FIP-XX: Diversify ETH Staking Exposure

FIP-XX
Status: Draft
Author: @Jstar

Summary

Diversification is key for the success of sustainable PCV revenue generation and hence the ability for FEI Protocol to protect the peg and drive FEI utility and adoption. Currently, ~17% of the PCV assets are exposed to Lido as the sole staking provider, equating to ~25% of the PCV’s ETH assets.

Staking provides a key foundation of the PCV investment framework given its attractive risk/reward (currently ~5% and expected to rise post-Merge). In order to maintain the benefits of ETH staking whilst increasing diversification of PCV assets, I propose that X% (tbc) of the current ETH staked with Lido is migrated to StakeWise. This will also allow the PCV to benefit from certain unique features that StakeWise provides, such as a higher staking yield and the ability to re-invest staking rewards to compound staking returns.

What is StakeWise?

StakeWise is a fast-growing liquid ETH2 staking platform with ~50k ETH staked by over 3.75k users (see Dune Dashboard). As a liquid staking protocol, StakeWise allows users to earn staking rewards without locking assets or maintaining staking infrastructure. Users of the protocol receive tradable liquid tokens (sETH2) in return for the deposited ETH. The current APR of sETH2, after fees, is 4.80%, compared to Lido’s stETH at 4.65%.

StakeWise provides:

  • Institutional grade infrastructure - StakeWise and Blockdaemon have recently announced a partnership to provide the first liquid institutional ETH staking platform.
  • Proven stability - strong history of validator performance, with no slashing/downtime penalties to date.
  • Competitive fee structure - 10% protocol fee on staking rewards.
  • Community governance - StakeWise is managed by a community-owned DAO.
  • Fully trustless staking solution - non-custodial staking combined with DAO owned exit signatures means node operators cannot refuse to close down validators and hold pool funds hostage.

Dual Token Model

StakeWise is the only liquid ETH staking provider to use a dual token model - paying rewards in a separate token, rETH2, mapped to ETH earned in the Beacon Chain. The other token, sETH2, represents the ETH deposit and acts as an interest-bearing wrapped ETH. Holders of sETH2 tokens earn daily rETH2 rewards no matter if their sETH2 is in their wallet or deposited in a contract.

Benefits of a dual token model include:

  1. Highest native staking yield and maximum capital efficiency - When depositing into a single token pool, like Lido, users are rewarded with stETH which represents a share of the total pool. That pool contains both staked ETH (interest bearing) and previously earned ETH rewards (non-interest bearing). If 10% of the pool consists of historic, non-interest bearing rewards, then only 90% of any new deposits are benefitting from staking yields. This problem grows larger as time goes on, with the percentage of rewards within the pool increasing. A dual token system solves this problem, with 100% of new deposits earning staking yields, as previous pool rewards are correctly assigned to the users who earned them. Hence, the yield for a dual token system is higher than single token models.

  2. Enables compounding - Users can take their rETH2 and swap it back into sETH2 via the secondary markets. It is a common misconception that ETH staking platforms are compounding yields using rewards, it is simply not possible with a single token system.

Risks/Considerations

Smart Contract Risk:

StakeWise launched on mainnet in March 2021 following a 9-month beta phase. Stakers are exposed to smart contract risk. To mitigate this, the protocol has undergone audits by 3 separate auditors: Runtime Verification, Certik and Omniscia (reports are here). The code is open-source and found here. The protocol has had no security concerns to date and an upcoming $200,000 bug bounty program with Immunefi.

Counterparty Risk:

StakeWise’s recent update is a key step for the decentralisation of the platform. Node operators can apply to run institutional-grade infrastructure courtesy of the StakeWise open-source deployment package. Applications undergo vetting by a DAO appointed committee before entering a testnet. A final DAO vote is required for a new node operator to go live on mainnet. Since the launch of this upgrade in January 2022, there are already 12 node operators at various points of onboarding to the testnet. Names include leading node operators such as Blockdaemon, Chainlayer, CryptoManufaktur and VeriHash. Further information on this recent update can be found here. Node operator quality is of paramount importance to mitigate security concerns and provide the staking pool with the highest possible yield.

Note that the StakeWise DAO is community owned and responsible for managing all aspects of the protocol.

Slashing risk

There is currently no slashing insurance in place, however it is a high priority within the 1Q2022 pipeline. There have been no slashing or downtime penalties to date. StakeWise is focusing efforts to on-board quality node operators to minimise risks of slashing/downtime penalties (see above).

DAO key management risk

StakeWise was the first platform to implement non-custodial ETH staking. The StakeWise DAO has control over its node operators’ validator keys, and hence decisions around the timing of withdrawals. Validator keys are registered by node operators utilizing StakeWise Deposit CLI, and are simultaneously split into shards that will be handled by the DAO elected Validator Committee. In case the node operator ignores the DAO’s decision to exit the validators and execute the withdrawal of funds, the Validator Committee can execute a forced exit of the validators. This would trigger a withdrawal by reconstructing the validator key from the shards and generating an exit signature that is submitted to the Beacon Chain on behalf of the StakeWise DAO. Thus, the DAO always remains in control of pool funds.

Liquidity Risk

ETH will remain locked within the Beacon Chain until after the Merge and validator withdrawals are enabled. Once withdrawals are possible (expected January 2023), StakeWise will allow users to burn sETH2 and rETH2 tokens in exchange for ETH.

Until then, stakers are required to exit the staking protocol via secondary markets. The StakeWise liquidity pools utilise concentrated liquidity on UniSwap V3 to maintain a tight peg between ETH and sETH2. This peg has proven to be stable and resilient over the lifetime of the protocol:

StakeWise incentivises its users to provide liquidity into the UniSwap pools using its governance token, SWISE. Pool APYs have consistently stayed above 12% and the DAO will continue to provide such incentives to grow liquidity alongside TVL.

There is currently $105M of liquidity within the sETH2 <> ETH pool, comprising 16.3k ETH and 17.9k sETH2. Slippage is detailed below and equates to ~7 weeks worth of staking rewards should the PCV look to liquidate its entire sETH2 position. This slippage risk is partially mitigated as Fei Protocol is overcollateralized and may not need the resources in the short term/prior to validator withdrawals being enabled. The allocation to StakeWise would represent only a small % of the total PCV.

SW SLippage

Slippage between rETH2 and sETH2 when reinvesting rewards to compound staking yields will not be an issue given the depth of the rETH2 <> sETH2 pool and the expected size of re-investments.

The price of sETH2 currently sits at a slight discount to ETH given the in-built liquidity premium. Each unit of sETH2 and rETH2 is backed 100% by ETH held within validators. Should the sETH2/rETH2 peg drop below a certain threshold, it is expected that entities would step in to purchase the tokens at a discount and hold them until withdrawals are enabled. This will maintain a lower price limit for both sETH2 and rETH2.

Economics of moving from Lido to StakeWise

The PCV will incur slippage whilst exiting the Lido staking pool, by converting stETH back into ETH via Curve. This slippage equates to ~9-11 days worth of staking rewards and will ultimately be re-paid by the increased yield offered by StakeWise.

Slippage:
Lido  Slippage

Dual Token Benefits:
Dual Token Benefits

Note: I have an affiliation with StakeWise and will be here to answer any questions regarding the protocol

Specification

Re-stake the voted amount of ETH from Lido to StakeWise.

Vote - Approval Voting

Restake

  • Re-stake 10k ETH from Lido to StakeWise
  • Re-stake 7.5k ETH from Lido to StakeWise
  • Re-stake 5k ETH from Lido to StakeWise
  • Let’s explore other options for diversification

0 voters

1 Like

Thanks for posting this!

The main factors here are decentralization, on-chain liquidity, and ease of integration. It seems that stETH has over 4bn in liquidity on Curve alone, not to mention the Balancer pool which is also very liquid. The native rebasing also makes it easy to use for our infrastructure.

It would be both a dev lift and a technical risk to migrate over some stETH to sETH2 and the benefits are not salient to me from this post. Curious to hear what the community thinks

1 Like

Hey Joey - appreciate you taking the time to review the proposal!

Got it - so the two concerns are depth of liquidity and ease of integration. If I could address them individually:

  1. Liquidity depth -
    If the DAO approved 5k ETH for example, then it would account for 2% of PCV assets. The liquidity depth issue would only arise should the PCV need to liquidate all ETH assets at the same time that StakeWise experiences a mass exit event. Should something happen for both events to occur simultaneously we probably have bigger things to worry about :sweat_smile: I am also aware that you are developing the sentinel internally, so that could help alleviate any liquidity concerns?

  2. Ease of Integration -
    I appreciate the token mechanics would be new to the PCV, however we have not experienced any issues with others DAOs we are working with regarding an integration. I’m conscious each DAO has unique infrastructure though - what specifically about rebalancing tokens suit you better? I’m confident any technical hurdle can be easily overcome though, the StakeWise dev team would obviously be on hand to assist with anything there.

I have the same opinion with you about that. With sETH2 APR 4.8% and Lido APR 4.65%, it seems very little increased in returns. And it may makes more potential risks than stETH.

@Jstar what about having sETH2 added to pool 146, the Tribe ETH pool on Fuse? this can allow your users to lever on interest rates https://app.rari.capital/fuse/pool/146

@joey we would love to integrate further with you guys via Fuse. I have responded on Discord with an idea we have to utilise FEI in an up-coming integration of ours. Would be very keen to expore this option further with you. It might be easiest to take the dicsussion offline so I can share the full details of the integration before our formal announcement.