Investment Strategy Framework - ETH2 Staking

Hi all,

Following on from this thread, I wanted to restart the discussion about generating yield on the Mutual’s asset pool. As Hugh mentioned, this is an area where community input is very valuable to both decide a framework and subsequently pick the individual options.

A few points to start off with:

  • The investment asset universe for Nexus remains limited to ERC20s on Uniswap due to the current technical implementation. As I understand, there are other priorities for protocol improvement before specific smart contract calls are developed for investment purposes.
  • There were a number of proposals for ERC20 assets in the initial thread linked above. However, these typically come with compounded technical risk (on top of that already covered by the mutual) and/or require trust in an asset peg or a centralised entity. These can be managed with exposure limits but the rewards for doing so don’t seem overly appealing at the moment.
  • The Mutual is in a position now where the liability exposure to Smart Contract Covers is significantly below the assets in the capital pool. While this is the case, exposure to the entire Ethereum ecosystem in the form of holding ETH seems preferable to picking out individual ERC20s for outperformance vs. ETH or hunting DAI/USD yield.
  • ETH2 Phase 0 has gone live with something like 1.6m-1.7m ETH deposited at the time of writing.

Therefore, believe it’s worth having a discussion about whether it’s worth deploying the Mutual’s assets to earn ETH2 staking rewards. Two aspects that we should establish:

  • A framework for evaluating any staking service or other route to obtaining the yield.
  • What percentage of the assets (if any!) and when should we be allocating to this. Assume that we would start out slowly and increase allocation over time.

Some possible routes are listed below.

Tokenised ETH2:

Buying tokenised ETH2-yielding ERC20s is the option that is currently the easiest to implement, but the landscape of options is sparse. Available volume and liquidity is not where it should be for meaningful amounts in the context of the Mutual’s assets.
However, this seems like a good medium-term option if the community believes we can do this within acceptable constraints. Hugh and team have good connections within a number of these projects so if the community feels it makes sense to explore further, there is scope to do so.

Some examples:
(1) https://lido.fi/
(2) https://stkr.io/
(3) https://stakewise.io/
(4) https://beta.rocketpool.net/ (when live)

A number of the centralised exchanges (Binance, Kraken, etc.) also provide staking opportunities but doesn’t seem worth considering in too much detail given the lack of transparency and risks.

Staking Services:

The system isn’t currently set up to do this, but this is worth discussing for the longer term. https://beaconcha.in/stakingServices seems to have a pretty comprehensive list outlining all the options for staking services.

One prominent option that popped out: https://www.attestant.io/
Probably more targeted towards wealthy non-technical users than projects wishing to deploy assets, but a decent example of an institutional-grade staking service.

Interesting that there is https://ethddc.org/#, whose first task is to evaluate the staking pools. Something to look at once they publish results if we do want to go down this path.

Framework Proposal:

Below are some considerations for the framework that could be used to evaluate any available options. Feedback welcome.

  • Yield:
    How much do we earn from putting the pool into one of these services? What are the fees they charge?

  • Governance:
    Evaluation of the governance framework of the asset/staking provider. Encompasses both centralisation/trusted counterparty issues and the governance framework of any decentralised solution. Can Nexus Mutual participate in the governance? Do we want to participate in any governance mechanisms and, if so, to what extent?

  • Technical/Smart Contract Risks:
    For any possible solution, need to assess how likely it is that something doesn’t function as intended and we lose a significant proportion of the assets.

  • Liquidity:
    What’s the possible amount we can actually acquire or provide to the staking service? How comfortable are we being a huge chunk of the assets held by each individual service? How big does each service have to be for us to consider it?

  • Infrastructure of Underlying Service:
    What are the risks that would lead to loss of funds due to slashing?

  • Decision point:
    How many of the above points and to what extent do we need to be comfortable with to push the trigger. How much due diligence do we want to do on each individual project?

Community Questions:

Keen to hear thoughts from the community on the following topics.

  • Should we do this at all?
  • Are we happy with the current state of the options available? Any other suggestions?
  • How much capital should we deploy and when?
  • Any feedback on the framework above, including specifics of due diligence you’d like to see performed.
  • Any other points?
4 Likes

I think Eth2 staking should be done with a trusted centralized exchange like Coinbase once it goes live in Q1. The reason being using a decentralized staking service exposes the investment assets to smart contract risk especially for a new project, which is the type of risk that the mutual should be selling covers for instead.

3 Likes

I second that re: tokenized eth2 projects–and the non-transferability of assets locked in eth2 staking makes other avenues (such as owned validators with Attestant, Bison Trails and so on) clunky if the Mutual is not running on fractional reserves (except if this is an opt-in feature for members?!).

That could potentially change if there was a standardized solution to collateralize validators (batches of 32 ETH) instead of BETH, but that’s not currently available.

On another note, enabling the pool as a feeder for flash loans requires smart contract calls enabled, right? cc: @Hugh

Just for clarity, we can’t actually use non-tokenised validators right now. Technically we aren’t built for it, but until withdrawals are allowed we can’t take the liquidity risk. It’s a possibility in the future though, after withdrawals are enabled.

We can however upgrade the pool contracts to allow others to take flash loans from the Nexus pool (and charge something for the pool). We had thought of it but in terms of priorities it is quite a bit lower than other core functionality. We will get there at some point.

1 Like

For tokenised staked ERC20s, stkr has already had slashings.

i finally signed up for an account just to echo @rchen8’s point

1 Like

Park those ETH with KEEP Network to earn 200% APY. They require ETH as collateral for BTC <-> tBTC minting process.

https://blog.keep.network/a-new-rewards-mechanism-deef3412c3e1

I imagine you guys can reduce risk by
(1) Increasing 150% ETH-to-BTC collateralization to 500%
(2) Charge a higher minting fee (5% instead of 0.5%) on $tBTC minting using your $200M $ETH for reinsurance?

@evanvanness @rchen8 I think the important point is we don’t cover the same risk that we invest in, or at a very minimum we manage exposure appropriately so we’re comfortable with total loss (claims + investment loss) in the event of failure.

So I wouldn’t exclude stETH just because it contains smart contract risk, we can choose to invest in it rather than offer cover for it.

Also, we want to provide cover for Coinbase etc so the same issue arises with centralised providers.

So it become a wider question on if a particular investment opportunity is more valuable to the members from the investment perspective or from a cover perspective.

As we can’t currently use non-tokenised staking providers (all assets must be on-chain) then our choice is use something like Lido to stake the ETH and not offer cover, or the other way around.

I want to offer the idea of using tranche risk pools for generating yield. I think risk tranching is the right fit when investing funds that are 1) not personal funds and 2) are extra sensitive to the risks involved. The leader in the tranche risk space is Saffron Finance. In a nutshell how it works is the risk and losses from the lower risk tranches are passed down into the higher risk pools. That means the high risk/higher yield pools essentially act as a hedge for the low risk/lower yield pools.

1 Like

I like the idea of low yield tranche of Saffron Finance. + we don’t cover them
I am not qualified to audit their code tho, maybe the protocol is perceived as high risk?
Definitely worth taking a deeper look and maybe allocating a low %

Since this isn’t a pressing issue, I wanted to also give a heads up on another potential solution that hasn’t come up yet: https://stafi.io/

Stafi is different than other staking solutions in that it’s its own blockchain as opposed to having a staking solution controlled by a centralized entity or a DAO as in the case of Lido. Staking solutions are being built for other PoS blockchains, so you will have rETH, rAtom, rDot rXTZ etc…

While still in the early stages of development and low liquidity, progress has been rapid and rETH has been released:

Since Stafi will be creating staking derivatives solutions for many PoS blockchains, incentivizing liquidity at first might be a challenge but I think once it gets going it will become a blackhole for staking solutions.

I view Stafi kinda like Thorchain in a way, instead of just building another DEX; you build a liquidity blockchain. Stafi is the staking blockchain.