdiction at a time, which is why it took a decade and why it looked like waiting. The fund launched in August 2026 not because blockchains improved but because the trust around them improved enough to sell to a bond fund. What is still missing is more of the same: the rules, not the code.
Primary sources
- American Banker's Bank Notes column by Paul Vigna, August 19, 2026, for the 2015 Securitize demonstration, the Neuberger Berman launch, the proof-of-concept caution, and the core-system conversion point.
- American Banker's piece on the Neuberger fund for the fund's structure, chain list, investor restrictions, and Securitize's $4.96 billion in tokenized assets, and CoinDesk's June 2026 coverage for the Clearing House consortium details, the 2027 target, and the stablecoin-competition motive.
- The Block's report on the Federal Reserve's proposed rulemaking for the GENIUS Act implementation timeline, and IOSCO's November 2025 Final Report on the Tokenization of Financial Assets for the adoption findings, legal-uncertainty analysis, and asset-class breakdown.
- Coverage of DTCC's Nadine Chakar at SmartCon for her market-trust remarks, RWA.xyz and Token Terminal research for the tokenized-asset figures and revenue-lag point, and CoinDesk's February 2026 report for the EU pilot-regime warnings.