I do not understand what will stop big validator from launching 10 small validators to avoid VP tax. If you want to distribute stake to smaller validators I recommend to implement dynamic commission same like offered in this proposal on Terra Classic: Commonwealth
1 Like
Did you check the equation? There are two scenarios here:
-
Most/all delegation is self-stake, example: a validator with 5% VP launches 5 validators of 1% VP each. Well, he doesn’t have to do it in the first place, because according to the formula 5% (VPi) - 5% (the VPsb^2/VPi), so the VP tax would be 0
-
Most delegation external, not self-stake: this other 5% VP validator could try to launch 5 new validators, but good luck attracting this 5% VP to the new validators out of the active set. If he manages to do this, according to the formula VPsb would be almost 0 so can be ignored, now let’s assume median is 0.17% and honestly with this 5 new validators the median is unlikely to change much so. 5% - 0.17%= 4.83%, and in the case of 5 new validators, 5*(1%-0.17%) → 5*0.83=4.15%, not very different
Edit:
I checked also what you mentioned about dynamic commission @Sephiroth, what this does is forcing a high minimum commission on the largest validators. The idea here with VP tax is similar, since this tax would be like a higher commission for the largest validators with low self stake. The difference is that in the case of the VP tax, this tax is then distributed equally for all validators to allow them to run many consumer chains. In the case of the dynamic commission idea it is a partial solution, there is no tax collected or distributed equally, just a hope that this higher minimun fee would lead to redelegations and increase decentralization. The VP tax gives results immediately after being implemented, at the least by distributing this tax to all validators and fostering the growth of the AEZ, and in the best case on top of this reducing centralization over time, thus increasing the security in the Cosmos Hub and hence its value
1 Like
VP Tax idea is not similar. This idea is about expropriation, while idea with dynamic commission is win/win for all parties.
About your math I still do not understand. Lets say there is 5% VP validator with 0 self stake but loyal delegators. So according to proposal this validator will pay 4.8% VP tax. If same validator launch 10 nodes with 0.5% VP each he will pay 0.3% tax in total. Compare 4.8% with 0.3%.
Incorrect, he would pay 10*(0.5%-0.17%)= 3.3%, and considering the costs of running 10 validators vs the costs of running one, this 3.3% is likely closer to 4%, so not so different.
The dynamic tax idea can also be called ‘expropriation’, since you are expropriating rewards from stakers to be given to large validators with a higher minimum commission. In the case of VP tax, instead of these rewards being earned by the largest validators, they are distributed across the whole set, this is to support the costs of many consumer chains and hence the growth of AEZ.
Can you please explain your math. Why does 0.33% become 3.3%?
1 validator scenario (5% VP)
VP Tax is 4.8%. Current APR is 22% so it will be 22% - 4.8% = 17.2% APR
1 validator scenario (0.5% VP)
VP Tax is 0.33%. Current APR is 22% so it will be 22% - 0.33% = 21.67% APR
10 validators scenario (10 x 0.5% VP)
VP Tax is 0.33%. Current APR is 22% so it will be 22% - 0.33% for each validator. 21.67% APR for each validator.
Dymamic commission cannot be called expropriation as every delegator can instantly redelegate their stake to any other validator with lower commission. In contrast, the VP Tax leaves no alternative for a validator except to lose rewards and delegators simultaneously without any other options.
Haha bro, it is not 10 validators scenario, it is 1 validator entity splitted in 10 smaller validators, so he pays the total combined tax of 10*(0.5%-0.17%)=3.3%, and again, now he has 10x the infra costs and more for the 10 validators, so considering this the tax would be closer to 4%, so splitting validators like this cannot really avoid the VP tax, even if the ‘loyal’ delegators you mention quickly redelegate to his 10 new validators
Also, the highest APR currently of validators at 0% fee is around 19.1%, from where do you get that 22% APR
In VP tax is similar, every delegator can instantly redelegate their stake to any other validator with less VP and hence no VP tax, and in doing this increase the decentralization of the Cosmos Hub.
The dynamic commission also leaves no alternative for a large validator, except to lose delegators. The difference is, in the case of this dynamic commision, the extra rewards from the higher minimum commission go to the large validators exclusively, in the case of the VP tax the combined extra rewards are equally distributed across all validators.
Seems like there are two potential centralization risks with a higher tax:
-
Becoming less competitive with CEX staking services which can bypass the tax entirely by being 100% self-bonded (this isn’t likely to be a problem, given Coinbase currently charges 35% on staked ATOM rewards)
-
Encouraging any validator with large amounts of delegated stake to attempt to take up multiple ‘validator’ slots, knocking real independent validators out of the active validator set
If you pay a 0.33% tax on $10 ten times, you’ve still only paid 0.33% on $100. The absolute value increased by ten, but the 0.33% tax percentage stayed the same.
And is there a reason a validator can’t cheaply spin up 9 extra ‘validators’ that don’t actually run a node, and just sign the same votes that the one real validating node does?
2 Likes
Why are you multiplying by 10? 0.33% from 100 ATOm is 0.33 ATOM. 10 x 0.33% from 10 ATOM is again 0.33 ATOM.
Regarding APR current APR with 0% fee is 22.2%. I recommend to use reliable sources.
Regarding VP Tax it is communism at its best. Choose wisely.
You can take a look at Polkadot type staking. Best examples are Moonbeam and Moonriver. Most of the small community validators were kicked out by big entities who can afford stake to get in active set.
For a long time I wondered why there is no mechanism in place that reduces rewards for the top validators to combat centralization of voting power. I like both the vote power tax and the quadratic voting idea. Though some have rightly pointed out loop holes that must be taken into consideration, what you proposed is a step in a good direction in my opinion!
Great post! Thank you. The VP tax seems like an appealing option. Here are a few thoughts:
-
I’m not entirely convinced about reducing the tax for self-staked ATOMs.
-
Without this reduction, the CEX can’t avoid the tax, as mentioned earlier.
-
Additionally, a validator will have a reason to re-delegate their self-staked position to another validator. This somewhat reduces the incentives for Sybil attacks (a validator may find redelagatng more profitable than the Sybil attack), but doesn’t eliminate them entirely.
-
If the VP tax also increases with self-stakes, Sybil incentives become even weaker. Validators creating new nodes with some of their self-staked ATOMs will still face a high tax (since the new node will basically have self-stakes).
-
Validators looking to re-delegate to another validator will do so if the commission fee with the new validator is lower than the additional tax. This would lead to more competition for lower commissions.
The VP tax is 0% if you’re one of the below-the-median validators. Where is the incentive to delegate to someone else (and pay a commission), when you can simply become several tiny validators? And what stops a CEX from doing a Sibyl attack and evading the tax, too?
Taxing self-stake would incentivize more validators to evade that tax, and validators who are mostly self-staked are the best-equipped to perform a Sibyl attack. It’s mildly difficult to co-ordinate a large chunk of your 3rd-party delegators to re-delegate to entirely new, smaller “validators,” but it’s trivially easy to split up your own ATOMs.
For some data, these are the commissions charged by some current custodial staking services:
There is no fee to stake or unstake. Coinbase takes a commission based on the rewards you receive from the network. Our standard commission is 35% for ADA, ATOM, DOT, SOL and XTZ (26.3% for eligible Coinbase One members) and 25% for ETH.
Binance.US deducts a 20% to 35% service fee from earned staking rewards, which is already reflected in the estimated Rewards Rate.
Our standard service fee is 20% for ETH; 30% for ADA, ATOM, DOT, SOL, and XTZ; and, up to 35% for other assets.
Kraken’s staking service seems to charge 15% on Ethereum staking rewards only, and 0% on all other staked assets, including ATOM. So Kraken already out-competes decentralized delegations right now, without any VP tax.
There are currently no fees for staking or unstaking. For Ethereum staking, due to the specific network rollout and complexity, Kraken retains an administrative fee from earned staking rewards and provides clients variable reward rate.
What is the Fee for Staking?
Kraken retains an administrative fee of 15% on all rewards received and the estimated RPY reflects this fee.
thanks for your response. Good points!
A Sybil attack has its cost. My point was just that, in some cases, a validator could be better by redelegating and paying the commission than assuming the cost of running new nodes. I agree that this would not be the case if you have to relegate too many ATOMs (in which case you would pay too much commission). That is why I said that there would be less incentives to do a Sybil attack (at least, for some validators, and comparing with the Syibil attacks incentives in the main post) but it does not kill these incentives.
Many participants mentioned that, with the scheme proposed in the post, a CEX could just self-stake and avoid the tax. If the tax is not reduced with self-staking, they cannot follow that strategy. However, as you said, they can still do a Sybil attack and split in many validators below the median validator.
Why do they need to pay this cost? They only need one real validating node, and can just add several extra signing keys for the extra validators, which all vote for the same blocks.
1 Like
Maybe I’m thinking too simple here, but aren’t the CEX validators known? Can’t we implement the vp tax in a way that it excludes staking as a service providers like CEXes from the self-stake vp tax reduction?
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.