Tokenized Treasuries Are Getting a Trading Venue. The Assets Are Held So Tightly There's Barely Anyone to Trade With.
This month, DTCC began clearing tokenized U.S. Treasuries for regulated trading — the machinery to make them change hands like ordinary securities. But a market needs many owners willing to trade with each other, and on the chain that trading touches, BlackRock's fund is held by five wallets. Across the largest tokenized Treasury funds, the ten biggest wallets hold 92–100% of the supply. The venue is arriving; the holders to fill it are not there yet.
Tokenized Treasuries are having their moment. Real short-dated government paper, issued on-chain, redeemable in minutes — and, this month, part of something bigger: the industry's central plumbing has begun to let these tokens trade like ordinary securities, with more than fifty firms lined up behind the pilot. The pitch is liquidity: assets that today mostly mint and redeem could soon change hands in a real secondary market. So it's worth asking a plain question the headlines skip — who actually holds these funds right now?
On Ethereum, the chain the new trading rails touch first, the answer for BlackRock's BUIDL — the largest tokenized Treasury fund — is five wallets. Not five thousand, and not five million: five addresses hold essentially all of the roughly US$963 million of BUIDL that sits on Ethereum. (BUIDL is a US$2.5B fund across nine chains; this is its Ethereum float, on the chain in question.) That is not a knock on BUIDL. It is what a market looks like the moment before it's asked to become liquid — and BUIDL is far from alone.
Liquidity needs many hands
A market is liquid when many participants stand ready to buy and sell from one another, so any single trade barely moves the price. That is the promise being built on top of tokenized Treasuries: a venue where they trade freely. But a venue is not a market on its own. If the entire float of a fund sits with a few institutions that hold to maturity and redeem directly with the issuer, there is no one on the other side of a trade — nobody to make the market the new rails are meant to host. The number that measures that gap is holder concentration, and across the largest tokenized Treasuries it is strikingly high.
Fund by fund, on each fund's primary chain, the ten largest wallets hold almost everything. Hashnote's JTRSY — a single-chain, roughly US$870M fund, so this is its whole holder base, not a slice — sits in fourteen wallets, its top ten at 100%. VanEck's VBILL and Spiko's USTBL are effectively all held by their top ten. WisdomTree's WTGXX (~US$752M) has fifty-seven Ethereum holders and a top-ten share of 99%. Franklin Templeton's BENJI, measured on Stellar where most of it lives, runs about 94%. Ondo's OUSG, at roughly 71% once the fund's own protocol wallets are set aside, is the most diversified of the group — and it is still concentrated by any ordinary standard.
Fig 1 — Top-10 holder concentration · Jul 22, 2026
Treasuries vs. gold: top-10 holder share
Across the largest tokenized Treasury funds, the ten biggest wallets hold 92–100% of the on-chain float. Tokenized gold (PAXG, ~24%) shows what a broadly-held token looks like — the Treasuries look nothing like it.
None of this is a failing. Most of these funds are access-gated products — BUIDL carries a US$5 million minimum investment — so a holder base of a few large allocators is exactly what you'd expect, on-chain or off. The traditional money-market funds these modernize are held the same way; you just could never see it. The point isn't that concentration is dangerous. It's that a fund whose entire float is a handful of long-term holders is, by definition, not yet a traded market — and that's the reality the coming trading venues will be measured against.
The number nobody else publishes
Plenty of trackers will tell you how big these funds are, and how their supply splits across chains. What they don't tell you is how the float is held— the concentration of the actual holder set, measured against 100% of on-chain supply. That is the read here. For funds on Ethereum we reconstruct the complete holder set and rank it; for funds on Stellar, where no public indexer exposes contract-token holders at all, we read the full balance map directly from the ledger's own data warehouse. The result is a fund-by-fund concentration figure that isn't available anywhere else — and it's the figure the “these are about to trade” story is missing.
It's also why the honest version of the number matters. We set aside a fund's own protocol wallets — an issuer's treasury multisig isn't a third-party holder — so the concentration you see is genuine outside ownership, not an artifact of counting the issuer against itself. Even after that, the ten largest holders of most of these funds hold nearly all of the float. The gold contrast makes the shape unmistakable: PAXG, a broadly-held token anyone can buy, has a top-ten share around 24%. The tokenized Treasuries are an order of magnitude more concentrated.
The starting line, made legible
So the story isn't that tokenized Treasuries are too concentrated to trade. It's that the market is being handed a trading venue while its ownership still looks like a private placement — a few institutions, holding, redeeming directly with the issuer. Whether the new rails change that is the real question: liquidity arrives when the holder base broadens, when there are many hands to make a market rather than a few to hold one. Today there are a few. That's not a verdict on any fund; it's the normal shape of a young institutional market that found its first large allocators before it found its crowd.
What's new is that you can watch it happen. In the funds this market modernizes, holder concentration is disclosed quarterly to a few, if at all. Here it's on the ledger, fund by fund, wallet by wallet, updated daily — the starting line the tokenized-trading era will be measured against, in public, the day it starts to move.
Top-10 holder share is computed against 100% of on-chain supply on each fund's primary chain (Ethereum; Stellar for BENJI), from the complete holder set. BUIDL 5 holders / 100.00%; JTRSY 14 / 100.00%; WTGXX 57 / 99.13%; VBILL 16 / 99.97%; USTBL 19 / 99.99%; BENJI 93.94%; OUSG 70.69%. Values are protocol-excluded (see below).
Verified Jul 22, 2026 · onchainbenchmark.com/indexesA fund's own protocol-controlled addresses (e.g. an issuer treasury multisig) are excluded from the holder set — they are custody infrastructure, not third-party holders. This is why OUSG reads 70.7% here rather than a higher raw figure that would double-count Ondo's own wallets. Funds with no protocol addresses in-set (BUIDL, JTRSY, WTGXX) are unaffected.
Concentration method · onchainbenchmark.com/methodologyEvery figure is a fund's PRIMARY-chain holder set, not its whole multi-chain fund. BUIDL spans 9 chains (~US$963M of US$2.5B on Ethereum); OUSG 2; WTGXX 3. JTRSY is single-chain, so its figure IS the whole fund. Holder counts on funds at the 100-holder collection cap (BENJI, PAXG) are stated as top-100 shares, not exact holder totals.
Coverage notes · onchainbenchmark.com/research/coverageDTCC began its first tokenized stock and US Treasury production trades on July 15, 2026, in a live environment, ahead of a full commercial launch in October — covering Russell 1000 equities, index ETFs and US Treasuries, with 50+ firms including JPMorgan, BlackRock, Goldman Sachs and Circle participating.
The Block · DTCC begins first tokenized Treasury production trades