Ethena Labs operates a synthetic dollar protocol (“USDe”) that has positioned itself as a scalable stablecoin alternative and its most attractive feature driving its rapid growth was its exceptional yield, with historical rates reaching up to 20% APR, significantly higher than traditional stablecoins that typically offer minimal or no yield.

USDe maintains its dollar peg and delivers these high yields through delta-hedging, using perpetual futures contracts to offset price volatility in its crypto collateral (primarily staked ETH and BTC). This strategy not only helps stabilize the peg but also generates yield from:

• Staking rewards from assets like staked ETH
• Funding rates from perpetual futures markets
• Basis spreads between spot and futures prices

However, the Bybit crypto-asset exchange hack in early 2025 illuminated critical weaknesses in Ethena’s risk management architecture, revealing how off-exchange settlement mechanisms protect custody but transfer counterparty risk to exchanges rather than eliminate it.

Multiple USDe depegging events also saw the “synthetic stablecoin” trading significantly below its dollar peg during periods of market stress, most recently during the crypto market rout of October 10, 2025.

These events have seen USDe shrink substantially from its original strategy of delivering a synthetic stablecoin with higher yields through delta-hedging. Today, USDe is largely backed by other stablecoins, compressing yields and raising uncomfortable questions about its overall value proposition.

In this case study, we examine how the scares caused by multiple depegging events has seen USDe shrink from its original mission, in particular illuminating the risks it contended with previously including:

• Ethena’s use of Copper’s ClearLoop product for custody protection, that concentrated trading counterparty risk on crypto-asset exchanges rather than distributing it;
• Ethena’s reserve fund, while sufficient for isolated incidents, also relied heavily on operational timing of the market, rather than acting as a comprehensive risk management framework; and
• recent USDe depegging events which highlighted issues with Ethena’s USDe stability mechanisms.

Many of USDe’s organic risks are now significantly lower, given the synthetic stablecoin’s shift into more liquid stablecoin backing.

Had Ethena not walked back from its earlier strategy, it would have effectively recreated pre-2008 traditional finance systemic risk architecture without access to bailout facilities large enough to address catastrophic outcomes and as more traders started to treat USDe as the equivalent of a dollar.

Ethena’s USDe is an innovative approach to a synthetic stablecoin, but its current iteration is also an acknowledgment that risks associated with such protocols can only be redistributed, and not hypothecated into non-existence.

As stablecoins continue to grow in popularity, Ethena’s USDe serves as a timely reminder of the risks baked into non-reserve-backed stablecoins, especially those with multiple cross-dependencies, and can help inform assumptions made by crypto-asset traders, as to the stability of such stablecoins.

2.1 Copper’s ClearLoop Product

Ethena initially launched using Copper’s ClearLoop solution, a UK-based custodian’s off-exchange settlement (“OES”) product that has been supporting institutional digital asset trading since 2018. The integration allows institutional clients to deploy capital and trade immediately on exchanges while keeping their crypto-assets in independent custody.

2.2 How Copper’s ClearLoop Works for Ethena

ClearLoop enables trading and settlement to clear in under 100 milliseconds while maintaining custody protection, with all crypto-assets held in Copper’s client segregated custody protected with multi-party computation (“MPC”) wallets. The process works as follows:

1. Ethena deposits collateral (for instance bitcoin) with Copper as custodian, then connects their Copper account to crypto-asset exchanges via ClearLoop.
2. Funds become available instantly on the selected crypto-asset exchange, providing immediate market access.
3. Ethena opens short perpetual futures positions to hedge their spot crypto-asset holdings.
4. Copper settles trades instantly between parties after execution.
5. The actual crypto-assets remain in Copper’s custody rather than being held on the crypto-asset exchange.

Copper’s users’ funds are held within a bankruptcy-remote trust, meaning that in the event of Copper’s failure, users’ funds are not expected to be part of Copper’s bankruptcy estate. This custody protection proved effective during exchange failures, with user funds remaining wholly available.

Ethena’s delta-neutral strategy attempts to capture funding rate payments as yield for USDe holders, marketed as providing “risk-free” yield through arbitrage between spot and futures markets. For instance, Ethena could hold bitcoin and holding bitcoin as collateral means they would also need to fund a short position in bitcoin perpetual futures (because as the price of bitcoin goes up, the price of the bitcoin perpetual futures goes down and vice versa).

Although Ethena still engages in this delta-neutral strategy, at the time this case study was prepared, the synthetic stablecoin has walked back significantly from this approach, with as much as two-thirds of the assets backing USDe held in other stablecoins.

2.3 The Assumption of Positive Funding Rates

2.3.1 What are Positive Funding Rates?

A positive funding rate means that traders holding long positions pay traders holding short positions. Perpetual futures contracts don’t have an expiration date (unlike traditional futures). To keep their prices anchored to the spot market price, exchanges use a “funding rate” mechanism – periodic payments between long and short traders.

When funding rates are positive (Bull Market Scenario):

• there’s high demand for long positions (traders betting on price increases);
• this buying pressure pushes the perpetual contract price above the spot price;
• to rebalance, longs pay shorts every 8 hours (typical funding period); and
• this discourages excessive longs, incentivizes shorts, and brings prices back in line.

2.3.2 How Positive Funding Rates Benefit Ethena

In bullish conditions, high demand for long positions can push the prices of these positions above the spot price, triggering a positive funding rate where long holders pay fees to short holders.

Since Ethena holds:

• Spot BTC/ETH (long exposure)
• Short perpetual futures (short exposure)

When funding is positive, Ethena:

1. Receives regular funding payments from longs (yield generation);
2. Maintains delta-neutral exposure (spot gains offset by futures losses, and vice versa); and
3. Earns the “carry” between spot and futures.

For instance, if the funding rate is +10% annually:

• You hold $100 million in spot ETH.
• You short $100 million in ETH perpetual futures.
• ETH price is neutralized (no directional risk).
• But you earn approximately $10 million per year from longs paying you funding.

2.3.3 When Funding Turns Negative (Bear Market)

When funding rates are negative, Ethena must pay traders with long positions to keep its short positions open, essentially reversing the cash flow – instead of earning yield, the protocol hemorrhages capital.

How Ethena Handles This:

• Reserve Fund Protection: An Ethena reserve fund exists specifically to step in when combined revenue from staking yields, funding rates, and stablecoin holdings turns negative, protecting the spot backing of USDe.
• User Protection: Ethena absorbs all payments due to negative funding rates – holders of USDe/sUSDe are not responsible for any losses and never receive negative yields.
• Dynamic Allocation: During periods of low or negative funding, Ethena shifts more backing assets into liquid stablecoins earning approximately the U.S. Treasury rate, reducing exposure to negative funding.

2.4 What Causes Negative Funding Rates

In bear markets when short positions become more expensive than spot prices, the funding rate turns negative and short holders must pay long holders. This happens when:

• Traders close long positions and open shorts during price crashes.
• Market sentiment turns bearish and demand for leverage shifts to shorts.
• Sharp corrections trigger mass liquidations.

The largest negative funding rate event occurred in September 2022 when Ethereum transitioned from Proof-of-Work to Proof-of-Stake, with funding rates dropping to -300% at one point. Other instances include the Iran-Israel crisis in April 2024 and the FTX collapse.

2.5 The Risks to Ethena’s Synthetic Stablecoin Model

2.5.1 The Reserve Fund Risk

A key risk faced by Ethena, or indeed any synthetic stablecoin considering a similar strategy is whether the reserve fund is large enough during periods of significant redemptions. Some analysts suggest that to withstand a bear market, Ethena would need to sustain a “keep rate” above 32% (meaning 32%+ of revenues go to reserves rather than users) at a $10 billion market cap.

Historically, funding rates have shown mean-reverting characteristics – negative periods don’t persist, with the longest streak of consecutive negative funding days lasting just 13 days compared to 176 consecutive positive days.

Even during the 2022 bear market, BTC and ETH funding rates averaged 7.8-9% annually, and when combining staked ETH income with funding rates, only 8.84% of days showed negative combined revenue.

The main risk isn’t whether negative funding can happen (it will), but whether the reserve fund grows proportionally with USDe’s market cap to weather extended bear market periods.

On the flip side, maintaining too large a reserve fund acts as a drag on yield generation, and balancing efficient capital allocation to these two competing requirements is a challenge. The challenge of maintaining this balance is exacerbated by the rapid settlement times in decentralized finance (“DeFi”) as well.

Unlike in traditional finance (“TradFi”), where it can take several days for settlement, DeFi participants expect instant settlement, allowing them to both enter and exit investments with near real-time settlement.

While DeFi may be more quick than TradFi, that efficiency comes at the price of stability and exaggerated volatility, especially during periods of market stress.

2.5.2 Risk Transfer to Exchanges Despite ClearLoop Protection

While Copper’s ClearLoop provides custody protection for Ethena’s collateral, the fundamental issue remains that the mechanism transfers counterparty risk to crypto-asset exchanges rather than eliminating it altogether:

Crypto-asset exchanges are required to post collateral to participate in ClearLoop, but they still accept exposure to Ethena’s trading positions. When Ethena (or similar protocols) maintain large short perpetual positions on a crypto-asset exchange:

• the crypto-asset exchange becomes the counterparty to those positions;
• if Ethena suddenly withdraws those positions during times of market stress, the crypto-asset exchange faces concentrated unwinding risk; and
• protocols using similar strategies to Ethena can create systemic pressure on crypto-asset exchanges collectively when the market moves downwards in unison.

Just as was demonstrated in the 2008 Financial Crisis, the risk has to end up somewhere ultimately. There are no magic solutions to hypothecate the risk into non-existence.

The perpetual futures exchange needs control over the crypto-assets to facilitate the marked-to-market process and settle up accounts. If a custodian like Copper can cut that access off, that perpetual futures exchange is at risk. If a custodian like Copper cannot cut that access off, then the crypto-assets are under the exchange’s control.

That allocation of risk is binary and there is no middleground.

If the crypto-assets are held with a custodian like Copper, it’s also possible that more than one crypto-asset exchange thinks they have control over the same set of crypto-assets and the bigger question is if it’s even possible to mitigate that risk?

The Bybit incident demonstrated this dynamic in practice.

While the Bybit integration with ClearLoop meant Ethena’s margin collateral was protected through Copper’s custody, Bybit still bore the risk of Ethena’s positions unwinding. When Bybit was hacked, Ethena’s custody protection worked as designed, but Ethena ultimately still relied on its reserve fund and on other crypto-asset exchanges to absorb losses through mutual insurance arrangements.

What would have happened if Copper publicly announced it would not allow any more crypto-assets to go to Bybit because it was hacked? It’s possible Bybit may have collapsed.

Alternatively, if Copper had said it would not stop transfers to Bybit and crypto-assets were then somehow swept into the ongoing hack and lost forever, was Copper ever really a custodian? These are uncomfortable but necessary questions.

2.5.3 The Combination of the Reserve Fund and Exchange Risks

While a reserve fund may be large enough to cover positions on any single crypto-asset exchange’s failure, this creates a problematic dynamic because if a reserve fund exceeds all its combined exchange positions, the synthetic stablecoin then simply earns the Fed funds rate plus a small spread, as idle cash dominates the yield profile, and this is what we’re seeing now in Ethena.

Alternatively, if a reserve fund follows the formula:

Reserve Fund > n × p

Where,

n = the exchanges which collapse due to coordinated ClearLoop withdrawals
p = the limit of assets on each exchange that the synthetic stablecoin has exposure

then this is the textbook definition of systemic risk.

The reliance on other crypto-asset exchanges to mutually insure losses (as happened with the Bybit hack) is itself a plan that depends on systemic risk not materializing.

However, for crypto-asset exchanges which deal almost entirely in crypto-asset derivatives like Deribit, or for larger products, such arrangements could prove catastrophic.

3.1 Impact Analysis

In early 2025, Bybit suffered a significant hack that put Ethena’s collateral positions at risk. The incident was manageable primarily because:

• Ethena’s collateral was protected through Copper’s ClearLoop custody solution.
• Bybit’s perpetual futures positions were relatively small compared to the crypto-asset exchange’s customer deposits.
• Bybit had sufficient capital reserves to absorb losses from the hack.
• Other crypto-asset exchanges agreed to mutual loss-sharing arrangements.

3.2 What the Incident Revealed

The Bybit hack illuminated three critical issues with a synthetic stablecoin model like Ethena’s:

Custody protection works, but counterparty risk remains. While ClearLoop protected Ethena’s collateral from the crypto-asset exchange going bankrupt, crypto-asset exchanges still bore the risk of concentrated positions unwinding.

Short-term user protection creates long-term bankruptcy claims in severe scenarios where multiple crypto-asset exchanges face similar stress.

Avoiding multiple failure modes is critical. The system’s stability depends on multiple failure modes not occurring simultaneously.

Bybit was manageable because the leveraged perpetual positions were small relative to customer deposits. Pulling the plug on perpetual collateral didn’t matter much overall in that case. But for a primarily-derivatives exchange like Deribit, or for a larger product with more concentrated exposure, a similar hack could have been catastrophic.

In that sense the Bybit hack was more of a warning of the risks associated with synthetic stablecoins, then a proper stress test.

The survival of perpetual futures crypto-asset exchanges now depends on:

• traders’ willingness to keep ClearLoop lines open during times of stress;
• retail depositors effectively bailing out leveraged trading positions; and
• crypto-asset exchange retained earnings always exceeding potential losses—essentially assuming no crypto-asset exchange can become insolvent.

Deribit, for example, does not have a custody business nor a large pile of customer assets used for unlevered trading. To the extent big problems get socialized across an entire business there are perpetual futures crypto-asset exchanges for which such derivative products are the entire business and there is nowhere to help cushion losses.

Note that in the 2008 Financial Crisis, it was the specialized finance companies, asset-financing funds, and mortgage lenders that blew up first at least in part because they had no other business lines to stave off bankruptcy.

Synthetic stablecoin frameworks, of the sort that Ethena’s USDe had previously developed, essentially recreate the pre-2008 TradFi architecture, but without a bailout facility large enough to address systemic failure scenarios.

While the current structure of USDe no longer resembles Ethena’s original design concept, note that there is nothing stopping Ethena from reverting USDe’s stabilization framework and reserve fund mechanics to its initial intent, bringing with it the same risks articulated above.

4.1 Recent Depegging Events

Ethena’s USDe has experienced several depegging events where the token traded materially below its dollar target, with the most recent being the market rout on October 10, 2025.

Such depegging events can occur when:

• funding rates turn negative, reducing yield attractiveness;
• exchange counterparty concerns increase redemption demand;
• market participants front-run potential position closures; or
• liquidity in secondary markets becomes fragmented.

Ethena’s strategy is to move into Fed funds-type yield-bearing products when funding is low or negative and is essentially market timing that can be best described as, “I’ll take risk when it’s a bull market and cash when it’s a bear market.”

The recent crash of October 10, 2025 clearly demonstrated that this transition can occur far more rapidly than it can be rebalanced, which is what led to more significant USDe’s depegging on the centralized crypto-asset exchange Binance, but more restrained responses on platforms like Curve Finance, a decentralized crypto-asset exchange on Ethereum.

Insofar as synthetic stablecoins go, DeFi is a double-edged sword – the speed and versatility afforded to DeFi traders works both ways, the ebb and flow of liquidity can be far more rapid than the ability to respond to such changes, and this is why synthetic stablecoins like USDe need to maintain a buffer.

4.2 Withdrawal Buffer Analysis

Ethena maintains a withdrawal buffer structure designed to handle redemption flows and at the time this case study was prepared:

• approximately $70 million USDT sat in the V2 Mint/Redeem contract;
• the buffer is replenished after significant outflows; and
• daily outflow capacity managed as much as $250 million without issues immediately following the Bybit hack.

In the aftermath of the Bybit hack, Ethena claims to have scaled their buffer up to as much as $250 million to cater for redemptions. Similarly, Ethena’s founder Guy Young also suggested that the platform was able to have withstood redemptions of over “$2b in a 24hr period with zero issues” during the more recent crash of October 10, 2025.

It’s worth noting that on both occasions when Ethena was severely tested, the stresses came from a single point of weakness – a single centralized exchange.

It is less clear if synthetic stablecoins would have sufficient resilience to withstand a systemic challenge, especially involving multiple centralized exchanges all using ClearLoop.

Stablecoins like USDe are also limited in how much liquidity can be buffered because that would necessarily bring down the yield available for distribution from the cash-and-carry trade that has made USDe attractive up to this point.

And this compromise is reflected in the current asset mix behind USDe today, which is heavily in liquid stablecoins. At some point this could change, as USDe seeks to preserve market share, and when that changes, the same myriad risks that led to USDe’s depegging could return.

4.3 Redemption Limitations

In addition to the quick replenishment mechanism detailed earlier, at the time this case study was prepared, there was a $10 million per block redeem limit. This hard cap means that during extreme scenarios:

• redemption demand can significantly exceed available capacity;
• secondary market prices for USDe can deviate substantially from its dollar peg;
• users bear timing risk and potential losses while waiting for redemptions; and
• the USDe’s “stablecoin” exhibits stability characteristics more akin to a limited-redemption structured product.

At the time this case study was prepared, many of these issues have been managed by having the bulk of USDe’s backing in stablecoins. But again, that’s an unideal compromise, given that the yields from USDe will come down dramatically from the expectations and perceptions of investors.

Again, the asset mix backing USDe could change at any time, and Ethena’s more risk-off approach may only be temporary.

To Ethena’s credit, the protocol has not, and does not claim to be riskless and acknowledges the limitations built into USDe openly. While the setup adopted by Ethena for USDe makes volatility unlikely, it also does not address the systemic risks built into the overall structure of their synthetic stablecoin design.

Ethena Labs has created an innovative synthetic stablecoin architecture with unavoidable and unideal tradeoffs, highlighting the challenges with building a synthetic stablecoin:

• Custody Protection But Counterparty Risk Transfer: Copper’s ClearLoop provides effective custody protection, but the mechanism concentrates trading counterparty risk on crypto-asset exchanges rather than eliminating it, creating systemic vulnerabilities.
• Reserve Structure: The reserve fund addresses incidents but creates impossible trade-offs at scale—either earning only Fed funds rates or accepting systemic failure risk.
• Depegging Vulnerability: Redemption limitations and structural dependencies can cause periodic depegging but also affect other interdependencies.

Synthetic stablecoins have the potential to recreate the foundation for a pre-2008 TradFi implosion in DeFi without the commensurate bailout facilities. The greater risk with every successful stress test of synthetic stablecoins is that traders start to treat them as the functional equivalent of other reserve-backed stablecoins.

As with Terra-Luna and other large-scale failures, the key is to identify what’s automated and operating faster than manual interventions can respond.