Tether’s Q2 2026 Attestation: A Thinner Floor, A Heavier Omission
KPMG has been inside Tether for four months. The audit is still in progress. In that same quarter, Tether’s excess reserve cushion collapsed from $8.23 billion to $4.11 billion. The company reported net operating profit of $1.5 billion. The ledger doesn’t lie, but it does omit.
Tether is the third-largest crypto asset by market cap, with 183.64 billion in liabilities against 187.75 billion in total assets. It added 30 million net new users in the quarter, reaching more than 650 million wallets. Daily trading demand remained strong even after Revolut announced it would delist USDT for European customers. None of those facts are disputed. The dispute begins when you ask what changed inside the reserve report between Q1 and Q2.
The July 31, 2026 attestation from BDO is not an audit. It is a limited-assurance engagement confirming that Tether was “over-collateralized” at a single point in time. It does not validate asset quality, mark-to-market accounting, or whether the reserves are accessible under stress. Four months after Tether announced that KPMG had begun a full audit, the word remains “in progress.” Attestation and audit are different instruments. The first is a pulse check. The second is an autopsy. Tether is still refusing to schedule the autopsy.
The most important number in the report is not the $1.5 billion net operating profit. It is the fact that the excess buffer fell from $8.23 billion to $4.11 billion in the same quarter. In percentage terms, the cushion over liabilities dropped from roughly 4.5% to 2.2%. A stablecoin issuer with a 2.2% buffer can still redeem everyone in theory. But a bank run is not a theory. It is a liquidity event that demands cash speed, not asset quality. And the buffer is the only spare cash visible to the outside world.
Now the accounting. Tether’s Q2 release called the number “net operating profit.” The Q1 release called it “net profit.” The difference is not cosmetic. Operating profit excludes unrealized gains and losses from the reserve portfolio. In a quarter where Bitcoin and gold moved sharply, excluding mark-to-market losses makes the income statement look cleaner than the balance sheet actually is. I built arbitrage scripts during the 2017 ICO mania and later manually audited early DeFi lending contracts. I learned to distrust any financial statement that changes its profit definition just before a volatile quarter. This is one of those statements.
Let me walk through the arithmetic. Tether generated $1.5 billion of operating profit. If profits are retained, shareholders’ equity should increase by at least $1.5 billion. Instead, the excess buffer fell by $4.12 billion. That implies roughly $5.6 billion of negative unexplained movement elsewhere in the portfolio during the quarter. Unrealized losses on Bitcoin or long-dated bonds could explain part of it. Write-downs on private loans could explain the rest. The report does not say. I don’t know where the $5.6 billion went. Neither do you.
This is where disclosure quality matters. In Q1, Tether broke out reserve components in dollar figures. In Q2, gold was disclosed as “over 146 tonnes” rather than its dollar value. U.S. Treasuries were described as “the majority of reserves” without a specific dollar amount. Bitcoin’s reserve position was not given at all. These are not rounding choices. They are deliberate removals of fixed dollar commitments in a period when Tether knows an external auditor is scrutinizing historical valuations. If the KPMG audit later revises any asset’s dollar value, Tether does not want a stale $200 billion gold line in its own attestation for the court record. Silence is the only honest signal in the noise.
The contrast with USDC is becoming structural. Circle provides monthly reserve breakdowns and a legal commitment to SEC disclosure. Tether provides a quarterly attestation with fewer details than it gave one quarter ago. I have no institutional loyalty to either coin. I care about the liability structure. A stablecoin is not a protocol with code you can fork. It is a promise backed by assets you cannot see. Code-first risk verification stops at the chain boundary. The moment you hold USDT, your real counterparty is a Cayman Islands entity, an Italian auditor, and an ongoing KPMG investigation. Risk isn’t a variable you control. It is a variable you price.
The contrarian part is this: almost nobody in the retail or emerging-market base cares. Tether added 30 million users in the quarter. Revolut’s European delisting did not produce a measurable dent in demand. That is not irrational. In Zimbabwe, Argentina, or Nigeria, USDT is a banking system. Transparency is an abstract demand from Western institutions. What those users need is a dollar-pegged bearer asset that can survive capital controls and inflation. They are not going to switch to USDC because of a footnote. The lesson of 2022, when I shorted Celsius and Voyager into liquidation cascades, is that demand is sticky until the price breaks. Users do not flee on warnings. They flee on the first frozen withdrawal screen.
But institutions are different. They read the Q2 attestation and see a shrinking buffer, hidden dollar-value disclosures, and a KPMG audit that is still not finished. They will not short USDT into a quiet market. They will simply widen the cost of carry, reduce collateral acceptances at lending desks, and shift a portion of treasury allocations into USDC or short-dated U.S. Treasuries. That slow drift is the dangerous one. It is not a depeg event. It is a steady decline in the marginal liquidity premium that Tether has always earned. Volatility is just unpriced fear wearing a mask. The question is whether that mask slips before or after KPMG signs the audit.
My forward levels are simple. I do not short USDT on the spot market because the peg can hold through any quarterly report. I watch the secondary DEX prices for sustained prints below 0.998. I watch the excess buffer. If the buffer falls below $2 billion, the equity cushion becomes a rounding error in a bank-run scenario. If the KPMG audit misses a second quarter, the delay stops looking like diligence and starts looking like negotiation. If either happens, I will treat USDT not as a stablecoin but as a distressed credit with a currency wrapper. The floor isn’t a feature. It is an option you haven’t bought yet, and the premium just doubled.