Most people think USDT is a crypto product. Wrong. It is a $187 billion money market fund that pays depositors zero, wrapped in an ERC-20 skin, and it just became one of the largest buyers of US short-term debt on the planet. Q2 2025 net operating profit: roughly $1.5 billion. Almost none of that came from crypto. It came from Treasury bills.
I spent four nights reconstructing Tether's June 30 reserve disclosure, line by line. Not because the token is interesting — it isn't — but because the balance sheet underneath it is now a macro instrument. When a private issuer sitting outside the US banking perimeter holds $114.96 billion in direct T-bills, you stop reading it as a stablecoin and start reading it as a systemically-relevant credit vehicle. The peg is the marketing. The portfolio is the product.
That gap — between what USDT claims to be and what it actually holds — is where the next round of risk lives. And almost nobody is pricing it.
Strip the branding. Tether's mechanism is a two-layer structure: an on-chain liability token (ERC-20, TRC-20, and a dozen other standards) sitting on top of an off-chain asset pool. Users deposit dollars. Tether buys short-dated US government paper and reverse repo. The spread is the business.
That structure is not a Ponzi. I want to be precise here, because the lazy take is "it collapses like UST." It doesn't — not on the same mechanism. UST was backed by an algorithm and a sister token's reflexivity. USDT is backed by actual Treasury bills earning actual risk-free yield. Tether's ~$1.5 billion quarterly profit is real income, not token emissions. That distinction matters, and I don't want anyone conflating the two failures.
What changed is the political framing. In October 2021 the CFTC fined Tether $41 million for misleading statements about reserve backing — specifically for claiming, between 2016 and 2019, that USDT was fully supported by bank fiat when it was not. In October 2022 Tether cleared its commercial paper book and swapped it for T-bills. Then in August 2025, KPMG US issued an unqualified opinion on Tether's financial statements.
Read that sequence again. A firm Washington fined for lying about reserves is now audited by KPMG and reported — via Bloomberg — as a potential counterparty in a Trump administration overseas stablecoin plan touching Treasury, State, and the DFC.
That is the story. Not the token. The reversal. Now the disclosure detail that the reversal obscures.
Here is the June 30 reserve composition. Total reserves: $187.75 billion. Total liabilities: $183.64 billion. Excess reserves — the buffer — $4.11 billion. That is 2.2%.
Break the asset side down and the "USDT is a T-bill fund" narrative falls apart. Direct T-bills: $114.96 billion, 61.2%. Overnight reverse repo: $18.63 billion, 9.9%. Good, liquid paper. But that is only about 71%. The rest is not Treasury.
Gold: $18.84 billion, 10.0%. Bitcoin: $5.8 billion, 3.1%. Secured loans: $13.45 billion, 7.2%. That last line is the black box. Secured loans to whom, against what collateral, at what haircut? The disclosure doesn't say. I've audited enough lending desks to know that "secured" is a word, not a guarantee. Collateral quality is where the tail risk sits, and it is exactly the line item with the least transparency.
Do the math on the buffer. $4.11 billion of excess equity against $187 billion of assets. If BTC and gold move 30% against the book, you lose roughly $7.4 billion — more than the entire cushion. The peg survives because redemption is orderly and reserves are mostly liquid. The peg does not survive a simultaneous drawdown in the volatile sleeve plus a confidence run. That is the tail. It is thin. Liquidity doesn't read attestations. It reads redemption queues.
Then there is the distinction almost nobody makes, and it's the one I keep hammering in my own client memos: an attestation is not an audit. A reserve report is a point-in-time snapshot, often prepared with agreed-upon procedures. A financial statement audit — which is what KPMG delivered — opines on whether the financials fairly represent the entity. It does not verify every reserve token-by-token. Tether upgraded from attestation to audit. That is real progress. It is not the same as per-coin reserve verification, and treating it as such is a category error. The unqualified opinion does not make USDT a government obligation. Tether remains a private issuer. Its liabilities are not America's liabilities.
Now the value capture, which is the part that actually annoys me. Tether earns roughly 100% of its income from holding US government paper. It pays USDT holders nothing. No interest, no governance, no claim on the reserve yield. The terms are set unilaterally, and USDT users have far less say over them — that is not a governance model, it's a take-it-or-leave-it contract with the strongest network effect in the sector.
Compare that to a money market fund, which distributes yield back to shareholders. Tether is an MMF that keeps the coupon. It is the most profitable asymmetry in crypto, and it scales directly with the Fed's policy rate.
The moat is not the reserves. It is distribution. USDT is the base quote asset on most exchanges, the default collateral in DeFi, and the practical dollar for anyone who can't open a US bank account. "Buying USDT is easier than opening an American bank account" — that is the entire pitch, and it explains why 60%+ market share has proven sticky. Network effects of that depth don't unwind on a disclosure. They unwind on a redemption panic, which is a different clock entirely.
The consensus read is: Tether won, it's too big to fail, the T-bills make it safe. I don't buy the safety half.
The bull case rests on the idea that USDT is now structurally tied to US debt demand — a win-win-win where Washington gets a buyer, Tether gets income, and users get a portable dollar. Fine. But that same tie is a single point of political failure. Tether's compliance standing depends entirely on the current administration's posture toward it. Policy reverses faster than reserves. And this is not hypothetical — the firm went from CFTC penalty to KPMG clean opinion in four years. That swung one way. It can swing back.
There is also a cleaner structural issue the narrative hides. If the overseas stablecoin plan becomes real — a joint venture, an official or semi-official issuer — it doesn't just validate the sector. It competes with Tether's gatekeeper role. Compliance-driven legitimacy tends to benefit the most compliant issuer, and that has not historically been Tether. Any regulatory tailwind for stablecoins as a category is a tailwind for USDC at least as much.
And the rate cycle. Tether's entire profit engine is the spread between T-bill yield and zero. Cut rates and you cut the income that funds the excess reserves that protect the peg. The buffer is already 2.2%. The earnings that thicken it are a function of the Fed, not of Tether's execution.
So where does that leave a trader? Track three numbers, not the narrative. First, the excess reserve line — if 2.2% falls under 1.5%, the tail is widening. Second, the gold-plus-BTC-plus-loan sleeve as a share of assets — every point above 20% is directional risk sitting inside something sold as cash. Third, the Fed path — every cut compresses Tether's income and, with it, the cushion.
The question worth asking is not whether USDT holds its peg tomorrow. It is whether a peg backstopped by a thin equity slice and a political mood survives the first genuine test. I'd rather size that position before someone else tests it for me.
