Kusama Validator minimum commission rate

Staking Admin
11 Comments
Executed
  • Content
  • AI Summary
Reply
Up
Share
Status
Decision14d
Confirmation3hrs
Attempts
1
Tally
94.5%Aye
52.8%Threshold
5.5%Nay
Aye
156.99KKSM
Nay
9.2KKSM
  • 0.0%
  • 0.0%
  • 0.0%

Threshold

Support(0.58%)
73.5KKSM
Issuance
12.6MKSM
Votes
Nested
Flattened
Calls
  • Call
  • Metadata
  • Timeline6
  • Votes Bubble
  • Statistics
    New
Comments
Sort by
Oldest

As we already posted in the discussion round, we are against this proposal and repeat our thoughts here again for people who did not see our reasoning yet.

There are two distinct issues at hand here: to have a minimum commission at all, and to raise this minimum commission from 3% to 10%.

The first discussion point for us is whether there should be a minimum commission at all or not. We have previously voted against this, and will continue to oppose it. The whole crypto ethos builds on the foundation that network-native currencies can provide sufficient incentives to run a globally distributed, decentralized network. These economic incentive structures should be robust enough to secure a stable network even in the face of well-capitalized and motivated adversaries. To ensure that a true game-theoretical equilibrium can be found by forces of the free market, interventions should be limited to the minimum. If in this intervention-free case, the true game-theoretical equilibrium leads to an end state that is not desirable and unintended by the protocol’s developers and community, then the protocol rules for incentivization should be changed, rather than introducing a subsidy of some kind.

Establishing a minimum commission does not allow for free market competition, or at least makes it harder as the minimum commission can still be circumvented on- or off-chain by means that involve some trust between the parties (operator of the validator and nominator). This is certainly not helpful for having a transparent commissions market. The minimum commission also guarantees a minimum rent that can be extracted by validator operators, which is opposite to the crypto ethos.

The second discussion point is raising the minimum commission from 3% to 10%. This would exacerbate the issues outlined above (more interference, higher rent to be extracted, etc.), and on top of it attempts to play “central bank” to provide a “soft landing” during worsening market conditions.

Now, we have only outlined what not to do so far, so here is what we think would lead to a better commissions market and actually fix the underlying issue rather than continuously attempting to treat the symptom of some validator operators being unprofitable despite providing good services. To do that, research into how users select validators and how to assist and improve this process is required (such as demonstrated in the recent talk by J. Gehrlein, see: https://www.youtube.com/watch?v=ym8t6cVHhgY and “An active preference learning approach to aid the selection of validators in blockchain environments”, Gehrlein et al., 2022). We would support further efficient efforts in that direction, from research through to deployment so that users can better balance out validator selection criteria rather than placing a perhaps irrationally high importance on commission only.

Reply
Up 1

There are so many reasons why we are opposed to this proposal. But I will address just two of them.

  1. It discriminates against inactive validators. Active validators will get more rewards and more exposure to nominators, as opposed to inactive validators who will only get losses. Ultimately, this will lead to an even more centralized network. A fairer solution would be to simply give away KSM from the treasury to cover the costs for any validator, even if they are not in the active set.

  2. This is basically risk shifting. Nominators are forced to cover losses caused by validators' bad business strategy. At the same time, validators will retain huge levels of profits in the bull market. This is the same thing as the bailout of banks with taxpayers' money.

Reply
Up

test123

Reply
Up

test1111

Reply
Up

test1234

Reply
Up