Deep dive into Ethena
Ethena has picked up some serious traction in its short existence with over $414 Million in market cap. Many users have been drawn in by the rather incredible stated 27% yields available on their stablecoin USDe, however some are looking cautiously, remembering the days they thought the 20% on UST from Terra Luna was a great allocation also.
As the stable-coin fund investment advisor, where we specifically focus on the available yields on dollar based assets, we thought it was a perfect time to get involved and share our thoughts. Is this the next Luna or just a great opportunity to ape on yields?
Our short summary:
Ethena is not going to be the next Terra Luna and UST.
The offered yields will decrease rapidly back to a normal rate.
Ethena contains a lot of risk which isn’t disclosed or fully understood.
There might also be additional legal risk in the US, especially when it comes to the guidelines proposed by the SEC.
Where the yield is coming from…
In very simple phrasing, they get it from two places, staked Ethereum yields, and then a ‘cash and carry trade’ on Ethereum futures. Their ETH is mostly allocated in stETH, this generates some 3.4% at the time of writing.
A ‘cash and carry’ trade is simply where a futures price on an asset in the market is above the spot price of that asset and so what you do is buy the spot (e.g 1 ETH) and sell one future on ETH. If there is a positive difference between their prices, you hold the ETH until the expiry date for delivery to the futures holder. This is a very common trading strategy in the world of crypto assets and is in principle ‘delta neutral’ the market price of ETH over the period will not affect the performance of the portfolio.
This is a form of arbitrage, and exists only in times of market inefficiency, however in crypto, it is a structural problem that a large number of private traders and larger funds engage In given the huge retail demand for futures and leverage in general.
To get more technical, we now have to look at the different types of futures used. There are traditional futures, and perpetual futures. The simple difference being the expiry date, traditional futures have a contract expiry, where the owner of the futures contract is entitled to delivery of the underlying asset.
In perpetual futures, there is no expiry date, and they instead simply use a ‘funding rate mechanism’ when demand for long futures is high and demand for short futures Is low, the funding rate is positive, this funding rate is paid from long futures buyers to short futures buyers periodically, to stabilise the price of the perpetual futures contracts to the underlying spot assets.
This yield can get pretty high depending on the market sentiment and activity. For example the Kraken ETH Perpetual futures funding rate for 8 hours is sitting at 0.035, extrapolating this out for a whole year, this equates to a nearly 32.2% yield. However, perpetual futures funding rates are volatile and subject to rapid change.
The critical take away is simply that, this APY offered, the vast majority of it comes from a market inefficiency that hundreds of private funds also exploit, as this market gains adoption, this spread regardless of underlying demand for futures will tighten in theory, and over a reasonable timeline, it should become effectively zero. At which point the yield will simply be the stETH yield less the cost of the futures contract, which in our opinion isn’t exactly appealing.
If you want to short term, generate some high yields, in our opinion it is not overwhelmingly negative, there are plenty of worse protocols out there, the risks we saw, we will outline nice and clearly below for you all, and you should do your own research as always.
How do they actually execute this?
Ethena has a company based in Rome (on checking the address online, it seems to be some apartment block), where they state that they have a Virtual asset service provider (VASP) approval from the Italian regulator.
They maintain accounts for crypto asset custody with Copper.co and other similar reputable crypto asset custodians.
A user buys USDe with a US dollar denominated stable-coin directly from Ethena or from another exchange like curve.fi. To get access to the yields underlying, they simply lock their USDe for sUSDe for a fixed period, with a minimum of 7 days.
Ethena in the background, use this USD stable-coin to partially acquire ETH, they then stake this on Lido mainly, to generate staking yields, they partially use the remaining balances to enter into short positions on futures both perpetuals and traditional futures on Binance, OKX, Deribit, Kraken amongst others.
All yields generated from the protocol currently go into the staking pool and are spread amongst all stakers. When a user unstakes to get their USDe back, the value difference in USDe returned and the USDe they initially contributed is their yield.
The process described above, if told to a regulator is our main problem:
It is an investment of funds (buying USDe and locking it for sUSDe), in a common enterprise (it’s a pool operated, owned and advised by an entity based in Rome), with the expectation of profit (it publicly states the yields available to users), and these users rely on the efforts of the Ethena team and its developed technology to generate profits in trading.
It fails every single part of the ‘Howey test’ immediately.
To make it worse. It looks strongly resembles a collective investment scheme, with derivatives based on crypto assets, sold to retail users. If an American has invested in this, and the SEC notices it, in our opinion, they will very quickly fined.
Situations where it would ‘break’:
Backwardation markets
Backwardation is the name for markets where there is a sustained discount on futures prices compared to the underlying spot market. Ethena is set up and designed only for positive differences between the futures price and spot prices, it takes in collateral in ETH or US dollar denominated stable-coins, it stakes the ETH as stETH to generate a small yield, and through this holding is ‘long spot’ it uses its stable-coin balances, to enter into short futures positions. When combined with the yield effective on ETH as stETH, this carry trade is positive on 89% of days according to their own research.
If a new bear market sets in, perpetual futures begin to see negative funding rates, and ordinary futures trade long term at a discount to spot, and this is sustained over a medium term, the ‘yields effective’ on their current strategy would become negative.
While their ‘advanced trading algorithm’ is executing these strategies for them, as with all automated strategies, they often struggle with edge cases like these, some further understanding on the mechanisms of the algorithm would help here.
In the best case, if it needs to maintain a perfectly dollar hedged portfolio, it would continue its current strategy while the lido staking yield is greater than the cost/losses on its short futures position. This would mean it would generate less than the lido staking yield, and then for its operations, including the cost of custody services and gas fees it would be making a very small effective yield, which would make it relatively unappealing to most users.
Counterparty failures
If one of the exchanges where they operate their futures has any issues of any kind, they will be at severe risk, any issues with their custody arrangements would also create severe risk for token holders. As the issuer and owner of the token, the received assets are held in their own custody on their balance sheet we would assume, so any issues regarding the solvency of Ethena labs would also cause potential issues for token holders.
Risks and fair disclaimers
Custody Risk
The protocol is partially decentralised partially centralised, it uses a number of large reputable service providers such as Copper, Fireblocks etc, to manage its assets under custody. However, if there were any issues with these custodians in regards to security breaches or similar, it would severely affect their business.
This is the case for nearly all relevant competitors who maintain fully physical/fiat collateralised centralised stable-coins, E.G. Circles USDC or USDT Tether, so is not unique to Ethena.
Some of its collateral however is also custodied as deposits in stETH, this has a slightly higher custody risk relatively compared to USDC in particular as a decentralised protocol, it has exposure to greater security risks contract risks and execution risks.
Regulatory risk
As previously described, Ethena receives customer deposits, issues ‘depository reciepts’ to its users (USDe), promises effective returns, and generates them from a very common trading strategy of the carry trade, based on underlying derivatives. (both stETH and the futures positions are de-facto derivatives.)
This to any regulator at best, could be seen as a collective investment scheme, with the majority of its user base being retail, not qualified investors so it has no Reg.d ‘work around’ in SEC terms, it would require full licences as the operator and distributor of a collective investment scheme in all jurisdictions where it operates in our opinion. In the worst case, this could be seen as a bank.
As it does have a ‘body corporate’ as Ethena Italia Srl based in Rome, and operates some elements in a centralised way, it falls under Italian and EU financial regulation.
Financial risk
Ethena is heavily weighted in derivatives on Ethereum (stETH) which while it has mechanisms to ensure its liquidity, under turbulent market conditions, it is not uncommon to see de-pegs or spreads widen between these liquid staking tokens and their underlying assets short term which could severely affect the performance of the portfolio if it tries to realise its returns.
The remainder of their assets is held in short ETH futures collateralised with USDC across a number of venues, as well as a minor set of positions in short perpetual futures.
While this is a common trading strategy and is in principle ‘delta neutral’, market fluctuations of funding rates on the perpetual futures, market liquidity to offload such large positions, futures volatility, de-peg risk in its collateral assets (stETH or its US-dollar stablecoins) can lead to potential severe portfolio losses.
Gas kills fees, could kill more.
Users contributing into the pool with the expectation of these high yields, must be aware, that these fees are gross fees, less any slippage and gas fees for execution of transactions.
For example if the yield stated was 30%, gas fees for simplicity at $10. User swaps USDC to USDe, they pay a gas fee and take some slippage of 0.1% on entering the positon, to get the yield, they then must stake their USDe for sUSDe, paying again $10 in gas fees.
Total cost $20 + 0.1% of volume.
After one year they claim their yield ($10 gas), withdraw their USDe from staking ($10 gas), swap it back to USDC (0.1% slippage +gas)
Total Cost $30 + 0.1% of volume.
So the yield stated is inaccurate to the true performance a user will see, below based on these over simplified numbers would be the ‘true yield’ under the highly unlikely condition that the yield stays at 27% over the period.
$100 Contributed = $76.746 net of fees (-23.25%)
$1000 Contributed = $217.46 net of fees (21.46%)
$10000 Contributed = $2624.61 (26.24%)
This is generally the case in a number of ETH based protocols, and is more a complaint about Ethereum and the poor UI/UX and clarity for end users of DeFi in general.
However, this may also have implications for the entity, currently Ethena doesn’t implement a fee system. Imagine a user interacts with the contract with 100 USDC to mint 100 USDe, the user pays a gas fee of say $10, they then stake this into sUSDe, and pay again a small gas fee, again for simplicity $10.
The entity itself then, is receiving the assets, transferring them to their custody solutions or exchanges (again costing gas fees) executing the trades (trading fees taken), while also acquiring ETH (swap and gas fees), then contributing the eth to stETH.
As a company, it seems like an expensive set of operations where they would lose some value in its operation, hypothetically without some fee system, a mad man with unlimited money just minting and claiming over and over on repeat could foreseeably cost the entity quite a bit in gas fees, it would be good to see if they have systems designed to prevent this.
MEV attacks
We know the on-chain transactions that they must implement, and under what conditions they will need to enter them, some actors in the space if so inclined could scalp some MEV transactions on their positions entering into ETH and into stETH with relative ease, depending on the volumes going through, some further research would be needed on the direct affects of this on the performance of Ethena, but on first look, more complex and smaller volumes have created some strong returns for MEV traders.
Front running
In regards to their futures positions, they publicly list their positions, their exchanges and their exposures, if linked with the on chain orders, it would be feasible, upon seeing a larger order for USDe and sUSDe, that you could ‘front run’ their trades, as they will have the lag of two block confirmations before they execute any futures transactions.
The nightmare scenario
MEV traders start to scalp out some fees on the ETH / stETH allocaitons, market makers and futures traders start to front run their futures positions, the effective yields of the USDe compress heavily below 10%, users switch to other stable-assets where it is possible to generate returns in excess of this with a relatively lower risk.
The reserves of the company are slowly depleting on development expenses, gas transactions, custody fees, trading fees, until they implement a trading fee into the system.
The SEC see the operations of the protocol, have some questions that seem similar to the ones we outlined above, and they begin an investigation, in partnership with local European regulators. After a short investigation if they came to a conclusion that Ethena represents a collective investment scheme, that has been sold to retail traders, they start a lawsuit.
Ethena has raised some $14 Million USD so far, US lawsuits about SEC and regulatory compliance would drain all the way through this relatively quickly, that would be before any potential fines.
Ethena labs is not a bank, users have no deposit protections, they are on their balance sheet, the assets in its custody in any insolvency event, would be included in the hypothetical liquidation. Ethena would be forced to unwind its entire futures position to meet creditor demands, potentially at a loss.
It would also need to unwind its stETH and ETH spot positions. Unwinding any significant value of either in a short time period may also incur trading losses.
Under this scenario, the entity would be entirely insolvent, there would be incurred losses, token holders could not be guaranteed any value back from their allocations. With the above known, any announcement by a regulatory agency into Ethenas USDe, would cause a pretty rapid ‘dump’ of USDe.
Our general summary
The yield will compress over time.
This yield is legitimate but comes with a large number of associated risks that some users may not be aware of.
It has a few weaknesses in their strategy from what we can see publicly.
We are very concerned about the possibility of regulatory actions against the protocol and the following affects on token holders.
We would generally avoid this until we see some regulatory clarity, there are plenty of other methods using US dollar stable-coins to generate solid yield, with lower relative risk.
This report may contain material that is not directed to, or intended for distribution to or use by, any person or entity who is a citizen or resident of or located in any locality, state, country or other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation or which would subject 512m AG or its affiliates to any registration or licensing requirement within such jurisdiction. All material presented in this report, unless specifically indicated otherwise, is under copyright to 512m. None of the material, nor its content, nor any copy of it, may be altered in any way, transmitted to, copied or distributed to any other party, without the prior express written permission of 512m. The information, tools and material presented in this report are provided to you for information purposes only and are not to be used or considered as an offer or the solicitation of an offer to sell or to buy or subscribe for securities or other financial instruments.
