Ethereum

Following recent developments in the ETH staking ecosystem, including:

- Ethereum (minority) client bug
- Protocols using primary LST price (1:1) instead of market price
- Introduction of @eigenlayer and restaking

Decided to write a revised leveraged ETH staking risk analysis.

DeFi protocols research analyst & product @defisaver

The strategy for boosting the ETH staking yield via lending protocols has become quite popular since the last write-up, so I'll give just a short overview of it here.

Increased yield comes from leveraging the spread between ETH borrow rates and ETH staking APY. 👇

Before delving into the risks, it's worth mentioning that the profitability of the leveraged staking strategy has increased in the last couple of months, with the average APY being over 10% despite the recent LSTs' drop in average yield (from ~5% to 3.5%).

The sustainability of the leveraged staking strategy has been questioned many times ever since the first additions of LSTs as collateral.

One of the main concerns was that the spread between borrow rates and staking yield was going to be arbed away.

https://dune.com/queries/2235238/3665104

As it turned out, the spread has not been squeezed at all.

With the max leverage for LST/ETH positions being increased across all money markets (up to 18x on Blue), the required spread got down to a minimum.

Even 0.2% is now enough to perform better than simply holding an LST.

The conclusion is that there are just enough people wanting to earn mild APY (<3%) on their vanilla ETH, alongside people longing ETH or just using it as collateral to short other assets (not interested in the supply rate), to make this strategy sustainable (long-term).

Alongside new underlying protocols and LSTs enabling this kind of strategy, the associated risks have also evolved in the meantime.

Some are fully mitigated, but there are new ones, too.
Let's explore those below.

Revised risk analysis of leveraged ETH staking strategy.

1. Borrow rate spikes.
2. LST provider (big) slashing event.
3. LST depeg & liquidation hunting.
4. SC risks.
5. Governance risks.
6. Restaking.
7. Ethereum execution client bug.

Breakdown of each point.👇

Borrow rate spikes.

Users can end up having negative net APY in case of high ETH market utilization on the underlying lending protocol (i.e. borrow rates > staking yield).

However, this needs to be severe or very long-lasting to have an impact (which depends on the leverage).

Protocols like Compound v3 and Morpho Blue (can) avoid having huge rate spikes after hitting the optimal utilization since no collateral is lent out guaranteeing liquidity for potential liquidations.

Those are thus deemed to be less risky for borrowers in this regard.

LST depeg.

LST market price represented a significant risk factor because of their use as collateral especially after the stETH/ETH Curve pool 'depeg' in 2022, *but it's not anymore*:

1. Can redeem 1:1 in a few days (withdrawals enabled).
2. Protocols now using primary rate. 👇

Shanghai upgrade (enabling withdrawals) resulted in the 'peg' tightening since stakers can now unstake at a 1:1 ratio anytime by waiting a few days in the exit queue.

However, the market price is still trading at a small premium since it provides an instant exit.

While withdrawals have improved the peg, this is not of much help (apart from liq. risk) when it comes to unwinding the leveraged LST/ETH positions since it needs to be atomic (in 1 tx).

It can be a factor if one has enough ETH for the whole debt repayment (withdraw + unstake).

Currently, the only unwinding option is to sell an LST back to ETH (can't wait 3 days), which implies potential swap fees and trade size impact.

Partial unwinds combined with waiting for pools to rebalance can be a solution for some, of course.

Therefore, on-chain LST liquidity is definitely a significant risk measure one wants to keep an eye on when deciding to ape into the lev-staking strategy.

Not sure if technically feasible, but it'd be cool to have a protocol matching the entry queue with the unwinding liquidity.

Lending protocols switching to LST primary rates.❗

Another (arguably) positive impact on the liquidation risk for lev-staking positions is the switch from price feed using the market price of LSTs to the primary LST rates to assume a 1:1 LST/ETH peg.

Find the full risk analysis and comparison of using primary vs secondary LST rates by BA below.

TLDR: Using primary rates shifts the risk from market manipulation (e.g. liquidation hunting) to the health of Ethereum infra (less volatile).

Find the full risk analysis and comparison of using primary vs secondary LST rates by BA below.

TLDR: Using primary rates shifts the risk from market manipulation (e.g. liquidation hunting) to the health of Ethereum infra (less volatile).

https://blockanalitica.substack.com/p/analysis-of-market-price-vs-exchange

In case of a big slashing event, the primary rate *could* also drop, but:

1. Significantly less volatile than the market rate.
2. Would not cause cascading liquidations.
3. DAO-governed protocols can react accordingly.

The primary rates are currently used by:

- Aave v3 (on both L1 and L2s)
- Spark (wstETH)
- Compound v3 is also considering this
- Morpho Blue is permissionless so technically anyone can create an LST/ETH market with a primary exchange rate as an oracle.

SC risks.

This one is probably what we all are most familiar with.
*Leveraged* ETH staking smart contract risks include:

- Liquid staking protocols
- Underlying lending protocols
- Tools for opening/managing such positions (e.g. http://ethsaver.com).

Governance risks.

- Aave v3, Compound v3, Spark: DAO-governed.
- Morpho Blue: Immutable.

DAOs impose a risk of a malicious proposal but also have the ability to react to market changes (e.g. disable additional ETH borrowing if primary rate drops).

Restaking.

Restaking is essentially leveraging the existing validators set to secure multiple networks (even non-EVM). Thus, staking yield can be increased, but also the risk of being slashed (according to the new slashing rules defined by new networks).

Here, we will cover the risks imposed by restaking for:

- Ethereum security in general
- Leveraged ETH stakers
- and also the potential LRT/ETH strategy.

When it comes to Ethereum security, @vitalik.eth himself outlined his thoughts about the risk of restaking.

TLDR: Makes sense if used for low-risk purposes. Avoid building complex financial primitives with the help of re-staking.

When it comes to Ethereum security, @vitalik.eth himself outlined his thoughts about the risk of restaking.

TLDR: Makes sense if used for low-risk purposes. Avoid building complex financial primitives with the help of re-staking.

https://vitalik.eth.limo/general/2023/05/21/dont_overload.html

As for leveraged staking strategy, restaking introduces a higher possibility of slashing. This extends to the % that can be slashed.❗

Under the current design of Ethereum, up to 50% of staked ETH can be slashed.

Eigen enables slashing the remaining 50% staked on the protocol.

Liquid restaking strategy.

Since Eigen Layer doesn't have its liquid restaking token, there have been a lot of projects building on top of EL introducing their LRTs.

If those LRTs are ever to be added as collateral on lending protocol(s), we'd have ourselves a new strategy.

Risks of potential liquid restaking strategy.

Clearly, potential leveraged LRT/ETH positions would be exposed to an additional 'Layer' of risk, alongside probably more complex process of leveraging up.

Will address those in detail if/when they become available ecosystem-wide.

Ethereum execution client bug.

TLDR:
- Minority client bug:
Penalties at the same rate as earning rewards (0.4% loss of the stake in 40 offline days).

- Majority client bug:
Increased penalties during *Inactivity leak* (90% loss of the stake in 40 offline days (no slashing).👇

This design actually incentivizes switching to a minority client by limiting the potential losses from 100% to 3.5% since this doesn't cause the chain to stop finalizing.

Team at Labrys wrote the best summary I've stumbled upon so far.

This design actually incentivizes switching to a minority client by limiting the potential losses from 100% to 3.5% since this doesn't cause the chain to stop finalizing.

Team at Labrys wrote the best summary I've stumbled upon so far.
https://twitter.com/Labrys_io/status/1749746305738703217

It's a more difficult decision for those whose business model depends on uptime.

After @dcinvestor's initiative to get their attention regarding client diversity, Coinbase replied they're reevaluating alternative execution layer clients, aiming to add them to their infra.

The risk of the majority client bug applies to all ETH stakers, including the leveraged ones, meaning we all can contribute by switching to a minority client.

Quick overview of the current landscape (geth):
- Coinbase, Binance: 100%
- Lido: 76%
- RocketPool, StakeWise: 42%.

To conclude, the risk/reward ratio so far has shown that a significant percentage of people are OK with the risks of (leveraged) staking with the APY offered in return.

As staking became more attractive after The Merge, so did the leveraged staking.

The top 3 liquid staking providers have seen a 2.5x average growth since September 2022 (The Merge).

Lido Finance stETH: 212% (2.1x) - growth of ~5m ETH.
Rocket Pool rETH: 355% (3.5x) - growth of ~900k ETH.
Coinbase cbETH: 202% (2x) - growth of ~720k ETH.

Management: Pooled vs Isolated approach.

If you wish to join the lev-staking bandwagon, there are two types of management tools for such positions:

- Pooled: Instadapp Lite, Sommelier Finance, etc.
- Isolated: http://ethsaver.com by @defisaver.

A short overview below.👇

Pool-based tools offer a 1-click option for depositing into a vault that's managed by a strategist.

Vaults usually have withdrawal fees (and limits!) to prevent spam, and likely a bit lower APY because of the idle liquidity and the potential unwinding losses.

Per-user approach (no asset pooling) introduces full control and no withdrawal limitations nor fees while having multiple choices (protocols and LSTs) available with 1-click switching to another one.

It also allows custom leverage management (in a fully non-custodial way).

To provide access to leveraged ETH staking strategies, we created:

- http://ethsaver.com - a non-custodial app with 1-click management options focused solely on leveraged staking.

Random stats:
• Launched in Sep '23
• 7300+ total ETH supplied
• 80% of Morpho Blue.

Summary (part 1):

- The sustainability of the lev-staking strategy is less questioned.
- Min spread of 0.2% to be more profitable than holding LSTs.
- Liquidation risk has become significantly lower (primary rates used).
- Consider partial unwinding & mevblocker bc of TSI & MEV.

Summary (part 2):

- Restaking increases slashing chances & potential losses (from 50% to 100%).
- Current Ethereum design incentivizes switching to minority clients by limiting losses from 100% to 3.5%.
- Isolated lev-staking management offers greater flexibility & control.

Agenda:

- The strategy and historical performance.
- Sustainability analysis: squeezing the spread (min spread at 0.2%).
- Risk analysis.
- Liquid restaking strategy.
- Minority vs majority client bug risks.
- Managing lev-staking positions: http://ethsaver.com.
- Summary.