Proposal: Unique shared validator commission rate, set through governance

Governance & Proposals
1

Disclaimer: I am part of the Setten validator team

Summary

There have been really interesting debates going on around how commissions affect the incentives for delegators and validators. Especially around the fight for the lowest commission rate and its negative effect on the overall validator ecosystem.

The main idea that came out is raising the minimum commission, materialized in Proposal: Minimum Validator Commission Rate - 5%.

Raising minimum commission rate is not a good solution
We don’t believe setting a minimum commission is a good long term solution.
If the minimum validator commission is arbitrarily set to 5% but the average validator needs a commission at 10% to break even, the problem remain the same.
In that case it’s even more beneficial than beforel to stay at the lowest commission rate, while it was at a loss previously.

With this proposal draft, I’d like to submit a different and broader approach:

Our proposal
Remove per validator commission settings
Have a unique global commission rate applied to all the validators.
This global commission rate would be set via conventional governance proposals.

Motivation

Staking game theory should be designed to encourage the stakers to delegate to the actors most beneficial to the chain security, decentralization and uptime.

The commission parameter has been since the start of Terra having the opposite effect. Just by proposing a 0% commission rate, delegators with no track records and/or poor performance could easily get delegations and high voting power.
It went to the point where some projects promised to exclude the biggest validators from their airdrops to try to rebalance delegations/

Stakers, looking for the best returns, will often stop their selection process at the validators with the lowest commission. Often not even considering historical performance or ecosystem contributions.

By completely removing the commission parameter from the equation, it opens again the possibility for validators to stand out for their skills and participation in the collective effort.

The main demarcation between validators would be their historical signing performance on the blockchain and how helpful they are to the overall community.

Pros

  • More decentralization by avoiding concentration of votes on low commission validators
  • Stop the unhealthy race to the lowest commission for validators
  • Easier for validators to break even
  • Encourage validators to stand out by means that beneficiate the ecosystem (improved infra, ecosystem contributions, etc)
  • Put emphasis back on performances
  • Removes malicious actors with “zombie” like behaviour
  • Lower the financial capital required to start a validator

Cons

  • Require some changes in Terra Core
  • The off-chain process to get an agreed upon rate and come up with the proposal might be a bit heavy.

Proposal

To implement this change, the following changes would need to be made to Terra Core:

  • Change the staking module to have a global shared commission rate
  • Change the governance module to add a staking commission rate change proposal

Timeline is to be defined (how to decide on the initial commission rate, etc).

FAQ

How about projects who use their staking commission for services (Stake to Subscribe, Charity, etc)?

Running a validator securely is a unique and separate skill set. There are many solutions available for these kinds of projects to raise money from their community.
The most similar way would be to create a staking contract that redelegates to trusted validators and take their commission on top.

How about validators who have more costly infra than the average?

Usually, those extra costs come from having a more specialized/advanced infrastructure and should translate directly into better performance in the long term.
With the commission factor being removed from the equation, this kind of validator should be getting more attention and thus more delegation and commission rewards.

9 Likes
Proposal: Minimum Validator Commission Rate - 5%
2

This is well reasoned and argued. I quite like this approach, also forces validators to focus on whats important: the sustainability of the network as a whole rather than profit or voting power driven motivations.

5 Likes
3

Heh… novel approach. By and large, I really like it.

However, after implementating this update, I believe we need to reset any re/delegation restrictions, to allow anyone who staked to a 0% validator be able to adjust their delegations accordingly (avoiding any situation where people feel it was a “bait and switch”).

Note: This kinda throws a spanner in the works for Angel Protocol, but the solution here might just be for other Validators to also pledge support for Angel Protocol though the “Angel Alliance” (similar to how projects do). Some mechanism to allow Validators to automate this Angel contribution would be amazing (and create a very special USP for Terra in the process), but that’d be a much larger change to the code.

@ChaunceyStJohn - think that’d work mate? :grin:

4 Likes
4

Elegant solution that would go a long way to preventing validators gaming the system.

This would help delegators become more informed about what validators actually offer.

6 Likes
Proposal: Minimum Validator Commission Rate - 5%
5

Would the global commission rate be able to be changed via governance? If so, then I’m okay with that. But a hard static cap without any ability to change it could cause problems long term.

2 Likes
6

Well stated, interesting solution with a floating standard across all validators.

2 Likes
7

Hello @moonplatoon
Indeed! As stated in the OP, it would need to be controllable by governance.

3 Likes
8

apologies, missed that part.

1 Like
9

Thanks for the feedback!

That makes sense, but also redelegation cooldown is like 21 days or something right? Maybe it would be ok to not over-complicate the implementation for such a “short” delay. If it’s trivial to do, why not tho.

As described in:

Projects like Angel Protocol could make a staking (or even liquid staking) smart contract that delegates on a set of trusted/partner validators.
I understand they’d only get (100 - [commission rate])% commission, but that would still be pretty high given a commission rate of 5% for example.

2 Likes
10

Where would you set the fixed rate at, knowing the costs involved? For example if some break even at 5% while others it will take 10%.

12

Commission alone doesn’t decide how much money is made, it’s commission + delegated stake.

So I’d fall back to:

Regarding how to decide on a rate, that’s an interesting question. That would mostly be left to whoever wanna put a proposal out there. So mainly off-chain consensus before bringing it to on chain.

I recon the process could be a bit heavy weighted. I’ll add that to the cons.

3 Likes
13

I would support this proposal. We validate in some chains with fixed commission rate and it is great to not see any discussions regarding the commissions, which does not offer any value to the chain, just validator economics.
Also, controlling this commission via governance I think is a great idea.

But a similar proposal was launched a long time ago in Terra Classic and it went nowhere. I think it will be hard that most validators agree on this.

2 Likes
14

So do you have a number in mind for this common shared commission ?

On a side note, I feel like this proposal will take time to be implemented and acted upon, not to mention the number needs to be decided first which will again take some discussions and time to get to a consensus.

Which is why I would prefer a floor commission proposal than the global one, considering the time crunch

1 Like
15

Looks like people prefer 10% so far (discussion from discord)

1 Like
16

Will update here once I get an answer from Yun on the feasibility in a short timeline.

When it comes to the %, maybe we could come up with a number quicker than expected. Seems like, as you said, some values are already discussed

17

like both proposals. My main point - avoid <5% commissions. it cause to lose delegants.
We are running IBC relayers. 1 transaction fee is about 0.05 - 0.3 Luna. It doesn’t matter how much Luna costs - if we couldn’t earn commission, we couldn’t support IBC. And I am not talking about infra costs.

3 Likes
18

Well this proposal was ignored by tfl so I guess we’ll go with min comm, which is ok for me too

19

From what I understood from the Discord discussion, this proposal was more complicated to implement - so the Dev feedback was to start with the minimum, then circle back to this when things calm down a bit… :blush:

I really like this idea, but if putting it on the backburner for a month or two is what we gotta do, I guess it just is what it is :person_shrugging:

1 Like