[CLOSED] ETH Vault Strategy

General Protocol
1

[Request For Comment] VST ETH Vault stETH Strategy

Authors: The Vesta Core Team + Review from Frens.

Note we recognize this is a period of immense market volatility and would look to implement this strategy after this calms down and the peg is held for longer.

Summary

With close to $10M worth of Ether in the ETH system on Vesta, our collateral reserves represent a huge pool of untapped potential. Currently, it sits idle in the pool, doing nothing of worth. While we stand by the service we provide by giving you on-chain utility with vaults, we believe that it is still possible to make your locked assets work for you even when they’re in a Vault.

This RFC suggests the implementation of depositing ETH collateral into a stETH strategy from the ETH vault and requests comment from the community on the functionality, design, and timeline. Currently there is ~3689 ETH sitting idle inside of the ETH Vault. While the ETH acts as collateral for the VST debt, the ETH is quite passive and not, as we believe, as capital efficient as it could be. However, sourcing sustainable ETH yield strategies, especially in this environment of yield compression is difficult.

While we are proposing what we believe is a close-to-optimal strategy for ETH utilization, it is important to note that you as the users have ultimate control over your own collateral. To that end, we encourage you to post your own thoughts on the risks, rewards, and alternative approaches. Community interaction is key in a large operation such as mobilizing $10M worth of collateral.

Currently, we believe that we can strike a desirable balance between security and yield with the Aave v2 (then eventually v3 which will allow us to squeeze even more ETH yield in high efficiency mode with high LTVs) stETH-Lido strategy. This is a widely used and trusted strategy, which would give the platform a substantial amount of extra revenue. This is a rather simple strategy to execute, which just involves depositing the collateral ETH into Lido.

Core risk is stETH-ETH depeg risk which leads us to being liquidated.

Current yield is 4.5% then 9% post merge which can be boosted by recursive borrowing in Aave. A key question for our community to decide is on the fee share between the earnings to the users in the vault and the governance token.

We could also offer two vaults, one of the vaults with a yield bearing stETH:ETH strategy and that doesn’t. This way the risk is separated and only placed upon users who are aware of it and want to take advantage of it. In the non-yield bearing vault we would change the lowest collateral ratio to 110% and in the yield bearing vault we would have 150% collateral ratio.

On the non-yield bearing vault, we have a 0 mint fee. The strategy would occupy a range of between 30-60% of the eth vst vault.

Finally, it is essential to note that stETH is redeemable for 1 ETH once the merge occurs. A stETH:ETH depeg may cause those with leverage to be liquidated in the short term, but as the merge date gets close any arbitrage opportunity becomes more valuable for those holding ETH.

Motivation

We aim to be the first L2 lending protocol that takes advantage of idle collateral to generate fees back to token holders. By taking advantage of relatively low risk strategies to mobilize collateral we increase capital efficiency of idle ETH.

Implementation/Execution

The ETH on Arbitrum would be bridged over through Synapse (incurring a bridging fee) then deposited into the generalized strategy through a fee different approaches:

  • Instadapp steth vault (note that this vault has limited capacity on the withdraw and deposit side)
  • Run the AAVE V2 strategy ourselves

Notes on liquidation risk and recent stETH depegs:

Due to recent market volatility, there have been some reasonable concerns regarding liquidation risks on strategies similar to this one. At Vesta, we are committed to providing you with the most upside on your liquidity while maintaining the least risk possible. To this end, we are committing to the following:

  • Voluntary risk-taking by implementing both yielding and non-yielding ETH vaults
  • Responsible recursive borrowing. In the context of recent volatility, we will recur in small and justifiable amounts with frequent communication.
  • Responsible timing. As of right now, stETH-ETH peg appears to remain off by around 400 bps. This has held for a few weeks now, and nothing indicates that this will widen in the future. In fact, we and many observers believe it is more likely that the peg will improve over the future weeks and months. Thus, we commit to timing any moves responsibly and safely.
  • Alternative strategy proposals. Part of the timeline of collateral mobilization includes giving more choices to the community. As part of this, we are committing to outlining multiple strategies when risk is involved such as with this ETH strategy. In the near future, we will be outlining alternative ETH staking strategies involving modules such as frxETH or squETH. Stay tuned!

Timeline

The implementation itself wouldn’t take that much time and be executed in weeks.

Community Sentiment Poll

After sufficient community discussion on this RFC, the authors of this proposal will open a poll and decide whether to move forward with execution.

5 Likes
2

One nitpick: stETH won’t be redeemable after the Merge; the next hard fork (Shanghai) is what enables staking withdrawals and will allow stETH redemptions en masse. This means another ~6 months or more where stETH could “depeg”, though I am of the opinion that a successful merge will remove a lot of execution risk and the stETH-ETH gap will close as time goes on and Shanghai approaches.

As for the strategy, I would definitely prefer either using a separate vault entirely or have this kind of yield be opt-in. People are already incurring risk by opening a CDP on a volatile asset; it should be up to them whether they incur more risk by having the vault leverage their collateral using multiple external protocols (Synapse, Aave, Lido, Instadapp, possibly others).

Would it be worth considering a liquid-staking-derivative basket instead of going all-in on leveraged stETH? It would lower the APR on the vault’s yield but insulate it from potential risk, specifically pool slashing. A non-leveraged strategy would also be more sustainable long-term, because eventually the ETH borrow rate on Aave will catch up to native staking yields (especially after Shanghai). Vesta could use ETH collateral to purchase stETH, rETH, Stakewise ETH, BETH, Kraken’s derivative, and Coinbase’s derivative whenever that comes out. Across the board this would result in roughly 75-80% of the native ETH staking rate, about 3.15% at today’s APR. Perhaps that could be yet another vault option; more options are nice but I don’t want to fragment liquidity overly much.

Fundamentally I do like the idea of using vault assets in a reasonably-leveraged staking yield strategy. As for the vault strategy fee, I think it should be kept fairly low given the amount of risk the user incurs and the fact that there are other assets on the market replicating this strategy. Index Coop’s icETH takes a .75% streaming fee while GalleonDAO’s ETHMAXY takes 1.95%; it would probably be reasonable to ask for 2-5% given that Vesta lets you borrow against your collateral while it earns yield.

Regarding timing, there are two significant events on the horizon: Arbitrum Nitro and the Ethereum Merge. I think it would be prudent to wait for both of these to pass before deploying this kind of strategy, though it may take that long to approve and create this vault anyway.

5 Likes
3

Agreeing with the sentiment expressed above, we should have separate vaults for stETH(or a mix of staking derivatives) and ETH. Even though the de-peg risk is decreasing as we move closer to the Merge, I would argue that not everyone is inclined to take it on.

Additionally, in theory, once stETH strategy becomes available on AAVE v3 it will also push the yields on pure ETH deposits. This with the increased staking yield from Merge might prove to be already attractive enough.

2 Likes
4

I am all for making the ETH productive and I like VST, but I have a problem with Lido. Lido is a victim of its own success and in my opinion has too big of a stake at the moment in the Ethereum validator network. I’ve always viewed keeping the Ethereum network sufficiently decentralized as one of the most important things we can do.

I’m probably in the minority, but I personally avoid stETH or anything that might create demand for it. I would likely avoid VST if any of the collateral were used in a way that would create demand for anything that threatens the decentralization of the Ethereum network.

I’m all for researching ETH staking strategies that don’t involve actors who control >15% of the validators.

3 Likes
5

This is certainly worth taking into consideration, and it’s partially why I’m interested in a more diversified approach to staking yield. My ideal ETH vault strategy would probably just be buying rETH, SETH2, stETH, and perhaps a few other derivatives to spread out the risk and avoid impacting the underlying staking network.

1 Like
6

I would think rETH is a better alternative than stETH for decentralisation, would be cool to have an index of ETH staking derivatives for better diversification exposure.

2 Likes
7

Noted, thanks for this very comprehensive writeup. Yeah, we’re seeing a lot of love for rETH in this RFC currently. We’ll do some more research and edit the governance post accordingly! Looks like stETH has sorta gone the way of the dinosaur. We’ll check out the pros/cons and come back.

We’re also seeing calls for user optionality on the ETH vault, as in being able to choose whether to have their ETH be productive or not. Part of our considerations here would be accessibility of collateral from Arbitrum as well, especially if there’s a lot of user optionality here which may impact the size we can actually deploy and migration concerns. We’ll do some further work here.

2 Likes
8

Hello,

Interesting proposal, I have a few comments to make:

  • I agree that two vaults for ETH should be offered, one for regular ETH and the other for stETH. People who desire to swap can do so, others can just leave their ETH in the current vault.

  • Regarding the 150% collateral ratio for the stETH vault, I think that it’s too high. Even if peg decreases by 5-10%, you’re not getting the same level of capital efficiency than with the ETH idle vault. Obviously, it’s risky to go to 110-120% collateral, but something like 120-130% would allow people to get to 150% and feel fairly safe. A quick glance over the risky vaults for ETH shows the lowest at 124%, while the second is at 133% collateral ratio. People like a buffer and with a 150% collateral ratio, we’re likely looking into 160%+, which would drastically decrease capital efficiency, even with the stETH yields! Obviously this strategy can bring additional fees for VSTA, and making sure it attracts a high TVL is key, in my opinion. 150% collateral ratio seems too high. You mention that it is a low risk strategy, hence my motivation to question why 150% collateral ratio is necessary.

  • Discussion whether having stETH or other liquid staking options is good. I tend to agree that supporting smaller liquid staking options can also be positive for the ecosystem (the main one mentioned above is rETH and could indeed be a good option.

  • Nonetheless, stETH is supposed to go live on Arbitrum and Optimism (Lido's stETH Comes To L2), so stETH vault could likely bring the most options currently!

  • Finally, what about the current stETH-ETH pool on Curve/Convex, which is yielding above 8%? VSTA could take fees directly in CRV & CVX, which would help it accrue a warchest for voting purposes? This could be beneficial long term for flywheel effects: fee rewards are accumulated and can be used to vote for VST pools.

  • What kind of yield does the Instadapp/Aave strat yield? As mentioned just above, if we assume that stETH will go back to peg, why not convert 50% of the ETH in stETH, for a discount, and farm it on CRV/CVX?

Overall, I think that this proposal is very interesting, but we should figure out the best way to put it to work, to make sure that it manages to attract a lot of TVL. Being able to borrow VST on “productive assets” can really help the protocol grow, both in terms of TVL, but also fees generated. It’s key to find a good balance between protocol revenue and attractivity for users (fees shouldn’t eat up most of the rewards or else the extra risks [mostly peg & smart contract] are not worth the rewards).

Also, maybe this is not the best place to ask, but have you considered the DPX vault? plsDPX has recently gained a lot of traction, and could be an other area to explore! In my opinion, making collateralized assets “work” can open up many options for Vesta, and further increase its fee revenues!

Thank you for your time and the work you guys are doing!

CD

3 Likes
9

Closing this topic for now as the strategy is outdated.