What Is Inside a Stablecoin's Reserves: Cash, Bills, Loans and the Liquidity Ladder

The short answer

A stablecoin reserve report gives you a ratio and, if it is a good one, a table of categories. The ratio tells you whether the issuer’s assets are worth at least as much as its tokens. The categories tell you whether the issuer can turn those assets into dollars on the timetable its redemption terms promise.

Those are different questions, and the second one is the one that matters in a stress event. This guide is about the categories. The reserve attestation guide covers the document’s scope and its limits; the two are meant to be read together.

How full reservation is actually defined

Tether’s FAQ states the definition without hedging: the platform is “fully reserved when the sum of all Tether Tokens in circulation is less than or equal to the value of our Reserves.”

Read that sentence as an accountant would. It compares two valuations:

value(tokens in circulation) ≤ value(reserves)

Both sides are numbers on a date. The left side is a count of tokens multiplied by one dollar. The right side is a portfolio marked at market, or at amortised cost, or at face — the report will say which. Nothing in the inequality refers to maturity, to the identity of the counterparties, or to how quickly the right-hand side could be converted into the left-hand side’s unit.

Tether’s transparency page states the same idea in a second form — “the Tether Issuer’s assets exceed its liabilities” — and describes the tokens as pegged 1-to-1 and backed 100% by the reserves. Both statements are about value. Neither is a statement about liquidity, and the difference between the two is the subject of this article.

Four categories, four liquidation profiles

Tether’s FAQ names the components of its reserves: “traditional currency, cash equivalents, and other assets, including receivables from loans made by Tether to third parties.” It separately explains why it holds bitcoin and gold. That is four categories, and they behave nothing alike when a redemption queue forms.

Cash and bank deposits. The fastest asset on the ladder and the one with the most operational friction. Converting a deposit into a payment to a customer runs through the banking system: cut-off times, business days, correspondent banks, and the receiving bank’s own handling. The blockchain leg of a stablecoin redemption is fast; this leg is not, which the USDC redemption guide sets out in detail.

Cash equivalents and short-dated government paper. Treasury bills are close to cash in credit terms and not identical to it in settlement terms. A bill matures on a date; selling one before maturity requires a dealer, a price and a settlement cycle. Under normal conditions that is a same-day or next-day operation. Under stress, the price is available and the buyer may not be.

Receivables from loans to third parties. An asset with a named counterparty and a term. It can be money-good and still illiquid: a loan that matures in ninety days does not pay a redemption that arrives tomorrow. This is the category where a reserve ratio and a redemption promise diverge most sharply, and it is the category a reader should look for first in a report’s notes.

Bitcoin and gold. Volatile, market-priced, and liquid in normal conditions relative to most assets. Tether explains the holding as an inflation hedge and says it maintains “a healthy balance of Net Equity (excess Reserves of the Tether Issuers) to act as a cushion against the volatility of bitcoin”. The same FAQ adds the sentence that matters most for a holder: “the dollar value of Tether’s Bitcoin holdings may change significantly following its most recent Reserves report.”

That is an issuer telling you that one slice of the reserve is marked at a price which moves after the report is published. The ratio in the report is therefore a ratio at a moment, and the cushion is the mechanism that absorbs the movement between moments.

Why a ratio can be right and a redemption still slow

Consider two issuers, both at 100%.

Issuer A holds overnight deposits and bills maturing within weeks. A redemption request is a payment instruction against assets that are already cash or about to become cash. The constraint is the banking rail, not the portfolio.

Issuer B holds a mix that includes term receivables and a volatile allocation. The ratio is 100% on the report date. A large redemption arrives, the fast assets are used, and the remainder has to wait for a loan to mature or for a buyer to appear. The ratio was never wrong. The maturity profile was never disclosed in the ratio.

Neither issuer has done anything dishonest. The point is that “100% backed” is compatible with both, and a holder deciding where to keep working capital is choosing between them.

The cushion, in arithmetic

Tether’s FAQ says the issuer maintains net equity as a cushion against bitcoin volatility. That sentence deserves an example, because it is the part of the structure that most readers misread as a guarantee.

Suppose an issuer holds reserves of 102 units against 100 units of tokens, with 6 of the 102 in bitcoin. The ratio is 102%. Now let bitcoin fall by 30%:

• The bitcoin slice falls from 6.0 to 4.2, a loss of 1.8 units.
• Reserves become 100.2 against tokens of 100.
• The ratio falls to 100.2%.

A 30% drawdown in the volatile slice consumed 90% of a 2-point cushion. Had the cushion been 1 unit instead of 2, the same move would have put the issuer below 100% on a mark-to-market basis — not insolvent in the sense of being unable to pay a small redemption, and below the line the issuer itself publishes as its test.

The example is constructed arithmetic, not a description of any issuer’s portfolio. The structure it illustrates is general: the cushion absorbs a percentage move in the volatile slice, and the size of the move it can absorb is the cushion divided by the size of the slice. Those are both numbers a reader can find in a report, and both are more informative than the headline ratio.

The cadence gap

There is a second, quieter problem: the two sides of the ratio are not published on the same schedule.

Tether states that it typically publishes token circulation daily and reserve reports quarterly, prepared by BDO Italia in accordance with criteria established by the IAASB, and that it “does not commit to update the information at any particular interval, regardless of any changes in the number of Tether Tokens in circulation or the value or composition of the Reserves.” It also warns that the information may be delayed.

So between quarterly reports, one side of the inequality is refreshed daily and the other is not. If circulation grows, the ratio computed against a stale reserve figure mechanically falls; if the volatile slice moves, the ratio moves without any report being published. Circle’s transparency page, read on 2026-10-29, describes its reserves as reported weekly, a shorter interval and a different compromise.

Neither cadence is dishonest. Both mean the ratio you read is a historical figure with a publication lag, and the publication lag is longer than the volatility of the assets in the reserve. That is exactly why the category mix matters more than the last published percentage.

A checklist for reading the reserve section

1. Find the categories, not the total. Cash, bills, receivables, volatile assets. If the report gives one line, it has given you nothing to work with.
2. Look for maturity or term language. “Cash equivalents” is a category; “maturing within 90 days” is a constraint. The second is what tells you about redemption capacity.
3. Identify the counterparties behind receivables. A loan to a bank and a loan to an affiliate are different assets with the same label.
4. Check the mark. Market value, amortised cost and face value produce different ratios for the same portfolio.
5. Find the cushion and divide it by the volatile slice. That is the percentage move the structure can absorb.
6. Compare cadence with volatility. A quarterly report on a portfolio containing bitcoin is a slower signal than the asset it describes.

The 1:1 promise itself is a contractual term, not a balance-sheet fact, and the minting versus buying guide sets out what direct issuance and redemption actually cost at the issuer. For how the two largest tokens differ in structure, the USDT and USDC comparison covers the rest. And for the closest traditional-finance analogue, the money market fund comparison shows what a fund does with the same assets under a different rulebook.

Risk note. The reserve descriptions in this article were read from Tether’s FAQ and transparency pages and from Circle’s transparency page on 2026-10-29, and are quoted as those pages state them. This article does not reproduce any issuer’s reserve percentages, does not audit any report, and makes no claim about the current composition or solvency of any issuer. The arithmetic examples are constructed for illustration. Stablecoins can trade away from their peg; digital assets are not deposits and are not insured; nothing here is investment advice.