Ethena is the most successful DeFi protocol in history. It grew from less than $10M in TVL roughly a year ago to $5.5B. It is integrated in various ways into @aave, @SkyEcosystem IE Maker / Sparklend, @MorphoLabs, @pendle_fi and @eigenlayer. There are so many protocols that work with Ethena that I had to change the cover multiple times upon remembering another partnership. Six of the top ten protocols by TVL either work with Ethena or is Ethena (Ethena is #9). If Ethena were to fail, this would have dramatic implications for many of these protocols, especially AAVE, Morpho and Maker, which would functionally become insolvent although to different degrees. At the same time, it has dramatically increased the usage of DeFi in general by billions of dollars, akin to the impact stETH had on ETH DeFi. So, is Ethena doomed to destroy DeFi as we know it, or is it bringing DeFi into a new renaissance? Let’s dive into it

How does Ethena Actually Work?

Despite being out for over a year, there is still widespread misunderstanding about how Ethena works. Many people claim it is the new Luna and then refuse to elaborate more. As someone who warned about Luna, I find this argument very misinformed, but I also believe that not enough people actually understand the minutiae of how Ethena works. If you feel you FULLY understand how Ethena manages Delta Neutral positions, custody, and redemptions, then please skip this section, otherwise this is important reading to fully understand.

Broadly, Ethena benefits from financial speculation and the crypto bull market the same way BTC does, but in a far more stable way. As crypto prices increase, more and more traders want to go long BTC and ETH, and less and less traders want to be short. Because of supply and demand, traders who go short are paid by those who go long (more in depth explanation / example here). This means that traders can hold BTC, go short the same amount of BTC that they hold spot, and have a neutral position, IE if BTC goes up, the profit and loss of each position cancels out while the trader gets paid an interest rate. This is all that Ethena does; it takes advantage of the natural lack of sophisticated actors in crypto who want to earn yield instead of simply going long BTC / ETH.

However, a major risk with this strategy, as seen with the collapse of FTX and its impact on the first generation of delta neutral managers, was custody risk of exchanges. In the event an exchange failed, all the money could be lost. This is how major managers, who had managed capital efficiently and safely, were dramatically negatively impacted by the collapse of FTX, most notably @galoiscapital, through no fault of their own. Exchange risk was a major reason for why it was important for Ethena to use @CopperHQ and @CeffuGlobal, custody providers that function as trusted middlemen, holding assets and facilitating Ethena’s interaction with exchanges while not exposing Ethena to the custody risk of these exchanges. The exchanges in turn can rely on Copper and Ceffu as they have legal agreements with the custodians. Total Profit and Loss, IE how much Ethena owes to traders that are long or how much traders that are long owe Ethena, is settled periodically by Copper and Ceffu, and Ethena systematically rebalances its positions based on these rebalances.

Minting and Redeeming USDe / sUSDe is relatively simple. USDC / other major assets can be used to buy or mint USDe. USDe can then be staked to make sUSDe, which earns yield. sUSDe can then be sold into the market with corresponding swap fees or redeemed for USDe. The redemption normally takes seven days. USDe can then be swapped for backing assets 1:1 to $1. These backing assets come from a reserve of assets as well as the collateral used by Ethena (primarily BTC, ETH/ETH derivatives). Given that there is some USDe that is not staked (much of it going into Pendle or AAVE), the yield earned by the assets backing these unstaked USDe help bolster sUSDe yield.

So far, Ethena has been able to process large amounts of withdrawals and large amounts of deposits relatively easily, although at times it had slippage on USDe-USDC of up to .30%, which for stablecoins is relatively high, this is nowhere near a significant depeg, and is far from dangerous to lending protocols, so why are people so concerned?

Well, in the event of major need for withdrawals, such as say 50%

How can Ethena “Fail”?

Given that we now understand that Ethena’s yield is not “fake” and how it works at a more granular level, what are the major genuine concerns about Ethena? Well, there are largely (#) scenarios. First, funding rates could go negative, in which case Ethena would end up losing money rather than making it if its insurance fund, which is currently ~$50M, or roughly enough to take 1% slippage / funding hit at current TVL. This seems relatively unlikely, as most users would likely stop using USDe in the event of lower yields and which has been seen in the past.

Another risk is custodian risk, i.e. the risk that Copper or Ceffu tries to run with Ethena’s money. This is mitigated by the fact that the custodian does not completely control assets. The exchanges don’t have a signing permission or control of any of the wallets holding the underlying assets. Both Copper and Ceffu are “omnibus” wallets meaning all of the institutional users’ funds are commingled together across hot/warm/cold wallets with safeguards for governance (ie control) and insurance over various potential risks. Legally, this is structured as bankruptcy-remote trusts, so even if the custodian were to go under, assets held with the custodians are not a part of the custodian’s estate and the custodian has no claim over the assets. In practice, there are still simple malpractice and centralization risks, but there are indeed numerous safeguards to avoid this issue, and I would consider the likelihood of this happening equivalent to a black swan event.

The third and perhaps most talked about risk is that of liquidity. In order to manage redemptions Ethena has to sell its derivative position and its spot positions simultaneously. In the event of sudden and dramatic movements in the price of ETH / BTC, this could be a difficult / expensive and potentially very drawn out process. At the moment, Ethena has hundreds of millions ready to enable redemptions 1:1 for USDe to USD values as it holds significant stable positions. However, if Ethena is a larger and larger portion of total open interest (IE all derivatives outstanding), this risk is relatively serious, and could represent a multi percent drop in NAV for Ethena. However, in this event, it would likely be filled by the insurance pool, and in and of itself would not be enough to cause catastrophic failures for the protocols using it, which leads neatly into the next section.

What are the Risks of using Ethena as a Protocol?

Broadly, Ethena risks can be separated into two core risks: USDe liquidity and USDe Solvency. USDe liquidity is the actual cash available and willing to buy USDe for its $1 base value or up to 1% lower than that. USDe solvency is the idea that even if Ethena may not have the cash in its hand at a given moment (say after a period of prolonged withdrawals), it can get this cash if given enough time to liquidate assets. Take the example of if you lend $100,000 to your friend who has a house worth $1M. Sure, your friend might not have that money available, and he might not be able to get it tomorrow, but given enough time, he likely can get enough money to pay you back. In this case, your loan is healthy, your friend just isn’t liquid, IE his assets may take a long time to sell. Insolvency inherently means that liquidity should be non-existent, but limited liquidity does not mean that an asset is insolvent.

Some protocols that Ethena works with such as EtherFi and EigenLayer only face significant risks if Ethena is insolvent. Others, such as AAVE and Morpho, can face significant risks if Ethena’s products are illiquid for extended periods of time. On-chain, there is only around $70M of on-chain liquidity for USDe / sUSDe. Although if you use aggregators you get quoted that you can sell as much as $1B of USDe for USDC 1-1, this is likely due to massive demand for USDe that exists at the moment as it is intents-based demand, this liquidity, in the event of massive withdrawals from Ethena, will likely dry up. When this liquidity does dry up, the pressure lays with Ethena to manage redemptions to renew liquidity, but this may take time, time that AAVE and Morpho may not have.

To understand why this is the case, it is important to understand how AAVE and Morpho manage liquidations. Liquidations happen when debt positions on AAVE and Morpho are unhealthy, i.e. going past the required loan to value (ratio between loaned amount and collateral). Once this happens, the collateral is sold to cover the debt obligation, with a fee being charged and any remaining funds being returned to the user. In short, if the VALUE of the debt (principal + interest) compared to the value of the collateral approaches a predetermined ratio, a position will be liquidated. When this happens, the COLLATERAL will be sold / converted into the DEBT asset.

Currently, there are a large number of people using many of these lending protocols to deposit sUSDe as COLLATERAL to borrow USDC as DEBT. That means, in the event that there are liquidations, a large amount of sUSDe / USDe will be sold for USDC / USDT / DAI. If this happens all at once alongside other dramatic market movements, it is very likely that USDe will lose its peg to the dollar (if the scale of liquidations is very high of course, somewhere around the $1B mark). In this event, there could be a theoretical risk that there will be significant bad debt accrued, which in the case of Morpho is ok because vaults are used to segregate exposure, although certain earn vaults would be negatively affected. In the case of AAVE, the entire core pool would be negatively affected. However, in this potential scenario, which is purely an issue of liquidity, there could be a potential amendment to how liquidations are managed.

In the event a liquidation is going to incur bad debt, instead of selling the underlying asset immediately into an illiquid market and AAVE holders having to foot the difference, the AAVE DAO could take on the liability of the tokens and the position but not immediately sell the collateral. This would allow AAVE to wait until prices and liquidity from Ethena had stabilized, allowing AAVE to make more money from the liquidation process (instead of being net negative) and for users to receive funds back (instead of receiving nothing back as there would have been bad debt). Of course, this system would only work if USDe did return to its previous value, in the event it didn’t the situation with bad debt would be even worse. However, if there is a potential highly likely chance that the token will be worth zero that nobody has found yet, then it is highly unlikely that liquidations would be able to capture that much more value than waiting, there may be a 10-20% difference as individual holders realize and start to sell positions faster than parameters can be changed. This design choice is important for assets with potential liquidity issues during frothy markets, and could also have been a good design choice for stETH before withdrawals from Beacon Chain were enabled, and could be a tremendous way to increase the AAVE treasury / insurance system in the event it works successfully.

The risk of insolvency is relatively mitigated, but not zero. For example, say that one of the exchanges Ethena uses goes under. Sure, Ethena’s collateral is safe with its custodian, however it has suddenly lost its hedge, and now must attempt to hedge in what is likely a turbulent market. There is also the risk that a custodian may go under, as pointed out by @CryptoHayes when I had the pleasure to talk with him in Korea. Regardless of all the safeguards around the Custodian, there could be a serious hack or other issue, crypto is still crypto, there are still potential risks, even if they are incredibly unlikely and potentially covered by insurance, the risk is still non-zero.

What are the Risks to NOT Using Ethena?

Now that we have discussed the risks of Ethena, what are the risks for protocols that don’t use Ethena? Let’s look at some statistics. Half of Pendle’s TVL (at time of writing) is attributable to Ethena. For Sky / Maker, 20% of revenue is in some way attributable to Ethena. About 30% of Morpho TVL comes from Ethena. Ethena is now one of the largest drivers of revenue and new stables to AAVE. Most notable platforms that don’t use Ethena or interact with its products in some way have largely been left in the dust.

There are some interesting parallels to be drawn between Ethena adoption and Lido adoption among protocols. Around 2020 and 2021, the fight for the largest lending protocol was much more competitive. However, Compound was far more focused on minimizing risk, potentially to an absurd extreme. AAVE integrated stETH early on in March of 2022. Compound had begun discussions about adding stETH in 2021, but failed to do so until July of 2024 when a formal proposal was made. This was also around the same time AAVE began to overtake Compound. Although Compound is still relatively large with $2B in TVL, it is now just over one tenth of the size of AAVE, which it at one point dominated over.

To an extent, this can also be seen with the relative approaches of @MorphoLabs vs @AAVELabs to Ethena. Morpho began to integrate Ethena in March of 2024, whereas it took until November for AAVE to integrate sUSDe. This is a difference of 8 months, and during those 8 months Morpho grew significantly, and AAVE lost its relative grip over the lending space. Since AAVE’s integration of Ethena, TVL has increased by $8B and yields have significantly increased for users of the product. This has led to the “AAVETHENA” relationship, where Ethena’s products create higher yield which incentivizes more deposits which creates more demand for borrowing etcetera.

The “risk free” rate for Ethena, or at least its “normal” rate is approximately 10%. This is well over twice the value of the FFR, which currently stands at approximately 4.25%. Introducing Ethena into AAVE, especially sUSDe, functionally increases the equilibrium rate for borrows, as now the “base” rate of AAVE inherits the base rate of Ethena, even if not exactly 1-1, it will be somewhat close. This can again be seen before with how the introduction of stETH into AAVE made the ETH borrow rate roughly equivalent to the stETH earn rate.

In short, protocols that do not use Ethena risk having lower yields and thus lower demand in exchange for avoiding what likely amounts to minimal risk of a serious depeg or collapse in USDe’s price. Some systems, like Morpho’s, may be better adapted in that they can avoid potential collapses given their siloed nature, and for this reason it is understandable that larger pool based systems have taken longer to adopt Ethena. Now, while most of this has been looking back, I would like to posit something looking more into the future. Recently, Ethena has been moving to integrate DEXs. Most perps DEXs suffer from a lack of short-side demand, IE users who want to go short perps. In general, the only user type that can do this consistently at scale are delta neutral traders, of which Ethena is the largest. I believe that the perps platforms that can successfully integrate Ethena while maintaining a good product could pull away from their competition in a very similar way to how Morpho pulled away from its smaller competitors by working closely with Ethena.

Disclaimer:

1. This article is reprinted from [DC | In SF]. All copyrights belong to the original author [DC | In SF]. If there are objections to this reprint, please contact the Gate Learn team, and they will handle it promptly.
2. Liability Disclaimer: The views and opinions expressed in this article are solely those of the author and do not constitute any investment advice.
3. The Gate Learn team does translations of the article into other languages. Copying, distributing, or plagiarizing the translated articles is prohibited unless mentioned.