KETJU Research

← The Register

synthetic-yield

Ethena (USDe)

Rejected
Max sleeve
—
Reviewed
2026-07-30 · v1
Next review
2026-10-30
Research basis
Individual research
Chains
Ethereum · sovereign

Rejected venues wait the longest for re-review; a rejection has to earn another look before the scheduled date.

REJECTED — published deliberately. USDe is a synthetic dollar whose backing has expanded beyond the original long-spot/short-perpetual trade into stablecoins, lending positions, tokenized funds, and other Risk Committee-approved assets. The core distinction remains: its dollar value depends on an actively managed reserve, centralized derivatives venues, off-exchange custodians, privileged mint/redeem roles, and market makers that alone receive direct redemption access after KYC. sUSDe yield is protocol revenue passed through, sourced from variable funding and basis, staking, and returns on reserve assets; it is not a deposit rate and can fall or turn economically negative before the reserve fund absorbs the difference. The system has handled real stress—$1.9B of redemptions over 2025-10-10 and 11 without using its reserve fund—but USDe also printed as low as 0.65 USDT on Binance during that event because venue-specific liquidity and oracle design failed. Calling this cash invites the wrong sizing. It remains a sophisticated basis-and-reserve strategy wrapped in a transferable dollar token, not a fit for a client whose mandate prioritizes sovereignty and simple, independently enforceable exits.

The research file

The mechanism

Ethena describes USDe as a synthetic dollar rather than a fiat stablecoin. A whitelisted market maker transfers an accepted backing asset and receives newly minted USDe; Ethena normally opens an offsetting short perpetual or deliverable-futures position for volatile crypto collateral, seeking to neutralize its dollar delta. The collateral stays with off-exchange settlement providers such as Copper and Ceffu and is delegated, rather than transferred, to centralized exchanges as margin. That separation reduces exchange-bankruptcy exposure but does not eliminate custodian, exchange-access, hedge-execution, margin, oracle, or legal risk.

The reserve is no longer one homogeneous ETH basis trade. Ethena’s current documentation permits BTC, ETH, SOL, liquid-staking tokens, USDC, USDT, USDtb, and other governance-approved assets, and governance updates document lending and tokenized-fund allocations. Revenue can come from perpetual funding and dated-futures basis, staking rewards, and yield on liquid stable assets. USDe itself does not automatically accrue that revenue. A holder must enter sUSDe, whose share value reflects rewards transferred into its staking contract. Ethena reports that BTC and ETH funding averaged 11% and 12.6% respectively in 2024 and sUSDe averaged 19% that year; those historical figures demonstrate where the headline yield came from, not a guaranteed forward return.

Who controls it

USDe issuance is permissioned operationally. Ethena’s docs say only addresses whitelisted after KYC/KYB can mint or redeem directly, and the protocol’s own backend is the caller that submits signed user orders on-chain. The USDe owner can replace the single minter address; the minting contract’s admin can change supported assets and custodians and grant or revoke minter, redeemer, and gatekeeper roles. Gatekeepers can halt minting and redemption. Ethena says the owner and admin are multisigs with keys distributed inside and outside the core team and seven-day delays for core changes, while its published role matrix also identifies operating minter/redeemer EOAs and multiple gatekeeper EOAs.

ENA governance and the Risk Committee influence reserve policy, but they do not make the system autonomous. The Risk Committee approves eligible assets and risk frameworks; as of the 2026 fourth term, governance reporting names Kairos Research, Blockworks Advisory, and LlamaRisk as its three voting members, with Ethena Labs non-voting. The separate reserve fund is controlled by a 4-of-10 multisig whose keys, according to Ethena’s documentation, are held by Ethena Labs contributors. The legal terms add a final control layer: Ethena BVI may suspend or decline mint/redemption services and may surrender reserve assets under a valid government order.

The record

The strongest evidence in Ethena’s favor is operational stress, not an absence of risk. In April and May 2024 the protocol processed more than $100M of USDe redemptions during a roughly 15% one-day contraction in market open interest; Ethena says the market price stayed within 20 basis points of one dollar for most of the selloff. The harder test came on 2025-10-10 and 11: Ethena’s governance report records about $1.9B redeemed, roughly 13% of supply, without reserve-fund use. On-chain pools stayed near $0.99 while processing about $790M of volume and related USDe/sUSDe liquidations on Aave and Pendle totaled about $47,000.

The same event is also the clearest warning. USDe traded as low as 0.65 USDT on Binance and 0.92 USDT on Bybit before normalizing in roughly ninety minutes. Ethena attributes the deepest print to Binance using its own order book rather than a weighted redemption-value oracle, and Binance subsequently added the redemption price to its index. Reserves were not shown impaired, but leveraged holders using a venue price were liquidated anyway. That is not a USDe solvency failure; it is proof that a synthetic dollar’s practical stability includes market structure and oracle paths outside its reserve ledger. No reviewed source identifies a successful exploit of the core USDe minting contract.

The exit

Retail exit is market-based. Anyone can sell USDe through an AMM or exchange, but only approved, KYC/KYB-cleared counterparties can use direct mint and redeem. This concentrates peg arbitrage in market makers: when USDe trades below one dollar they may buy it, redeem backing, and close the corresponding hedge. Ethena keeps liquid assets in the minting contract for hot redemptions and uses per-block limits; larger orders may need several blocks, and dependency checks can automatically pause minting and redemption. The USDe terms reserve the right to suspend or decline service and warn that reserve impairment can produce a pro-rata haircut rather than par redemption.

sUSDe adds a second queue. Governance replaced its static seven-day cooldown in March 2026 with a one-, three-, five-, or seven-day schedule tied to reserve liquidity tiers. The governing liquidity study used historical 99th-percentile redemption coverage of 4.1% for one day, 9.2% for three days, and 12.0% for seven days, each with a 1.5-times safety multiplier. That is thoughtful asset-liability management, but it confirms that the savings token is not at-will cash.

The comparison

USDC and USDT mainly transform bank deposits and short-duration securities into issuer claims; their central failure modes are custodian, bank, legal, and freeze powers. DAI or GHO mainly transform overcollateralized on-chain debt into dollars; their dominant risks are collateral, oracle, liquidation, and governance. USDe instead combines an issuer token with an actively managed portfolio and derivative hedge. It reduces direct dependence on one bank but adds off-exchange custodians, centralized derivatives venues, continuously managed hedge execution, funding and basis, and a permissioned redemption network.

A cash-like asset should become simpler as diligence goes deeper. USDe becomes more complex: the holder must underwrite every reserve asset, every venue and custodian, hedge effectiveness, legal control, the reserve fund, and secondary-market liquidity. The protocol has managed those layers competently so far. The rejection is that this client should not be paid a variable spread to own all of them while describing the position as dollars.

Open questions

The next review should reconcile each monthly custodian attestation to circulating USDe, hedge notional, venue concentration, and reserve-fund assets on the same timestamp; test whether attestations establish legal control as well as asset existence; and quantify unsecured exposure created by lending and tokenized-fund allocations. It should also track whether the three-member Risk Committee is sufficiently independent, whether live multisig signer identities and thresholds still match the documentation, and how quickly non-whitelisted holders can exit during simultaneous exchange outage and negative funding. The verdict does not depend on one bad print or one month’s yield. It changes only if the structure offers a permissionless, independently enforceable redemption path and materially reduces the active-management and centralized-counterparty stack.

Sources

The claims above trace to these. Where a number could not be independently verified, the thesis says so.

Inherited controls

The verdict above grades the protocol layer. Every position also inherits the asset it holds and the chain it settles on. The least safe layer sets the position’s grade, and the position table names which one that is.

ChainVerdictGradeControl constraint
EthereumApproved sovereign No sequencer, no upgrade key, no operator who can be compelled — rule changes require social consensus.
The memo is public. The watching is the product: the terminal reads your clients’ wallets against this Register and flags the events above when they fire. $49 per advisor per month, first 14 days free. Start the trial.