Tokenized treasuries are widely described as the safe corner of crypto, and the description is broadly fair. The underlying asset is short-term United States government debt — the reference point against which the safety of everything else is measured. But the question "are they safe" cannot be answered by pointing at the bills, because you do not hold bills. You hold a token, issued by a company, representing an interest in a fund, holding securities at a custodian, priced by an administrator, recorded on a blockchain. The bills are the safest thing in that sentence. Everything else is where the risk lives, and this report takes it layer by layer. For how these products work, start with what are tokenized treasuries.
At a Glance
- The asset — About as safe as it gets
- The risk — Lives in the wrapper, not the bill
- Layers involved — Six or more, each a dependency
- Chain-specific — Contract, oracle, whitelist
- Tested? — Not yet, by a real failure
- Verdict — Safe-ish, for specific reasons
01 — Safe from what?
The word does too much work. In finance, safety is never a single property — an instrument is safe against some risks and exposed to others, and calling something "safe" without saying which is how people end up surprised. Six distinct questions hide inside the one being asked here.
Will the borrower repay? Will the fund holding the debt be run properly? Will the custodian holding the securities keep them segregated? Will the issuer of the token honour the claim? Will the technology recording your ownership function correctly? And will you be able to get your money out when you want it? Those have different answers, and the honest assessment is that the first is close to a certainty, the middle three are ordinary financial risks handled by regulated firms, and the last two are where something genuinely new has been introduced. Anyone evaluating these products should keep the six apart rather than reasoning from the reassurance of the first.
02 — The part that is genuinely safe
Take the underlying asset seriously, because it deserves the reputation. Short-dated United States government debt carries negligible credit risk in its own currency, and its short maturity means negligible interest-rate risk too — a bill maturing within months barely moves in price when rates change, unlike a long bond. It is the most liquid debt instrument in the world, tradable in enormous size at any moment the market is open.
That combination is genuinely exceptional and it is why this category grew while others stalled. Compare it with what sits inside the other real-world asset categories — a building with a tenant, a business that might not repay, as in tokenized private credit — and the difference is not one of degree. If the wrapper works as designed, you are holding the safest financial asset there is. The entire question is whether the wrapper works as designed.
03 — The stack above the bill
Between you and the government debt sit several distinct parties, each of which must function. The fund must be managed competently and within its mandate. The custodian must hold the securities properly segregated from its own balance sheet. The administrator must value the fund accurately. The transfer agent must maintain the ownership register correctly. And the issuer must honour the terms of the token.
None of these is exotic and all of them are ordinary financial infrastructure, usually operated by regulated firms with long records — which is precisely why the reputable products in this category are reasonable things to hold. But each is a dependency rather than a guarantee, and the aggregate matters: a chain of six sound institutions is not as safe as one, because any of them failing is a problem for you. Two consequences follow. Structures where the assets sit in a bankruptcy-remote vehicle, legally separated from the issuer, are materially stronger than those where the token is a claim on an operating company — the distinction that our note on real-world assets puts at the centre of assessing any of these. And concentration matters at the sector level, since a small number of custodians and administrators serve a large share of the market, so a single operational failure would reach further than one issuer.
The Treasury bill is not the risk. It never was. Everything you are actually assessing sits between the bill and you — and it is a longer chain than owning the bill, or a fund holding it, would have been.
04 — The layer that is genuinely new
Here is what tokenization adds that a conventional fund does not have, and it deserves more attention than it gets. Three things.
The smart contract that issues and transfers the tokens is software, and software has bugs. A flaw in the token contract, or in the whitelisting logic that controls who may hold it, is a risk with no equivalent in a conventional fund register. The oracle problem is subtler: the fund's value is calculated off-chain by the administrator, and that figure has to reach the blockchain somehow. Whatever mechanism carries it — a feed, a signed update, an operator transaction — is a dependency, and a stale or incorrect value flowing on-chain would misprice every downstream use of the token. And the bridge between records matters: the authoritative register may be the transfer agent's books rather than the chain, so an on-chain transfer that the register does not reflect conveys nothing. None of these has produced a major loss in this category to date. All of them are real, and they are the specific price you pay for the settlement speed and composability that make the product interesting.
05 — Can the price break?
A tokenized treasury can trade away from the value of its underlying holdings, and understanding why is more useful than being alarmed by it. The fund itself is valued on a schedule — typically daily — and creations and redemptions with the issuer happen in batched windows on business days. The token, meanwhile, transfers continuously.
So between valuation points, and especially at weekends, the token's secondary price is set by whoever is trading it rather than by the fund's arithmetic. If a holder needs to sell urgently into a thin market, they will accept a discount, and nothing arbitrages it away until the primary window reopens — the same two-tier structure that governs how stablecoins stay pegged. This is not a defect in the backing and it is not a depeg in the stablecoin sense; it is the predictable consequence of wrapping a business-hours instrument in a continuous one. The practical implication is that these products are sound for holding and imperfect for urgent liquidity, which is close to the opposite of how the marketing positions them.
06 — What has actually happened
The honest answer is: very little, and that cuts both ways. No tokenized treasury product has suffered a major failure — no significant loss of principal, no custodian collapse, no contract exploit of consequence, no issuer default. Redemptions have been processed, valuations have held, and the products have functioned as described. For a category this young that is a genuinely good record and it should be counted.
It should also be read carefully. The absence of a failure is not the same as proof of resilience, because the category's entire life has coincided with a benign environment for its risks: short-term rates were attractive, the underlying market was orderly, and no participating institution came under stress. The interesting test would be a period in which a custodian or issuer faced difficulty at the same time as holders wanted their money — and that has not occurred. Anyone reasoning from "nothing has gone wrong" to "nothing can" is making an inference the evidence does not support, in either direction. The correct statement is that the structures are sound in design and untested under stress.
07 — Against the alternatives
Safety is comparative, so the question is safe relative to what. Against a stablecoin, a tokenized treasury is generally safer in one specific respect — you hold a claim on a regulated fund's assets rather than an interest-free claim on a company that keeps the reserve income, as our note on how stablecoin issuers make money explains — and less convenient in another, since redemption runs on business days rather than instantly.
Against a conventional money-market fund held through a broker, the tokenized version is generally less safe, because it adds the technology layer and often sits under a lighter regulatory regime, without the investor-compensation protections a regulated brokerage relationship provides. And against directly held bills, it is less safe again, since direct ownership removes every intermediary. The pattern is consistent: each step toward convenience and composability adds a dependency. That is not an argument against the product — it is the trade the product exists to make, and for a treasurer wanting yield on an on-chain balance there is no conventional alternative that does the same job.
08 — The honest verdict
Tokenized treasuries are among the soundest instruments in crypto and they are not risk-free, and both halves of that sentence matter. The asset is excellent. The reputable structures are conservative, run by serious institutions, and have worked. The risks that remain are real but ordinary — counterparty, custody, operational, technological — and they are concentrated in the wrapper rather than the contents.
Three practical conclusions follow. Assess the issuer and structure, not the asset, because the asset is not the variable: ask whether the vehicle is bankruptcy-remote, who custodies, who computes value, and what your claim is in writing. Match the product to the horizon — these suit balances held for weeks or months, not money you may need on a Sunday. And keep the sizing sensible: safer than most of crypto is a low bar, and a token wrapping a Treasury bill is still an instrument with more moving parts than a Treasury bill. Used for what it is good at, by someone who has read the structure, it is a reasonable place to hold cash on-chain. Treated as a risk-free savings account because the words "US Treasury" appear in the description, it is being misunderstood.
"The name of the LORD is a strong tower: the righteous runneth into it, and is safe." — Proverbs 18:10
Methodology & Sources
This report assesses a category's risk profile as at the date of publication; it names no product, issuer, custodian, administrator or platform, and deliberately contains no yields, assets under management, fee levels or market-size figures — consult current issuer documentation and offering materials before acting. Structural descriptions reflect standard arrangements: tokenized treasury products are typically fund vehicles holding short-dated United States government securities or equivalents, involving separate investment-management, custody, fund-administration and transfer-agency functions alongside the token issuer, with the blockchain record substituting for or mirroring a conventional share register. The characterisation of short-dated US government debt as carrying negligible credit and interest-rate risk in its own currency is a standard description of the asset class and is not a guarantee. Technology risks described — smart-contract defects, whitelisting logic, the dependency created by transmitting off-chain valuations on-chain, and potential divergence between the authoritative off-chain register and the on-chain record — are recognised categories of operational risk in these structures rather than descriptions of any incident. The statement that no tokenized treasury product has suffered a major failure reflects the absence of publicly reported incidents at the time of writing and is expressly not presented as evidence of resilience; the category has not operated through a period of stress affecting its participating institutions. Comparative safety statements relative to stablecoins, conventional money-market funds and directly held securities are general characterisations that depend on the specific products and jurisdictions compared. Nothing here is a recommendation regarding any product, issuer or transaction, and nothing here is investment advice.