Proposal: Rebalance the DVT Rewards Split Between Node Operators and the Lido Vault

TL;DR: Proposal to adjust the share of DVT rewards from 80/20 to 33/67 in response to changes in costs and rewards.

SUMMARY:

This proposal addresses the way DVT rewards (particularly SSV and Obol rewards) are divided and shared by the node operator and the Lido Vault.

It is particularly relevant to stakers who participate in the Community Staking Module (CSM), but should also be of interest to members of the SDVT cohort and potentially members of the Curated Set who are using SDVT as a validation tool.

The Problem:

The impetus for the proposal is the declining amount of Incentivized Mainnet Rewards being distributed by the SSV network, and how it compares to the SSV Network Fee as a cost to use the SSV network.

The Incentivized Mainnet was designed a mechanism to reward early adopters of the SSV Network, to make it worth the effort of migrating validators into the network, and to offset the Network Fee that supports the cost of network operations. As the amount of ETH staked on the network grew, the incentive was set to decline.

Over time the reward has declined from the initial 30% rate down to 20%, then 10% and now is approximately 6%. (The actual reward rate is less than this – more like 3% – since the program calculates rewards based on a $10 minimum SSV token price, but SSV is currently around $5.)

At the same time, the SSV network fee has increased from 0.5% to 0.75%, and is about to increase to 1.0%.

For most SSV operators, the network fee at 1% is less than the incentive reward at 3%, so it’s a net positive.

The twist comes from the way Lido shares the rewards with the operators who run CSM validators. Because CSM validators are a joint effort between Lido (who provides the funding) and operators (who provide expertise, labor, and an ETH bond), the incentive rewards are shared between them. Currently the split is shared with Lido receiving 80% of the rewards and the operator receiving 20%.

At the current rates, the CSM operators are now receiving 20% of a 3% incentive, which is 0.6%. So the 1% fee to run the SSV validator is greater than the amount earned by the operator. Bear in mind that the SSV network fees are solely the responsibility of the operator, not Lido, and those fees must be paid in advance to the network.

The emergent trend that I believe we will see is CSM operators REMOVING their CSM validators from the SSV network, and reverting to simple solo staking.

While solo staking will be marginally more profitable for the operators, it effectively ends the free incremental rewards being received by Lido currently. In short, it kills the goose that lays the golden egg.

Proposed Solution:

In order to reset the equilibrium so that SSV Mainnet Rewards are once again profitable for all parties – operators and Lido – I propose that we update the reward split, and allocate a larger share to the CSM Operators.

At current rates, roughly 1/3 of the rewards are going to the cost of the Network Fee. (1% fee out of 3% effective rewards).

I propose that we earmark 33% of the rewards specifically to cover the network fee, and then we find a reasonable split of the remaining 66% between Lido and the CSM operator.

I think a reasonable case could be made for giving the CSM operator anywhere from 50% to 67% of the total incentive rewards, which would net them between 17% to 34% of the total rewards after they have paid 33% to network fees. Lido in turn would receive between 50% and 33%, depending what the final split was. I personally think the 67/33 split makes the most sense, because the net reward to the CSM operator needs to be large enough to compensate for the fees they must pay in advance plus the additional work of splitting keys and running operators (and maybe paying other operator fees as well).

In short, I recommend that

  • 1/3 of the rewards are paid to the CSM operator to cover network fees
  • 1/3 of the rewards are paid to the CSM operator for their time and trouble
  • 1/3 of the rewards are paid to the Lido vault

In Closing:

I think the current problem can be resolved quite easily by adjusting the fee split, which is fortunate.

I’m much more familiar with the SSV Network rewards and fees, but I suspect the same issues may apply to the Obol Network rewards. And it probably makes sense to structure the split the same for both programs.

A discussion can be had about exactly what the new fee split should be to best align with what all parties bring to the table, time or treasure.

Keeping the status quo will likely cause CSM operators to revert to solo staking, in which case Lido gets no rewards at all, which is clearly a much worse outcome.

Full Disclosure:

I am a member of the SSV DAO, a Lido CSM operator, a Lido SDVT operator, a solo staker, and a past contributor to the CSM program.

3 Likes

Hi @gbeast, thanks for bringing this issue forward.

While I agree that changes to the configuration are likely needed, I don’t support the proposal with the current split structure.

I do agree there is an issue, and I believe that it would make sense to mitigate the impact of the network fee cost exceeding the amount of incentives generated.

However, part of the issue here lies with the SSV fee increase coming at the same time as a significant decrease to the amount of SSV incentives generated due to the SSV DAO’s inflation cap. The SSV network fees are increasing at the same time as the value of incentives (in both USD and SSV terms) are decreasing.

Lido doesn’t share rewards. Neither the protocol nor the DAO receive any of these rewards. The SSV IM incentives are simply split between stakers in the Mellow DVV and Node Operators who run the underlying SSV-based validators as a part of the Lido protocol. Lido (the DAO) does not provide any funding or receive any incentives, the capital for validators (apart from the bond) comes from users of the protocol (stakers).

CSM Node Operator’s provide cover of just 7.5% of the ETH required for their first 32 ETH validator, and only 4% after that, however, receive 20% of the IM incentives for the validators they operate currently.

The main stakeholder in terms of incentive distribution should be the main capital provider, which in this case is stakers, with those incentives flowing to the Mellow DVV (again, not the Lido DAO).

While I am open to the idea of increasing this split, I do not think that it makes sense for the majority of incentives to flow to the Node Operator, or at least for the stakers share to cover both the SSV Network Fee and also provide significant additional incentives to Node Operators on top of that.

Personally, I think a better option would be for the SSV DAO to consider a delay to the Network Fee increase until the SSV/USD price increases to such an extent that the inflation cap is not limiting payment of the Network Fee, alongside a possible increase to the CSM & Curated Node Operator share of incentives.

We expect to have additional conversations with SSV DAO contributors at Devconnect over the next week and will follow up with potential solutions that hopefully can address these issues.

9 Likes

Hi @GBeast - a related proposal to this discussion has just been published. Please take a look and provide feedback at your convenience - Proposal: DVT & DVV Incentive Allocation Changes.

1 Like

This makes complete sense to me. A lot of Lido CSM keys are being run on private SSV clusters. There should be a mechanism where SSV operators that choose to use their SSV clusters to run Lido keys receive an amount of the SSV rewards to cover the SSV network costs. And everything else could go to Lido. Right now running Lido CSM keys on SSV is a net negative for the node operator.

1 Like

really make sense for who running with SSV cluster.

Rebalancing the split is a necessary recognition of the “Operational Overhead” inherent in DVT. Running a distributed cluster isn’t just about hardware; it’s about the increased coordination cost and the redundancy management that a single-node setup doesn’t face.

If we keep the Vault’s take-rate too high during this transition, we risk disincentivizing the very operators who prioritize decentralization over pure margin. Has there been an analysis on the “Breakeven Threshold” for smaller Obol or SSV clusters under this new split, especially considering the potential for lower MEV smoothing efficiency in a distributed environment?