Summary: Add a per validator staking cap for each delegator to encourage large delegators to spread their stake more evenly to encourage decentralization.
Background: Currently large delegators such as exchanges and hedge funds control a large percentage of staked ONE. Unfortunately these large wallets have not spread their stake as evenly as many in the community would like. To encourage these delegators to diversify their stake, we propose a maximum cap on how much ONE a delegator can stake to a single validator.
If Binance spread their stake to 5 million unelected validators each, this would mean that this single wallet alone could allow for 300 more validators to join the Harmony ecosystem.
This rule will help with the node decentralization goal on the roadmap by introducing more validators into the Harmony ecosystem and thereby increasing the amount of nodes per shard: Decentralized Nodes
Specification: Add a maximum cap on how much ONE can be staked by a delegator to each validator. For delegators with existing stake over X for each validator, they will only receive rewards for up to X stake each where X is the cap. This proposal passes if the combined stake-weighted votes for any cap reaches 67% and the total votes is 51% or greater. The final cap will be a stake-weighted average of all the caps that were voted on rounded to the nearest million.
@leo Would the Harmony team have any recommendation on how to carry out this proposal? Is it feasible to enforce for current staking amounts or could it only be applied to new staked ONEs going forward?
Same exact thing I want to know as Fortune asked. Is there a retroactive ability to force this on existing delegators if this is passed? Because if not then I believe the impact will be minimal at best.
If implemented, then maybe a transition phrase of having rewards based on only the threshold (Ex: first 5 million ONEs) and people will move the funds themselves. Anything exceeding the threshold will not be counted towards reward.
Interesting proposal and I think it’s a step in the right direction. Instead of using fixed caps, what about dynamic caps? In this case it would be Effective Median Stake. 1 EMS, 1.5 EMS, 2 EMS, etc.
Ideally this would have to be retroactive to have an immediate impact, but if I was a delegator that trusted Harmony and came back after a year to check on my rewards, I would be pretty annoyed If I found that I’ve lost several months worth of rewards.
Finally, Harmony has some partnerships like Stake DAO where the fundation is staking 50 million with their validator. What would happen here?
I thought about dynamic caps at first but then realized it could become frustrating if some delegators lose rewards when the cap unexpectedly lowers. The advantage of having a fixed cap is that there won’t be any surprises.
I’m not sure how this would even be possible by the chain. The wallet owners control their own private keys making this impossible to change across the chain. The whole idea behind the cryptographic chains. The person in charge of the private key hold full control over the assets assigned to that wallet. So in my humble opinion, retroactively this would not work if my understanding of the whole crypto system is accurate. Which I could be wrong.
One way we could do this is to cap the staking rewards to 5-20m per validator. Their stake to the validator would not change but only the rewards would be capped.
This certainly seems more feasible and would motivate the big delegators to distribute more of their holdings to various validators. Although I could see an issue with the hard cap as the number of tokens are made available. A percentage of the total staked value might be a better way to cap it, and it would adjust dynamically as the total staked value is changed with each epoch. @RoboValidator
Another variation of this rule could be to allow staking over 5-20m but only cap the rewards – with the caveat that validators still receive the same reward fees.
So if Harmony wanted to, they could go ahead and stake 50m with StakeDAO. StakeDAO would earn still rewards as if the full 50m were staked however Harmony’s own reward would be capped at 5-20m.
Exceptions create problems with coding and lead to unexpected results. Keep it simple and consistent across the board. I think the hard cap should be based on the percentage of staked tokens from the previous epoch. It would be up to the delegator to ensure that they stay below that cap to remain eligible for any rewards that exceed the cap. These rewards could be distributed to the remaining delegators throughout the ecosystem that are staked but didn’t exceed the cap.
I think for anyone who is currently staked pre-cap, those are grandfathered in, anyone new or if they unstaked and moved to another delegator, the cap would be applied. Could give it a year (or a set number of epochs) to allow folks an opportunity to not be penalized by a cap penalty.
Another option to decentivize the delegators in staking all their ONEs with a specific validator by reducing the rewards after the hardcap.
Lets assume there is a 10m cap on each delegators
The delegators will still earn their usual reward on the first 10m but after that, anything that goes above the 10m is still counted for reward but they will earn less.
This way, delegators can decide by themselves if they want to delegate all of their ONEs to a single validator knowing that they will be earning a bit less reward for anything above the hard cap.
I think this approach is safer for those who do comes back after 1 year or left their ONEs to stake without being much around as pointed by @Rutilant_Hub
If we do hardcap, the delegators can just spread their ONEs to different wallets and still delegate to the same validator tho. So for a delegator with 10m ONEs, and the hardcap being 5m, he can simply create 2 wallet and still delegate to the same validator.
What can be done? the hardcap should be lower (~1m), creating and maintaining 3-4 addresses will be more of a pain and might incentivize the delegators with only 1 wallet to delegate to other validators, maybe.
You can’t stop delegators from doing what they want. If they want to delegate all their $ONE than so be it, they can make multiple wallets. The big things is that most will take the path of least resistance. So if a cap was in place they might just delegate to another validator.
I understand the proposal’s objective and agree with it. But I feel like this will add complexity to the protocol when the same outcome could be achieved with easier methods. I also don’t see how this could be applied to large wallets who are already delegating to the largest validators. How could you enforce this retroactively? Because if you can’t enforce it retroactively, then I’m not sure this proposal will actually have much of an impact
I would rather have the APY decrease as validators grow their delegations (jacksteroo outlined this in a previous discussion thread). That way delegators are incentivized to stake with smaller validators, enhancing protocol decentralization with an increased number of validators. This seems like the cleanest method, imo
It may or may not have much impact on wallets who are currently delegating with large validators (see below). But I think it would have a large impact on where new delegations are made
I want to see Harmony reach 1,000 validators. Harmony literally lists this as one of the protocol’s features on it’s website’s home page. So I think it must be part of our objectives here as well. And we simply won’t ever come close to 1,000 validators with the system we currently have in place
.
These wallets would have the option to remain with their current large validator but receive smaller rewards, or move their delegation to a smaller validator in order to receive higher rewards again. But the protocol wouldn’t forcefully redelegate these funds