[Proposal] Increase Validator Commission From 11.7% to 20%

[marcotor]

Give your proposal a title:
[Proposal] Increase Validator Commission From 11.7% to 20%
Proposed by lovali (www.lovali.xyz)

Summary:
We propose to increase the validator commission on delegators’ share of staking rewards from its current value of 11.7% to 20%. We believe that this change is in the best interest of the Vega network and community.

There are a number of justifications for this:

In the short term, now and immediately following the launch of Alpha Mainnet, rewards paid to smaller validators (by stake) are unlikely to cover operating costs, let alone provide any significant return.

• Currently the smallest validator is making considerably less than $500 a month in staking rewards.
• It is very likely this will remain below $1000 a month for some time after launch, even as fees begin to grow with trading volume.
• The hosting costs alone of running a performant validator node and data node in mainnet likely exceed this amount for at least some validators.
• These figures are based on 13 validators. As the expectation over the next 6-12 months is for the network to grow to 20–25 or more validators, there will be further downward pressure on the rewards available per validator.

In the medium term, the project team have signaled their intention to iterate rapidly to add features and improve the protocol. We strongly support this, however it requires ongoing and responsive engagement from all validators, which we believe will be more forthcoming if incentives are properly aligned.

• Currently, validators operate both a testnet and a mainnet.
• There are regular releases requiring deployments and other activities, both to prepare to launch new code on mainnet, and to deploy to mainnet itself.
• Incidents and issues also require validators to be engaged and responsive.
• This represents a significant time and effort commitment compared with majority of token holders who receive staking income passively and are dependent on validators for the future success of the network and continuation of such income

Increasing the commission share would therefore:

• Increase the chance of rewards covering validator costs in the first 6-12 months after launch
• Ensure that serious professional operators want to validate the Vega network
• Increase validator engagement and responsiveness to incidents and upgrades
• Reduce the susceptibility of validators to corruption from malicious parties looking to exploit the network
• Ensure competition to validate the Vega chain

For these reasons, we believe an increase from 11.7% to 20% is justified now given the current token price, reward amounts, and phase of the project.

We do not necessarily believe the commission level should remain at 20% indefinitely and would welcome future proposals to reduce the commission again if appropriate, once there is significant fee revenue on the network.

We welcome feedback, comments and questions below, and look forward to discussing this proposal openly with the community to increase the likelihood of a successful vote.

For reference, details of the current share and the schedule for staking reward payments can be found in this blog post: Vega Staking Rewards & Inflation. Staking rewards on Vega are paid to all… | by Vega Protocol | Vega Protocol.

Thanks

[Ed-Commodum]

Marco, thanks for sharing your proposal.

We think this proposal makes a lot of sense and should help to secure the short to mid term future of the Vega network. It is important that Validators are properly incentivised to operate the network to the best of their abilities.

The full suite of infrastructure required by the vega network for each validator is significant, in addition to the validator node, there is also a requirement for sentries and at least one data node. Taking this into account, and given the current state of the market as well as the ongoing decrease in VEGA issuance, it will certainly be difficult for some of the smaller validators by stake to break even. Soon the network will be generating fee revenue that will supplement the decreasing VEGA token rewards. However, since these revenues are dependant on trading volume, there is a high degree of uncertainty in how much a validator can earn in the initial months after the launch of trading. To temporarily increase the commission in these initial months would help give validators confidence that they will be duely compensated for their service.

Additionally, given that Vega is a highly complex system, there is a reasonable amount of technical overhead and knowledge, as well as significant time investment, required by each validator to fulfil their role and coordinate upgrades. We believe this necessitates proper incentives.

It is also prudent to point out that the Validators could bear a significant amount of the regulatory risk that is associated with Vega by operating the network. Given that there are no clear guidelines as to what constitutes operating an exchange with respect to decentralized derivatives protocols we think it makes sense that there be adequate compensation for validators that are willing to take on these risks.

The other way to increase the validator revenue without increasing the commission would be to increase the infrastructure fee. However, we believe that this would not be in the best interest of the network since the infrastructure fee is already set very high at 10 basis points on mainnet. As such, it seems that the most viable way to support and incentivise validators is to temporarily raise the commission.

Commodum is in support of this Proposal.

[peloclick]

Good night here,

posting as the SysOP of GPValidator Vega validator.
To be honest, I can only support what lovali is writing. We came here after a long wait (BillO can confirm that…) and for us it has been a surprise that there is little chance to cover the costs. Opposite to what other chains do, here rewards depend on the validator’s stake and those, like us, and probably other coming after us, joining the validation set at a later stage of the project, struggle to reach a good balance between costs and net result.

We will stick around but we would deeply appreciate the change if it happens.

Regards

[Jubi]

Thanks for creating this proposal @marcotor !

As a token holder and long-term community member of Vega, I agree with the points you have made. Currently, the Node Operators are the backbone of the network and without them, Vega would be in a dire state.

20% seems like a fair number however, I do think that once meaningful volume comes in this should be adjusted lower and I’m curious if the code implementation supports validator-chosen commission rates? If this is true, we should see commission rates trend downward once trading goes live.

I by no means take away any of the hard work that the Vega team are doing, they’re amazing. But I hope they do realise that running validators comes at a cost and validators won’t stay around forever with the constant delays to the launch of mainnet trading.

[pathrock]

PathrockNetwork supports and totally agrees with the proposal. We also only joined the Vega validator set a few weeks ago and it’s quite hard to break even considering the amount of delegations needed, the time operating a Vega validator on mainnet and testnet etc.

Also agree with the re-evaluating the commission at a later stage and change it as needed.

[marcotor]

Thank you for your comments. Here is the proposal I’m going to submit:

{
  "rationale": {
    "description": "As per prior discussion: https://community.vega.xyz/t/proposal-increase-validator-commission-from-11-7-to-20/4216 We propose to temporarily increase the validator commission from 11.7% to 20% by modifying the network parameter 'reward.staking.delegation.delegatorShare' from 0.883 to 0.8"
  },
  "terms": {
    "updateNetworkParameter": {
      "changes": {
        "key": "reward.staking.delegation.delegatorShare",
        "value": "0.8"
      }
    },
    "closingTimestamp": 1682294730,
    "enactmentTimestamp": 1683294730
  }
}

[barney]

Hi @marcotor, thanks for this. Great to see the active engagement here and the thoughtful analysis of the economics of the network at this point.

Obviously I, and the project team, aren’t well placed to evaluate the costs of operating a validator, however given that the launch and therefore the potential for fee revenue to flow to validators has been significantly delayed while staking rewards have dropped, this seems like a sensible proposal to me.

I’ll be watching how things evolve over the next few months, and would like to see more discussion about whether to reduce the validator commission again in future, but I plan to vote for this proposal.

[barney]

@marcotor Looking at the supplied JSON above, it would be useful to include at least a summary of the rationale you articulated in this thread in the description field.

The reason being that the post and content on this site can’t be verified by a user. It may be edited or deleted and keep the same URL, so summarising the key points in a sentence or two on the proposal itself (in addition to keeping the URL) would be ideal.

[marcotor]

Thank you @barney for your support and your advice: it makes sense to have a little bit more of context onchain.

Here is the revised proposal:

{
  "rationale": {
    "description": "As per prior discussion: https://community.vega.xyz/t/proposal-increase-validator-commission-from-11-7-to-20/4216. We propose to temporarily increase the validator commission from 11.7% to 20% by modifying the network parameter 'reward.staking.delegation.delegatorShare' from 0.883 to 0.8. We propose this to ensure the network's success and provide proper incentives for validators to engage and respond to incidents and upgrades. This increase is justified due to the current token price, reward amounts, and phase of the project, but the commission level may be reduced in the future if appropriate."
  },
  "terms": {
    "updateNetworkParameter": {
      "changes": {
        "key": "reward.staking.delegation.delegatorShare",
        "value": "0.8"
      }
    },
    "closingTimestamp": 1682294730,
    "enactmentTimestamp": 1683064800
  }
}

[marcotor]

I’d like to confirm I submitted the proposal. Here is the transaction: Explorer

[AnyValid]

Hey @marcotor. Looks like you provided an incorrect date in the “closingTimestamp” parameter, so the proposal immediately received the Rejected status. I think you should choose a date later than 72 hours from the date of submission of the proposal

[marcotor]

Just fixed it:

proposal_id: d2157929132456dbc66eecbd478307156066243cd8769306e71f31882c22344f
link to vote for it: Vega Token dApp

Thank you

[lemonty]

It would be great to see the exact cost and not The hosting costs alone of running a performant validator node and data node in mainnet likely to exceed this amount for at least some validators.
I perfectly understand that you want to make money, but you want to do it by taking income from stakers, most of whom have already lost much more than any validator by buying tokens on a token sale at 15, 10, 5 dollars.
I will vote against.

[marcotor]

hi @lemonty
I fully understand your point and that’s the reason why we are here: to discuss all together.

If you check the validators’ profiles, we are a community validator: our aim isn’t really to maximize our profit in the short term. We really want Vega to succeed (exactly like you) and we see the potential of the project in the long term.
It’s complicated to estimate the infrastructure costs for a validator: we all have different HW setups and different operational procedures.
Since the beginning, we spent on average 50h/month not only installing, securing, and updating our node, but also creating tools to help Vega users and other validators. Many other validators are also helping the project evolve daily and I think that effort might be somehow rewarded.

But again: I understand and respect your choice.

[lemonty]

They need to be rewarded, but Vega is lame at the moment. It;s not like anyone outside of the team or seed investors have been rewarded here. We all (ico investors especially) got a huge cut. All we can do is wait and hope that something will change. Cut vega holders even more is not a good idea.

[CryptoSteel]

Transparency

Would you mind share your cost base? While not as lucrative as before I cant see that the top validators like you - Commodum (@Ed-Commodum) and Lovali (@marcotor) are not breaking (at-least) even. Given that your rewards are currently around 1000-1500 Vega a month. At time of writing this amounts to around $1000 to $1500.

History

Secondly if we look at cumulative stake/rewards this proposal is kind of ridiculous on the border to being offensive. See last column for cumulative rewards, and add that the token price was trading at $10+ during the beginning when you got rewards for basically running an empty mainnet.

Small and new validators

However, AnyValid has a point regarding smaller and newer validators but this could potentially be solved by tweaking other parameters such as optimalStakeMultiplier and competitionLevel to increase the rewards for smaller and newer validators.

[Marcin]

While I appreciate the reasoning behind this proposal, I agree with @CryptoSteel point about the cumulative validator rewards. It’s important to remember that running a node in volatile markets means that profits are often heavily skewed towards bull markets, and breaking even during bear markets is a real possibility.

A good example of this is NalaDAO, which accumulated 60k VEGA tokens over a two-year period before deciding to stop running a validator. If we take into account the price fluctuations that occurred over the past 24 months, those 60K VEGA tokens would have been worth roughly between $120k and $400k over the two-year period.

I understand that this problem has mainly arisen due to the lack of real-world protocol usage and the long testnet period. This has had an impact on the token value, which is also influenced by wider market trends. However, it seems like we are on the cusp of launching the mainnet, and the market trends are looking more favorable.

Taking these factors into account, we need to carefully consider the long-term implications of any decisions we make about validator rewards. While it’s important to incentivize participation in the network, we also need to ensure that rewards are fair and sustainable over the long run.

I support the idea of tweaking parameters such as optimalStakeMultiplier and competitionLevel to increase the rewards for smaller and newer validators.

[Ed-Commodum]

Hello CryptoSteel, our goal as a community validator is to provide the best service we possibly can to the Vega network and its ecosystem. As mentioned previously the required infrastructure is extensive, but I can go into a little more detail in this post. Across both the mainnet and the testnet a validator is expected to operate 2 validating nodes, 6 sentries, and 2 data-nodes. The recommended spec for each node, as per the docs (Infrastructure requirements | Vega Protocol Documentation),
is 12 CPU cores and 32GB of RAM with 2TB of NVMe SSD storage. As you can imagine when running this many nodes the cost quickly adds up. These costs also do not include redundancy, security features, and other infrastructure that may be provisioned as part of a highly-available setup, for example; remote signers/key managers or Hardware Security Modules.

In addition to the infrastructure itself we also must provide a significant time commitment to properly monitor and manage our infrastructure, assist the Vega team, and communicate and collaborate with other validators. We invest time in DevOps, performing network upgrades, troubleshooting, testing monitoring solutions, and testing Infrastructure As Code for deployment. Additionally, we are on call 24/7 should anything go wrong with our infrastructure or in the event of a critical bug or a network halt on mainnet. I hope this demonstrates that we are dedicated to fulfilling our role as a validator to the best of our ability.

I also feel I should emphasize that despite having the opportunity to migrate all of our infrastructure to the cheapest available cloud provider (Hetzner), we have not done this because we feel it would undermine the security and reliability of our setup, as well as the decentralization of the network. Instead, we have distributed our infrastructure across multiple providers using a combination of VMs and dedicated servers.

Taking all of this into consideration, I’m sure you can understand that the current level of revenue generated is not enough for most validators to achieve the required quality of infrastructure, security, and availability, as well as the required level of engagement. And certainly not enough to do so while also providing the multiple members of each team with a decent level of personal income to compensate them for their work. Despite this, we still strongly believe in the future of Vega and we will continue to dedicate ourselves to supporting the network, the ecosystem, and the community, irrespective of the outcome of this proposal.

[david]

Thanks Ed for the detailed and thoughtful post.

I have no experience in running network infrastructure. But I do know that behind the scenes all the validators have significant amount of work imposed on them by the project:

• the validator testnet and its restarts aren’t always smooth because the node software is being rapidly developed by the team and hence things change; there are also rough edges(*)
• mainnet maintenance (you all saw there was a need to react when it temporarily went down due to a bug)
• general coordination…

My point is that running a validator (apart from the HW costs) isn’t just “fire and forget and collect rewards”. It’s a significant commitment and we appreciate the work the validator community puts in.

So I strongly support this proposal.

As you’ve seen in the weekly updates - vega bytes we are now close to alpha mainnet. Once trading is live we’ll see what the trading fee revenue will look like (I am feeling optimistic) and then we can re-visit this again, via a new governance vote.

(*) Of course we’re taking validator feedback and working on improving things, so running a node should be becoming a smoother experience over time.

[olympus]

Validators will sell Vega to cover operating costs, which will then put pressure on the price of Vega again. Raising the commission rate to 20% becomes meaningless if the value of Vega drops.

The distribution of $Vega rewards is proposed to remain as is. And it would be good to increase only the reward for $USDx generated by protocol transaction fees to 20%.