• Largest notional liquidation event ever, with the largest ever drawdown in open interest

• Highest ever 24hr volume in perpetual futures markets

• $14bn drawdown in open interest roughly confirms estimates of $10bn in liquidations

• Despite market volatility, zero issues on USDe which moved in line with fiat stablecoins

• Perpetual futures contracts discounted by as much as 5.8% vs spot markets during the sell-off

• Ethena uniquely positioned to capture the discount via shorting perpetual futures. Over $500k of protocol revenue was realized last week from unwinding contracts and capturing the discount

• Automated unwind of Ethena’s underperforming contracts helped funding revert positive, with over $1bn of Ethena perpetual futures unwound and allocated to yield bearing stablecoins

Historical Liquidation Event

Perpetual futures markets just underwent their largest notional liquidation event ever, with most of the volatility concentrated in ETH markets. It marks the ninth drawdown of at least one billion dollars in ETH open interest since Ethena launched, and the second drawdown over $2 billion. In just 24 hours, ETH open interest fell by $2.3bn, the largest 24hr drawdown ever.

Over the course of a week, ETH open interest is down over $5bn, a weekly drawdown of over 25%.

Total open interest across all assets fell over $20bn in the last week, with BTC contracts losing notionally less than ETH, despite the BTC market being almost twice the size. $14bn of that total open interest was wiped in 24hrs between Feb 2nd and Feb 3rd.

The dramatic sell off resulted in the highest 24hr volume in perpetual futures markets ever.

It's hard to judge how much of that volume and open interest was a result of liquidations - as a result of API limitations on how many feeds are pushed out per second, exchanges often underreport their liquidation figures and third party data providers present understated data as a result.

Out of the ~$14Bn of total open interest that was wiped on the 3rd of February, apparently only $2.3bn of that was liquidations.

However, as the founder and CEO of Bybit tweeted after the sell-off, Bybit alone was responsible for $2.1bn of liquidations vs the $333m reported by Coinglass.

He estimated total liquidations to be in the range of $8-10bn. That more closely correlates to the open interest drawdowns observed on each exchange over the 24hr period, with Bybit total OI down $4bn from Feb 2nd to Feb 3rd, and $5bn on Binance alone.

Assuming that logic holds, this was one of, if not the largest liquidation events in crypto history. A stress test like that gave Ethena the opportunity to prove the resiliency of USDe once again in the face of volatility.

USDe Resilience

Despite being the largest notional liquidation event in crypto derivatives to date, everything functioned as expected, with no disruptions to redemptions and virtually no losses suffered from unwinding perpetual futures contracts, which were trading at a discount to spot markets.

$50 million in USDe redemptions were processed in 24hrs via primary markets with zero issues. Secondary USDe markets saw over $350m in 24hr volume as USDe remains within 10bps of USDT.

USDe price moved in line with USDC and DAI during the volatility as market makers stepped in to close any arbitrage opportunities between primary and secondary markets.

sUSDe and USDe collateral across DeFi functioned as normal (e.g., Aave, Morpho, Fluid, Curve, Pendle), with no material liquidations, unwinding, or liquidity issues.

As prices moved lower, many perpetual futures contract were trading for less than their corresponding spot markets, presenting an opportunity for Ethena to realize profits from unwinding perpetual future positions.

Capturing Profits during the Sell off

What is commonly under-appreciated about Ethena’s design is that during market volatility to the downside, Ethena is positioned on the “right” side of the trade, being short perpetual futures and long spot. During market sell offs, it is quite common to see perpetual future contracts discount in value versus their equivalent spot contracts due to liquidations and general panic in futures markets. These discounts have surpassed 5% in previous market sell offs, and the same behaviour was observed this time around.

Being short a perpetual futures contract that is trading at a discount leads to higher unrealized PnL and consequently more margin collateral buffering Ethena’s already unlevered positions. As Ethena’s system automatically unwinds from perpetual futures, Ethena’s positioning naturally benefits from these dislocations, placing the protocol on the stronger side of the market.

By closing perpetual future contracts that have dislocated to the downside, the protocol can capture that discount as realized PNL and is able to use that revenue for the benefit of sUSDe holders.

For example, the Binance ETH coin-m contract saw a discount of -5.8% vs spot during the sell-off, charted below in bps. With an Ethena short position of approximately $200m on this contract, the theoretical unrealized PNL at the peak discount was over $11m - which could have been captured as protocol revenue were the full position to be unwound.

The discount compared to spot wasn’t as significant on other exchanges such as Bybit and OKX, who had discounts closer to 1% on ETH contracts. BTC contracts across the three exchanges held up better than ETH, as the sell off was mainly concentrated on ETH contracts and specifically coin-margined (inverse) contracts.

Binance perpetual futures generally saw wider discounts across all contracts compared to other exchanges.

The dispersion occurring among contracts and assets puts more emphasis on execution quality and Ethena’s automated execution ensures dislocations are taken advantage of.

Over the course of the week, over $500k of Ethena’s $5.5m of protocol revenue was attributable to both capturing the discount between perpetual future contracts and spot and unwinding the lowest funding rate contracts, adding an estimated extra 50bps to sUSDe’s weekly APY. This realized PNL also provided an additional buffer for any negative funding cycles we witnessed over the weekend.

Adapting to Lower Funding Rates

With volatility picking up and a wide dispersion in performance across both exchanges and assets, it is important Ethena reacts quickly to any structural changes in perpetual futures markets. Over the last week, funding rates on ETH contracts moved significantly lower than BTC, with both moving lower than the levels observed at the start of the year.

Structural changes in funding rates are an indicator for Ethena to begin shifting the allocations of the assets backing USDe. With yield-bearing stablecoins like USDS available as eligible backing assets, there is an opportunity to move more of the backing into liquid stables earning up to 8.75%, and less towards underperforming perpetual futures contracts. Funding rates improved as Ethena began to unwound short ETH positions.

Over the last week, the protocol demonstrated its ability to react quickly to market moves, shifting an additional $1bn into stablecoins and away from BTC and ETH contracts in the face of low funding rates.

Ethena’s dynamic approach to allocation is illustrated in the below comparison of the backing assets at the start of February compared to today. USDe is now backed by almost $1.8bn in liquid stables, or 29% of the backing, some of which yielding more than perpetual futures funding rates. As long as these perpetual future contracts underperform, Ethena will continue to allocate to liquid stables and optimize the assets backing USDe.

Despite the largest notional liquidation event in history, USDe proved its resiliency and seamlessly processed redemptions while remaining at peg. The ability of Ethena to capture profits during market sell-offs further demonstrates the structural design advantage of being short perpetual futures, and how it can benefit the interests of sUSDe holders in market downturns.

With every stress test, USDe proves itself as one of the most resilient assets in the space and Ethena can hopefully continue to earn the trust of our users.

This post walks through the history and evolution of the allocation strategy for USDe’s backing starting from launch through today. The makeup of the assets backing USDe has adapted over time based on observations and market feedback in order to mitigate risk, improve liquidity, and provide diversified revenue sources for the protocol.

Additional checks and more transparent decision-making processes regarding asset allocation and overall strategy have been implemented with the Risk Committee. In line with evolving market conditions, Ethena has made adjustments to how it approaches constructing a synthetic dollar and this post aims to highlight the thought process behind those adjustments.

The Beginning

When the initial framework of Ethena was created in April of 2023, the vision was to primarily use ETH and ETH LSTs with the delta being hedged by perpetual & dated futures contracts. The idea was borne out of Arthur Hayes’ article Dust on Crust, outlining a vision for a synthetic dollar backed by BTC. The core team felt this was an idea worth exploring, and as far as backing assets go, potentially improving on the idea with forms of staked ETH as backing to naturally increase revenue and mitigate potential negative funding risk.

At the time, the stETH yield of 6-7% provided an ample buffer for instances when funding rates dipped negative. Funding rates historically average approximately 8% annualized, and so with a spot backing anchored by stETH in a hypothetical scenario it would take almost a 2x funding rate move to the downside before the protocol was exposed to negative revenue. In theory, this was the ideal foundation upon which to build a synthetic dollar.

However, as the amount of ETH staked increased, staked ETH yield fell from 6-7% to roughly 3% by the time USDe launched in February 2024. As those yields fell, staked ETH made less sense as the backbone spot asset for USDe from a risk perspective for a few reasons:

1. As a “derivative” of ETH, stETH is naturally less liquid than its underlying, exposed to risks of depegging and market discounts due to diminishing secondary market liquidity or potentially a slight delay in redeeming via a withdrawal queue. On a handful of occasions, these issues have come to a head in the market and resulted in stETH’s market price deviating from that of the underlying ETH. Since stETH yields are hovering around 3% and will likely only decline, holding a less liquid asset as a significant portion of the backing of USDe became less appealing from a risk perspective, particularly in an environment where short term US T-bills were earning close to (or in excess of) 5%.

2. As demand for USDe grew, it became clear that other assets would be needed sooner than expected to keep pace. As USDe became the fastest growing USD asset ever to reach both $2bn and $3bn in supply, it was clear that limiting Ethena to the ETH perpetual future market would eventually be a bottleneck for growth - one day having too large of an impact on ETH markets and lowering funding rates as a result, while exposing the backing to liquidity risk if Ethena-related positions constituted too great a portion of open interest. To address this, Ethena began to support BTC as a backing asset to open up an untapped futures market to hedge with.

Prioritizing Liquidity and Stability

As USDe grew to become one of the biggest USD denominated assets in the industry, minters began to show a preference to mint & redeem USDe with stable assets, such as USDT and USDC, rather than using ETH or LSTs. As a result, to facilitate on-demand 24/7 redemptions, the protocol began to maintain a balance of stable assets, such as USDT. This buffer value was initially set at ~5% of the total circulating supply of USDe with the ultimate goal of limiting downside to protocol revenue while still being able to manage large scale redemptions.

At the time Ethena launched publicly, funding rates were extremely elevated, in some weeks reaching as high as 60% annualized. As a result, Ethena generated over $8m in weekly protocol revenue on launch week. The significant levels of revenue enabled the protocol to grow the Reserve Fund by as much as $5m per week to enhance overall protocol stability, ultimately growing the Reserve Fund to over $46m today.

In April/May 2024, Ethena underwent its first stress test as it handled the first large scale redemption event in the face of a significant market open interest reduction, which reached approximately 15% in just one day (the largest single-day open interest reduction in over a year) - over 100m USDe redeemed. In the face of this extreme market event, USDe market price remained within 20bps of $1 for the vast majority of the sell off. This event was a valuable demonstration that liquidity should be a constant priority for the backing of USDe, and since that event the protocol has maintained a sufficient liquid buffer to handle mass redemption events.

Where We Are Today

Staked ETH

In line with the above principles of prioritizing liquidity and allocations that align with risk profiles, the protocol has reduced the proportion of ETH LSTs compared to the total backing assets to less than 10%.

Liquid Stables

During the last three to four months, the return profile offered from hedging the delta of the spot digital assets on certain perpetual futures contracts has been less than the risk-free rates offered on US treasuries. Adapting to this reality, Ethena began to explore alternative means to produce a positive (or at least neutral to risk-free) return profile, reduce volatility of the return profile, as well as address potentially long term & structurally negative/poor funding rates if that were to eventuate.

In an environment when certain perpetual contracts are yielding less than US treasuries, it is naturally unappealing for the protocol to hold perpetual futures positions.  Additionally, in phases of market downturn, funding rates are at higher risk of turning negative, which could put additional pressure on the protocol.

With that in mind, Ethena started to allocate a greater portion of the USDe backing assets to stablecoins, specifically USDC on Coinbase and Sky’s USDS. Each stablecoin offers a compelling yield, with USDC on Coinbase closely tracking US treasury yields, while Sky’s USDS offers a 6.5% yield at the time of writing.

Introducing stablecoins as backing assets for USDe is expected to ease pressure on the protocol during down markets, while maintaining the flexibility to transition to perpetuals when funding improves:

1. Bull markets: perpetual future funding likely improves and more of USDe’s backing will shift towards perpetual futures, capturing the basis and the upside that funding rates have to offer. Liquid stablecoin allocations would be lower in this scenario.

2. Bear markets: funding rates lower and US treasury yields outperform. More of the backing is allocated towards liquid stablecoins, easing pressure on the protocol while potentially earning close to U.S. treasury rates on a portion.

The recent reduction in USDe supply coincided with funding rates moving significantly lower than was observed during the first quarter of 2024. As depicted in the charts below, as the spread between sUSDe APY and U.S. treasury rates compressed, USDe supply started to contract.

Ethena’s Risk Committee

As of August 2024, Ethena launched protocol governance, taking the initial step of forming and tasking a Risk Committee with identifying, evaluating and guiding risk management within the Ethena ecosystem. The Risk Committee is made up of industry leading risk and advisory firms, already advising some of the top protocols in the space (such as Maker and Aave). For the first six months, the Risk Committee members are:

• Gauntlet​

• Block Analitica​

• Steakhouse

• Blockworks Research

• LlamaRisk

• Ethena Labs Research (non-voting member)

The Risk Committee is responsible for conducting extensive research and approving new backing assets for USDe, as well as monitoring the allocation of the backing assets to ensure they stay within acceptable parameters as pertains to risk matters like exchange exposure, asset exposure and counterparty risk. By way of example, the committee recently deliberated on and approved SOL as a backing asset for USDe.

Tasking the Risk Committee with managing risk provides the Ethena community with greater visibility into the decision making process, with members posting analysis in the governance forum for each matter.

UStb

Last month, Ethena announced the impending launch of a new product in collaboration with Securitize: UStb, a stablecoin product backed by Blackrock’s USD Institutional Digital Liquidity Fund token, BUIDL. Subject to governance approval, a portion of USDe’s stable backing could be allocated to UStb once live, allowing that portion of the backing to scale up and down as necessary during varying market conditions to enhance stability.

Before UStb is launched, Ethena has implemented the below dynamic allocation approach between funding rate and stable deployments. Those allocations combined account for roughly 25% of the backing assets of USDe today. This allocation % will change in line with the below chart, depending on the performance of perpetual future funding rates.

Improving Scalability

BTC and ETH presently comprise about 70% of the backing assets of USDe. BTC open interest is $26bn today excluding CME, while ETH open interest stands at $10bn. SOL has just been approved by governance as a backing asset for USDe, unlocking a further $2bn of open interest to access.

At an end state, USDe supply might be constrained in size to roughly 30-40% of that total open interest. At its peak of $3.5bn, the BTC backing of USDe represented about 5% of BTC open interest, and the ETH backing about 15% of ETH’s. At 30% of total BTC, ETH and SOL OI, USDe would have an estimated supply cap of ~$13bn based on today’s levels of open interest across relevant venues. However, that cap assumes no further growth in market size of perpetual futures - over the last year alone BTC open interest has grown 3x in size, from $8bn to $26bn today.

Once any supply cap in perpetual and deliverable futures markets is reached, Ethena has the ability to allocate towards UStb and/or liquid stables , which can match virtually any level of demand for growth.

USDe could theoretically add ~30bn in additional liquid stables backing if perpetual future supply is exhausted, and still maintain a spread of roughly +2% to RWA rates, assuming funding stays at baseline as Ethena builds a larger position.

Summary

Ethena adopts a dynamic allocation approach, in line with feedback from members of the Risk Committee, with the aim of making USDe more stable in varying market conditions while producing protocol-level revenue. Forms of staked ETH comprise significantly less of USDe’s backing today compared to when Ethena started as the protocol prioritizes assets that aim to improve liquidity, stability and scalability.