USDe is a dollar-denominated token issued by Ethena that holds its value through a hedged trading position rather than through cash in a bank. It is routinely described as a stablecoin and routinely filed alongside algorithmic designs, and both classifications mislead. What it actually is — a tokenised basis trade, packaged so that anyone can hold it — is more interesting than either label, and understanding the mechanism explains both why it grew so quickly and why it has since shrunk considerably from its peak. It assumes you know what a stablecoin is.
01 — Not a stablecoin in the usual sense
The company itself uses the term synthetic dollar, and the precision is warranted. A conventional stablecoin holds dollars or government paper and promises to give them back. USDe holds crypto assets and an offsetting short derivatives position, and it is the combination that is designed to be worth a dollar. There is no bank account holding your dollar and no reserve of Treasury bills backing your token in the ordinary sense.
This also separates it sharply from the algorithmic designs it is frequently grouped with. Those attempted stability by issuing a companion token and manipulating supply, with the reflexive failure examined in our note on algorithmic stablecoins. USDe has real assets and a real hedge; nothing about its design creates a death spiral. It is a genuinely third category, and the risks that apply to it are the risks of running a leveraged trading position at scale, not the risks of a confidence game.
02 — How the hedge works
The mechanism is a delta-neutral position, and it is old technology borrowed from traditional finance. The protocol holds a crypto asset — principally staked ether and similar collateral — and simultaneously sells an equivalent notional amount of perpetual futures on that asset. If the price falls, the spot holding loses and the short position gains by roughly the same amount. If the price rises, the reverse. The combined value stays approximately constant in dollar terms regardless of direction.
That is the whole trick, and it genuinely works as arithmetic. The complications are operational rather than conceptual: the hedge must be maintained continuously as prices move, the short positions live on derivatives venues that hold the margin, and the collateral must sit somewhere while serving as the long leg. Each of those requirements is a dependency, and the dependencies rather than the mathematics are where the risk actually lives — the leverage mechanics being those described in position sizing with leverage on a perps DEX.
03 — Where the yield comes from
USDe attracted attention because it paid a substantial return, and the source is worth stating plainly because it is unusually legible. Two streams feed it. The first is the yield on the collateral itself, where staked assets earn network rewards. The second, historically the larger, is the funding rate on the perpetual futures: when more traders want leveraged long exposure than short, longs pay shorts a periodic fee, and the protocol — being short — collects it.
This is not magic and it is not a subsidy. It is payment for providing something the market wants: leverage. In a market where traders are enthusiastically long, being the counterparty is a profitable service, and USDe industrialised the provision of that service and passed the proceeds to holders. But notice what that makes the yield: a function of how bullishly positioned the derivatives market happens to be, measured by the funding rates covered in our note on open interest. That is a real income stream, and it is not a stable one.
The yield is a payment for taking the other side of everyone else's leveraged bullishness. That payment is generous when the market is greedy, thin when it is cautious, and negative when it is fearful — which means the product pays best exactly when it is least needed.
04 — Two tokens, and why
Holding USDe itself earns nothing. To receive the yield you stake it, receiving a second token that accrues value over time — the staked version, which appreciates against the base token rather than paying a stream of income.
The split is not arbitrary. It concentrates the yield among holders who actively opt into it, which means the base token can function as a plain medium of exchange while the staked version behaves as the yield instrument. It also matters legally: a dollar token that automatically pays interest to everyone holding it looks considerably more like a deposit or a security than one where yield requires a separate, deliberate act. Since the United States framework prohibits authorised issuers from paying holders interest for merely holding a payment stablecoin — the provision set out in what is the GENIUS Act — the structural separation between a non-yielding token and an opt-in staked one is doing real work.
05 — The cyclicality nobody priced
Here is the most important thing to understand, and it is visible in the protocol's own trajectory. Because the yield depends on funding rates, and funding rates depend on how many traders want leveraged long exposure, the product's economics are tied directly to market sentiment. In an exuberant market, funding is strongly positive, the yield is high, capital floods in, and supply expands rapidly. When sentiment cools, funding compresses, the yield falls toward or below what an ordinary money-market instrument pays, and the capital that arrived for the yield leaves.
That is not a hypothetical cycle — it has now happened. USDe's supply reached a peak substantially higher than its current level and has contracted materially, while the staked yield has compressed to something in the general vicinity of conventional short-term rates. The instrument did not break; it simply stopped being exceptional, and much of the capital that had come for the exceptional part departed. Anyone assessing this product should treat supply growth as a sentiment indicator rather than an adoption metric, because a large share of the balance is yield-seeking capital that will leave when better opportunities appear.
06 — The risks that actually matter
Four, in rough order of severity. Sustained negative funding is the central one: when the market is persistently bearish, shorts pay longs, and the protocol's core position becomes a cost rather than an income. Brief episodes are absorbable; a prolonged regime would erode the buffer and eventually pressure the peg.
Venue risk is the second and is structural: the short positions live on derivatives exchanges, and the collateral margining them is exposed to those venues' solvency and operational integrity. The protocol uses off-exchange custody arrangements to mitigate this, which genuinely reduces but does not eliminate the exposure. Execution and liquidation risk follows — the hedge must be maintained through exactly the violent conditions in which derivatives markets become disorderly, and a hedge that cannot be rebalanced is not a hedge. And there is collateral risk, since staked assets can trade away from their underlying and are not always instantly redeemable. None of these is a reason the design cannot work; together they are the reason it is a trading strategy rather than a savings account.
07 — The buffer question
The protocol maintains a reserve fund intended to absorb periods when the position costs money rather than earning it, and the adequacy of that buffer is the single most contested question about the product. The fund has been a low single-digit percentage of supply, and the protocol's retained profit in recent quarters has been modest relative to the size of the position it insures.
Whether that is sufficient depends entirely on the severity and duration of an adverse funding regime, which nobody can specify in advance. The honest framing is that the buffer is comfortable against ordinary volatility and untested against a prolonged bear market of the kind this product has not yet experienced at scale — because its entire life has coincided with conditions that were, on balance, favourable to being short perpetual futures. That is not an accusation; it is an observation about the sample size available for judging it.
08 — The pivot, and what it signals
The protocol's recent direction is the most revealing development. Faced with compressed funding yields, it has moved to generate returns from real-world assets — tokenised government debt and institutional credit — alongside the basis trade, and it operates a separate, conventionally Treasury-backed token designed for regulated markets and to reduce reliance on funding income.
Read that carefully, because it is the same pattern documented in fiat-backed versus crypto-backed stablecoins: a novel design, under competitive and regulatory pressure, drifting toward holding the same government paper as the incumbents it was built to improve upon. It is a sensible response to a real problem and it is also a concession — the basis trade alone did not prove sufficient as a durable foundation. For anyone weighing USDe today, the practical summary is that it is a well-engineered, genuinely innovative instrument whose returns are cyclical, whose risks are venue- and market-structure risks rather than reserve risks, and which is in the process of becoming somewhat more conventional than it was. Judge it as a yield strategy with real tail risk, not as a place to keep money you cannot afford to lose.
"Boast not thyself of to morrow; for thou knowest not what a day may bring forth." — Proverbs 27:1
Methodology & Sources
This report describes a protocol as at the date of publication; it deliberately contains no supply figures, yield percentages, reserve fund amounts, collateral proportions, or venue names, all of which change continuously — consult the protocol's own documentation, dashboards and current market data before acting. Structural claims reflect the protocol's published design: USDe is a synthetic dollar maintaining value through a delta-neutral position combining crypto collateral, principally staked ether and similar assets, with offsetting short perpetual futures positions; yield derives from collateral staking rewards and from funding payments received on those short positions; holders must stake USDe into a separate token to receive that yield; the protocol maintains a reserve fund intended to absorb negative funding periods; and off-exchange custody arrangements are used to mitigate exchange counterparty exposure. Risk characterisations — sustained negative funding as the principal tail risk, alongside venue solvency, hedge execution and rebalancing, and collateral liquidity — reflect the protocol's own published risk disclosures and independent analysis. Statements that supply has contracted materially from its peak and that staked yields have compressed toward conventional short-term rates reflect reported market data at the time of writing and are directional; both figures move continuously. The described expansion into real-world asset and institutional credit backing, and the operation of a separate Treasury-backed token for regulated markets, reflect publicly announced protocol developments. Nothing here is a recommendation regarding USDe, its staked version, the protocol's governance token, or any trade; the instrument is not a bank deposit, carries no deposit insurance, and is not a reserve-backed stablecoin.