Draft Proposal: Expanding Rocketpool's Market Reach with No-RPL-Bonded Megapools

Draft Proposal: Expanding Rocketpool’s Market Reach with No-RPL-Bonded Megapools


We propose changes to the Rocketpool protocol aimed at expanding the total addressable market (TAM) of Rocketpool Node Operators (immediately) and ETH => rETH stakers (eventually). In addition to the options they currently have, Node Operators would be able to

  • Lock RPL tokens in a vault for fees
  • Run a new type of pool that trades off RPL price volatility for borrowed-ETH rewards

Some Personal Background

As a Node Operator with an RPL collateral ratio under 10%, I won’t be starting new minipools despite having idle ETH I’d like to stake. Going into rETH is not an option for me as the tax liabilities triggered would dwarf the current (low) APR. I’ve looked into starting new validators as a solo staker, but was put off by the nitty-gritty of doing so. And there’s just no way I’m buying more RPL after seeing the volatility RPL exhibits.

The current UX is unfriendly to those of us who aren’t interested in speculating on RPL and just want to stake their ETH using Rocketpool’s tried-and-true framework. Truth is, with Megapools set to vastly improve the cost-effectiveness of staking incremental ETH, I’d willingly forego most of my 14-15% cut of borrowed-ETH rewards if doing so would shield me from RPL volatility while allowing me to stake my idle ETH with Rocketpool’s best-in-class.

  • Software
  • Dashboards
  • Smoothing-pool
  • Rescue-node
  • Community
  • Lindy

I think there are others like me in the node operator market. See for example,

How can Rocketpool expand its TAM and capture our business?

Details:

  1. Establish a No-RPL-Bonded (NRB) Megapool: As the name suggests, this megapools does not require RPL to be bonded. NRB pool operators borrow 24 ETH for every 8 ETH bonded.
  2. Establish an RPL Vault: Node operators with one or more pools (eb8, leb16, or nrb) have the option to lock RPL tokens in a vault for up to 1 year.
  3. Lockers’ Cut of Borrowed-ETH rewards: 10% of an NRB megapool’s borrowed-ETH rewards goes to the vault as a Vault Fee.
  4. NRB Operator Cut of Borrowed-ETH rewards: NRB megapool operators receive 2% of borrowed-ETH rewards.
  5. Vault Fee is Sent to the rETH contract: The vault fee is simply added on to the ETH sent to the rETH contract whenever the operator takes a distribution from the smoothing pool, their fee distributor, or their NRB megapool.
  6. Vault Fee is Claimed as rETH: Depositors into the vault can claim their share of the vault fee as rETH at the current protocol rETH/ETH exchange rate at the end of each epoch.
  7. Locked RPL is Eligible to Vote: RPL locked in the Vault is eligible to vote on a formal proposal (e.g. on Snapshot) if the RPL
    • Was locked at the time the proposal was posted
    • Will remain locked for 2 weeks past the end of voting

Notes:

  1. The current tokenomics do not offer a clean separation of roles—one needs to either be lucky or be an RPL tokenomics/market-dynamics maven to profitably operate a Rocketpool node. This proposal creates space within the Rocketpool umbrella for a pure play node operator role that’s competitive with solo-staking. The provision of such a role will expand Rocketpool’s TAM.

  2. Because NRB operators do not bond RPL, there’s no need for unsustainable RPL inflation to compensate for decreased capital efficiency. Indeed, NRB pools can play a central role in Rocketpool’s path to sustainable inflation.

  3. A 10% Fee on borrowed-ETH rewards is double the take rate of LDO holders. So, theoretically speaking, if NRB megapools were the only minipools available, RPL could sustain the same marketcap as LDO with half the amount of borrowed-ETH.

  4. The 2% Fee on borrowed-ETH is a nominal amount intended to nudge would-be solo-stakers to stake with Rocketpool. It also serves as compensation to them for taking on an increased penalty risk.

  5. With rETH holders losing only 12% of their ETH rewards as opposed to 14% with eb8 and 15% with eb16 pools, rETH becomes more attractive to stakers if NRB pools draw significant interest.

  6. The requirement that RPL be locked for 2 weeks past the end of voting to be eligible for voting is intended to thwart vote-and-run type governance attacks.

FAQ

  1. Why is RPL locking limited to Node Operators with at least one pool and not extended to all RPL holders?

    The limitation is in place because locked RPL could be eligible for voting. To calculate a voting power accurately, we need locked RPL to be associated with a node address. The requirement that the node operator have at least one pool acts as an anti-sybil measure.

  2. Aren’t Megapools weak anti-sybil measures due to their capital efficiency?

    Yes, this is a challenge for all megapools, not just NRB ones. Any anti-sybil solution designed for megapools in general should be applicable to NRB megapools.

Open Questions:

  • Do the Vault and Node operator fees (10% and 2% of borrowed-ETH respectively) seem right? Ideally these rates would be adjustable by the pDAO and the adjustments would apply to all NRB minipools (not just new ones). Unfortunately, making this so poses a technical challenge: When borrowed-eth rewards are distributed, how do we ensure that borrowed-ETH rewards are taxed at the fee rate prevalent when the rewards accrued?

Next Steps:

  1. Community Discussion: We encourage the Rocketpool community to review this proposal and provide feedback.

  2. Technical Analysis: A technical feasibility study is needed to understand the potential impacts and complexities of the proposed changes.

  3. Formal Proposal: If the community and technical analysis support the proposal, it can move to the formal proposal stage for voting.

5 Likes

Discord thread: Discord

looks like a interesting proposal. would be good to see some theory crafting on wether this or a similiar model could be a sufficient demand driver for RPL to replace the currently forced collateral requirement.

in regards to FAQ 1. Why is RPL locking limited to Node Operators with at least one pool and not extended to all RPL holders?

what if we allow Non node operating RPL holders to delegate their RPL to Node Operators and thereby also earn rewards? wouldnt that eliminate the sybil voting attack vector while making RPL attractive for non operators? i think that could be a key point.

1 Like

I like this proposal but not sure if the 2% makes sense - that would be 0.024 ETH or 44 USD per annum (under price of 1850 USD/ETH) for a LEB8 pool, I don’t think anyone will care about this.

Could we give an incentive in the form of RPL instead?

1 Like

I like the idea, something must be done to RPL tokenomics to increase rETH supply in the market share of LST

1 Like

A simpler (and IMO stronger) version of this proposal is here:

The reasoning for the changes is covered in this Discord thread

Haven’t had time to argue for the proposal recently. Just want to make sure folks are looking at the latest version.

1 Like