Stablecoin Reserve Report Due Diligence: USDC, USDT, PYUSD, USDe

A stablecoin reserve report tells you what is supposed to be backing the coins you hold. It does not tell you whether the backing is structured the way you would want it to be, whether the custody arrangements are robust, or whether off-balance-sheet exposures exist that erode the headline reserve number. For operators using stablecoins in treasury or payment flows, the ability to read a reserve report critically is the difference between counterparty risk management and faith-based finance.

This briefing is the operator’s reference for reading stablecoin reserve reports — what the documents actually say, what they don’t say, and the five lenses that separate documented backing from marketing copy. The comparison covers the four issuers most relevant to operator decision-making in 2026: USDC (Circle), USDT (Tether), PYUSD (Paxos for PayPal), and USDe (Ethena Labs) as the most prominent non-fully-reserved instrument that operators encounter. For an operator-level comparison of these four issuers across risk, regulatory status, and use-case fit, see the stablecoin issuer comparison.

Attestation vs Audit — the Distinction That Matters

The first thing to know about a stablecoin reserve report is that it is almost always an attestation, not an audit. An attestation is a point-in-time confirmation by an independent accountant that a specific set of assets exists in specific accounts as of a specific date — typically the final business day of the month or quarter. An audit examines the issuer’s internal controls, the broader financial reporting, and the reasonableness of management’s representations over a full reporting period.

The practical implications:

• An attestation does not test whether the assets reported existed the day before or will exist the day after. It is a single-day snapshot.
• An attestation does not examine the issuer’s broader financial position, related-party transactions, or operational risks.
• An attestation does not provide an opinion on whether the issuer’s overall business is solvent — only that the named reserve assets existed in the named accounts on the named date.

A full audit covers these gaps. Circle publishes an annual audit by Deloitte; Paxos publishes annual audits by WithumSmith+Brown. Tether publishes a quarterly attestation by BDO Italia but does not publish a full annual audit as of early 2026. The distinction is the single most important thing to understand when comparing issuers.

Lens One — Asset Composition

The headline number on every reserve report is the total reserve value. The first lens is the composition of that number — what percentage is held in genuinely liquid, redemption-ready assets versus less liquid instruments that may not be available at par on short notice.

Under the GENIUS Act, MiCA, and the MAS stablecoin regime, compliant payment stablecoin reserves must be in high-quality liquid assets:

• US Treasury bills with maturities under 90 days. Settle T+1, redeem at par at maturity, deep secondary market liquidity.
• Cash deposits at federally insured (or jurisdictionally equivalent) banks. Available on demand up to deposit insurance limits.
• Overnight reverse repurchase agreements collateralised by Treasuries. Settle daily.
• Money market fund shares invested in HQLA, where permitted.

The composition issues to look for:

• Commercial paper and corporate bonds. Less liquid than HQLA, can trade at meaningful discounts in stressed markets, and are excluded from compliant reserve structures under recent regulatory frameworks. Tether reduced commercial paper holdings substantially between 2022 and 2024 but historically held over $20 billion in commercial paper at peak.
• Longer-dated Treasury holdings. Treasury bills under 90 days are the standard; anything longer carries interest rate risk that affects the mark-to-market value of reserves. Most compliant issuers stay short, but the maturity profile should be disclosed.
• Gold and other non-cash equivalents. Tether has historically held gold positions disclosed in reserve reports. These are not redeemable for fiat at par on demand — they are commodity exposure dressed as reserves.
• Secured loans. Tether has historically disclosed “secured loans” in its reserve composition. The counterparties are not always disclosed, and the loans are typically collateralised by Bitcoin or other crypto assets — meaning the reserve is exposed to crypto market price movements.

USDC’s monthly attestation reports 100% HQLA composition: short-dated Treasury bills and cash deposits, with no commercial paper, no longer-dated bonds, and no non-cash equivalents. PYUSD is similar. Tether’s most recent attestations show HQLA majority but with disclosed “other investments” that vary quarter to quarter. USDe does not publish a reserve report in the same format — its backing is staked ETH and perpetual futures hedge positions, which require a different analytical framework entirely.

Lens Two — Custodian Concentration

The second lens is where the reserves are actually held. A stablecoin issuer can have $50 billion in HQLA, but if all of it sits at a single custodian bank that fails, the stablecoin loses access to its backing for the duration of the resolution process — even when FDIC insurance ultimately covers the loss.

The canonical case is USDC at Silicon Valley Bank in March 2023. Circle held approximately $3.3 billion of USDC reserves in cash deposits at SVB. When SVB failed and the FDIC initially declined to backstop uninsured deposits, USDC briefly traded at $0.87 as the market priced in the possibility of partial loss on those deposits. The situation only resolved when the FDIC and Treasury intervened to backstop all SVB deposits. The lesson: concentration at a single custodian, even an insured one, is a source of stablecoin peg risk that the headline reserve number does not capture.

What to look for in the attestation:

• Custodian list. USDC’s monthly attestation names individual custodian banks. Tether’s attestation does not. The disclosure asymmetry is itself information.
• Largest single custodian exposure. A diversified reserve structure should have no single custodian holding more than 25–30% of total reserves. Concentration above 50% at a single bank is a material risk.
• Geographic diversification of custodians. A reserve held entirely at US banks has different risk characteristics than one diversified across US, EU, and Asian custodians — both for resolution-process risk and for sanctions/regulatory risk.

Circle restructured USDC reserves materially after the SVB incident, diversifying cash deposits across multiple federally insured institutions and moving a larger portion to Treasury bills (which are held in custody arrangements that are not affected by individual bank failures the same way). Paxos has historically used a similar diversified structure. Tether’s custodian list remains undisclosed at the level of individual banks.

Lens Three — Maturity Profile

The third lens applies to the Treasury bill portion of reserves: what is the weighted average maturity, and how much interest rate risk does the issuer carry?

For payment stablecoin reserves, the safe range is 30–60 days weighted average maturity. Shorter is better — it means reserves can be liquidated quickly without realising mark-to-market losses in a rising-rate environment. The risk of longer maturities:

• A stablecoin issuer holding 180-day Treasury bills faces meaningful mark-to-market losses if rates rise sharply.
• If redemption demand spikes simultaneously with a rate shock, the issuer may need to sell at a discount, eroding reserve coverage of outstanding stablecoin supply.

USDC’s attestation discloses weighted average maturity of the Treasury portion of reserves, typically around 30 days. Paxos publishes similar detail. Tether’s attestation does not break out maturity profile in equivalent detail.

The maturity profile lens matters more in rate-volatile environments. In a stable rate regime, the difference between 30 and 90 days is small. In a rate-shock environment, it can be material.

Lens Four — Off-Balance-Sheet Exposure

The fourth lens is what the reserve number does not capture. Several types of exposure can erode the practical backing of a stablecoin without appearing in the headline reserves figure:

• Secured loans. Tether has disclosed secured loans in reserve composition, typically collateralised by Bitcoin or other crypto. The reported value reflects the loan principal, not the collateral coverage — meaning a sharp crypto price drop could leave the loan undercollateralised without the reserve number changing.
• Related-party balances. Some stablecoin issuers have operational entities or affiliated exchanges that hold balances against the issuer. These can be netted in reporting in ways that obscure true reserve coverage.
• Bitcoin and other crypto holdings. Tether has accumulated meaningful Bitcoin holdings disclosed in its assets but not always treated as part of reserves backing USDT supply. The distinction between “reserves” and “assets” matters for stablecoin holders.
• Derivatives and hedging exposure. Synthetic dollar instruments like USDe have hedging positions that are part of the backing structure but expose the issuer to derivatives counterparty risk, exchange custody risk, and funding-rate volatility.

The clearest tell is whether the issuer publishes a balance sheet alongside the reserve report. Circle does. Paxos does. Tether discloses summary balance information but not a full balance sheet at the attestation cadence. Ethena (USDe) discloses hedging positions and counterparty exposure but in a different format than traditional reserve reports.

Lens Five — Timing

The fifth lens is whether the attestation date is genuinely representative or potentially window-dressed. The standard practice is to attest to the final business day of the month. Issuers can theoretically structure reserve positions to look favourable on attestation dates and revert intra-month — though there is no public evidence that the major issuers do this systematically.

What to look for:

• Consistency of disclosure across months. If reserve composition fluctuates materially month to month, the issuer is either making meaningful portfolio changes or structuring around attestation dates. Either case warrants further questions.
• Quarter-end versus month-end. Some issuers publish more detail on quarter-end attestations than month-end ones. The information gap is itself a disclosure signal.
• Attestation timing relative to events. Attestations published shortly before major regulatory deadlines or licensing applications should be read with extra scrutiny.

USDC and PYUSD publish consistent monthly attestations with stable composition. USDT’s composition has historically varied more month to month, partly because of the larger and more diverse asset base. The pattern matters for analytical purposes even when no individual report shows a red flag.

What Operators Should Do

Reserve reports are public, free, and updated monthly. The operator playbook for using them:

1. Pull the most recent attestation for each stablecoin in your treasury or payment flow. Read it directly, not the issuer’s marketing summary of it.
2. Apply the five lenses. Composition, custodian concentration, maturity profile, off-balance-sheet exposure, timing. Note where the disclosure is detailed and where it is summary.
3. Track changes over time. Pull three consecutive months of attestations for each issuer. Material changes in composition, custodian list, or maturity profile are worth understanding before they affect peg stability.
4. Distinguish payment stablecoins from synthetic dollars. USDC, PYUSD, and licensed USDT competitors are payment stablecoins under their respective regimes. USDe and similar instruments are not — they require a different due diligence framework that examines hedging strategy, exchange counterparty risk, and funding rate exposure.
5. Build issuer review into the treasury policy cadence. Reserve report review should be a quarterly treasury task, not a one-time vendor onboarding step. The composition of stablecoin reserves changes; treasury policy needs to keep pace.

The compliant stablecoin issuers in 2026 publish enough disclosure that operator due diligence is genuinely feasible. The work is not glamorous — reading attestation reports is the unglamorous middle of treasury risk management — but it is the work that separates operators who understand their counterparty risk from those who hold stablecoins on faith.