Introduction In my last post , I detailed the current benefits and shortcomings of the Cosmos Hub’s Replicated Security (RS) framework. RS faces a unique and complex challenge that can simply be referred to as the RS Incentive Problem. This trilemma encapsulates the divergent, often conflicting interests of three primary stakeholders: ATOM stakers via the Cosmos Hub, consumer chains, and validators. Each group operates with distinct self-motivated interests, creating a nuanced relationship. The interwoven dependencies between stakeholders require careful balancing for the successful implementation of shared security models like Interchain Security (ICS) and Replicated Security (RS). • ATOM Stakers via the Cosmos Hub: ATOM stakers, driven by the pursuit of higher yields, naturally favor an increase in consumer chains. Their preference to extend economic security stems from a perceived lack of accountability - they do not bear any direct costs when voting to support additional networks. Their inclination is to consistently vote in favor of integrating more consumer chains, viewing each addition as an opportunity to increase their yield without incurring proportional costs.…

Excerpt (1199 of 36389 characters). Read the whole post on the forum ↗

VelvetMilkman1:

ATOM Stakers via the Cosmos Hub: ATOM stakers, driven by the pursuit of higher yields, naturally favor an increase in consumer chains. Their preference to extend economic security stems from a perceived lack of accountability - they do not bear any direct costs when voting to support additional networks. Their inclination is to consistently vote in favor of integrating more consumer chains, viewing each addition as an opportunity to increase their yield without incurring proportional costs.

what are you basing this assumption on? Validators have voted to include all of the poorly performing ICS chains thus far, despite their validator’s continued efforts to prop up stride by throwing as much of the CP as they could at it. Validators have voted to take on the added cost of loosing consumer chains, not stakers.

Thanks for reading. It’s based on some heuristics, my general sense from internal conversations, and some of these figures:

Launch Neutron Top 20 Validator voting results (by stake weight):

• Yes (10 vals) = 29.67% of total delegations

• Abstain (2 vals) = 5.81% of total delegations

• Didn’t Vote (8 vals) = 25.9% of total delegations

• Total Non-votes = 31.71% of total delegations

• Token Holder Vote (Yes : No/Abstain/Veto) = 99.68% : 0.32%

Launch Stride Top 20 Validator results (by stake weight):

• Yes (10 vals) = 31.2%

• Abstain (1 val) = 1.76%

• Didn’t vote (9 vals) = 28.42%

• Total Non-votes = 30.18%

• Token Holder Vote (Yes : No/Abstain/Veto) = 99.68% : 0.31%

how can you tell that it was token holders voting vs validators voting on token holder’s behalf and delegators not overriding their vote with their own or not affirming the same way their validator did with their own vote?

no, it is a flawed assumption. I know for a fact those stats consists of a large number of bots with almost nothing staked. you can look at the token vs account trends that the number of accounts topped out around the 17th and the tokens voting roughly doubled after that. so the votes appear to be based on validator vote rather than delegators voting.

It appears that validators volunteered to take on the added expense of an unprofitable chain. rather than subsidize an unprofitable venture and force users to pay validators for things users obviously dont use, why not secure profitable chains or something crazy like that?

I removed my message before I saw your response otherwise I would have left it.

I don’t think we’re in disagreement.

In an earlier version of my draft, I propose validators don’t have the ability to influence governance with delegated stake. In my opinion, all validators should Abstain from all Consumer Chain Proposals and only vote with their own tokens.

If this entire post was adopted, validators should not be able to influence the vote one way or the other beyond the stake own. They are being paid to serve the community and the community should decide the best path forward without this type of influence.

The assumption is that most delegators want to see consumer chains and shared security succeed and are less cost-sensitive to OpEx given the current model. For the record, I also want to see shared security succeed.

why make a minimum commission and not just allow different commission rates 0-100 rather than 5-100? it would be more competitive if the whole range no minimum were variable on all ICS chains. with a cumulative APR it gives validators greater flexibility to create a competitive business strategy based on their ICS chain preferences and remain overall competitive.

I explain why in the post.

VelvetMilkman1:

The assumption is that most delegators want to see consumer chains and shared security succeed and are less cost-sensitive to OpEx given the current model. For the record, I also want to see shared security succeed.

I dont think this is a fair assumption. My assumption is that the promise of a bull run is the only thing propping ATOM up

You’re asking for a race to zero or you’re asking for full optionality on Day 1, which is something I’m trying to avoid. It simply won’t be successful without training wheels, especially early days. I’m trying to build a sustainable model where everyone gives and everyone takes, on the margins.

Without opt-in/out (1.3), the only real competitive strategy available for operators is to reduce commissions (assuming anyone is even paying attention).

Once opt-in and opt-out are available, validators can actually differentiate and offer ATOM holders more optionality, greater yields, etc.

how is it a race to zero? it seems far more like a race for skin in the game. validators with more self stake can set their commission lower.

Validator costs are some fixed number (electricity + equipment + salary for support). Obviously they are different from place to place, but they can be averaged. Let’s for example’s sake use $1000 per month per validator. The revenue that validators get depends on the commission. Setting min commission rate to 5% would hand large validators a windfall. Validator A has $10 million staked, Validator B has $1 million staked. 5% commission would mean, Validator A has $500,000 of revenue vs Validator B who has $50,000 revenue. Both of these are well over the $12,000 per annum needed to run the computers. As such commission rates could easily be 2% and both validators will be operating quite comfortably. I don’t see how setting min commission rate of 5% makes any sense. This is handing the Allnodes, Binance and Coinbase validators free money. Actually Coinbase already charges an arm and a leg rate of 20%. More to the point, if you want to ensure validators don’t engage in race to the bottom on costs (although I see no problem with constant efficiency improvements, in fact they are required for anyone running a business), I think you should set some fixed dollar budget under which…

Excerpt (1198 of 2705 characters). Read the whole post on the forum ↗

Nice work, really appreciate the read. Also, agree with vixcontango on the fixed dollar budget based on the validator size. Seems like a good compromise for both parties.

Edit : Where are we with the Power voting Tax, Dynamic LS Tax and quadratic voting proposals?

vixcontango: I think the issue with the validator gripes for the most part are the large inequities in income generated between big validators and small validators. Small validators feel like they are lone warriors and not making much while the big validators like Coinbase are making a killing. Personally, I feel like setting a minimum commission does places small operators on a semi-even playing field with large, well-funded operators. Smaller validators are generally the ones that can’t compete against larger operations that can charge less. There are other factors at play as well. • Does it make sense for a company that charges 0% commissions to control 10% of the voting power? • Does it make competition a race to zero? (I would argue it does) • How do we expect small validators to run additional hardware for consumer chains, if they are running at a loss on the Host Chain? • What does the Cosmos Hub community value and want to promote when seeking a distributed validator set? I have my own opinions that I’ve shared in the post. This has to be a consensus decision. I personally don’t care what the minimum commission rate is set to, but I do see the benefit…

Excerpt (1197 of 2810 characters). Read the whole post on the forum ↗

Thanks for the read. Those are great questions that I don’t have the answer to but am interested to look more deeply into.

I think there is ample space to explore on how best to align the interests of every party in the equation. I do think more analysis is necessary. Putting a team in place to review and negotiate SLAs between consumer chains and host chains would definitely help give the Cosmos Hub a better negotiating position and I think Partial Set Security will also lead to very interesting outcomes.

We really have only scratched the surface and I’m open to exploring many alternatives. DMs open if you want to jam!

I personally don’t care what the minimum commission rate is set to, but I do see the benefit of setting one. Ideally, a minimum commission rate offers sustainable cashflow to all validators in the active set. This allows them to compete on other metrics besides “I have bigger bags.” If a validator with bigger bags is forced to compete at the same minimum commission as others that have smaller operations but better performance, why would anyone stake to the larger validator? There are obviously reasons why but in that case, it’s a personal decision.

Allowing validators with more self stake to reduce commissions to 0% ultimately is a race to zero. It squeezes out the smaller operators and gives greater political influence to those with the biggest bags. This flywheel inevitably increases delegations, which increases self-stake, which reduces commissions, which further increases delegations, which further increases self-stake, which further reduces commissions, etc.

This is the definition of a race to zero, where only the largest token holders can compete and smaller operators are squeezed out.

First and foremost, we’d like to commend @VelvetMilkman1 for the extensive effort put into crafting this post. The post is well-thought-out, well-written, and demonstrates a commendable focus on proposing thoughtful improvements, providing detailed rationale behind them. We appreciate this commitment to high-quality content and hope it sets a standard for the Hub. Here, we offer our feedback on several key points: 1.1 Minimum Commission Fee for Cosmos Hub Validators The 5% minimum validator fee has already been approved through an on-chain vote (referenced here: Mintscan ). We align with the reasons presented and supported this proposal during the on-chain vote. 1.2 Dynamic Commission Adjustment with Consumer Chain Integration • Incremental Minimum Commission Increase/Decrease: Your straightforward framework for addressing validator compensation complexities provides a solid foundation for discussion. Simplicity can be a baseline for further iteration, striking a balance between complexity and efficiency. • Interim Period: While supporting the proposition, we recommend redirecting early-stage fees to the community pool rather than stakers. This would enhance…

Excerpt (1188 of 4694 characters). Read the whole post on the forum ↗

I would argue that larger validators don’t set their commissions less than the average on a chain by chan basis which (from what I’v seen), hovers around 5% anyway. Just because they are larger doesn’t mean that they will automatically select an abnormally smaller commission rate. My point, I like the plan and I don’t se how setting a minimum harms stakes all that much. Most wont even notice it.

setting a 5% minimum commission reduces validators ability to compete within the set & reduces the value proposition for staker’s.

nothing is currently stopping validators from setting 5% as their commission rate other than being forced to like these proposed changes aim to do.

all this proposal seems to do is reduce competition for delegations among validators and reduce the appeal of staking ATOM.

Wouldn’t you say that as a result it forces decentralization?

no, why would a 5% minimum commission help decentralization when most validators already have at least 5% commission?

something like making the top 15 validators subsidize the rest of the lower validators with <1% of their rewards until your choice of decentralization KPI is met would increase decentralization by making the bottom of the set more competitive with their commission rates.

you said: common_spelling: setting a 5% minimum commission reduces validators ability to compete within the set How does this reduce validators ability to compete, other than forcing stakes to choose at random now that the base commission is 5% across the board? common_spelling: & reduces the value proposition for staker’s. again, I dont think the incentive of +/- 1-2% is a windfall for most takers. I’d even say that most expect to pay 5% as that is the average across the board from what I’ve seen. common_spelling: nothing is currently stopping validators from setting 5% as their commission rate other than being forced to like these proposed changes aim to do. This is why I said that really nothing changes. common_spelling: all this proposal seems to do is reduce competition for delegations among validators and reduce the appeal of staking ATOM. If I understand what yo mean by “competition”, the degree to which this changes would affect said competition and appeal for staking (even less so) is minimal IMO. Maybe you are a validator and have a purview I don’t as fas as validators but as a staker of ATOM…

Excerpt (1197 of 1235 characters). Read the whole post on the forum ↗

peopleschamp: common_spelling: setting a 5% minimum commission reduces validators ability to compete within the set How does this reduce validators ability to compete, other than forcing stakes to choose at random now that the base commission is 5% across the board? it eliminates a validator’s ability to differentiate itself within a metric that delegator’s care about peopleschamp: common_spelling: & reduces the value proposition for staker’s. again, I dont think the incentive of +/- 1-2% is a windfall for most takers. I’d even say that most expect to pay 5% as that is the average across the board from what I’ve seen. then why force a 5% minimum if its irrelevant to where delegator’s stake peopleschamp: common_spelling: nothing is currently stopping validators from setting 5% as their commission rate other than being forced to like these proposed changes aim to do. This is why I said that really nothing changes. you just said that 1-2% reduction in rewards to stakers…while not a change for validators, it is a change for the value proposition of delegators. it not being significant to…

Excerpt (1199 of 1962 characters). Read the whole post on the forum ↗

common_spelling: then why force a 5% minimum if its irrelevant to where delegator’s stake I’m saying it doesn’t matter either way but I the spirit of the OPs proposal I thin its a move in the right direction of the ecosystem as a whole. If you don’t think so, create another proposal with your proposed solution. common_spelling: you just said that 1-2% reduction in rewards to stakers…while not a change for validators, it is a change for the value proposition of delegators. it not being significant to you is immaterial. I said plus or minus, thats different than just a reduction as some will go up, some will go down. I said this is just my opinion, so whats immaterial is you opinion, and can be said to be immaterial as well. common_spelling: I mean one of the few ways validators can differentiate themselves and attract delegators other than their validator pfp and name. Some have websites with mission statements, specs for hardware etc. I wouldn’t say the pop is the end all, at least not for those who really care. common_spelling: You do see how it hurts, you have literally said you think the amount it hurts is irrelevant…

Excerpt (1198 of 1737 characters). Read the whole post on the forum ↗

peopleschamp: common_spelling: you just said that 1-2% reduction in rewards to stakers…while not a change for validators, it is a change for the value proposition of delegators. it not being significant to you is immaterial. I said plus or minus, thats different than just a reduction as some will go up, some will go down. I said this is just my opinion, so whats immaterial is you opinion, and can be said to be immaterial as well. what? peopleschamp: common_spelling: I mean one of the few ways validators can differentiate themselves and attract delegators other than their validator pfp and name. Some have websites with mission statements, specs for hardware etc. I wouldn’t say the pop is the end all, at least not for those who really care. only nerds visit validator websites or care about their hardware. I am talking about regular users, simple people that just want to gamble their fiat. peopleschamp: common_spelling: You do see how it hurts, you have literally said you think the amount it hurts is irrelevant to you…How does it help? You can fault me for having an opinion but your picking…

Excerpt (1197 of 2077 characters). Read the whole post on the forum ↗

common_spelling:

only nerds visit validator websites or care about their hardware. I am talking about regular users, simple people that just want to gamble their fiat.

Insults get you nowhere and doesn’t negate my point.

common_spelling:

only nerds visit validator websites or care about their hardware. I am talking about regular users, simple people that just want to gamble their fiat.

I don’t get your point. If all in all nothing really changes, how does a 5% minimum hurt???

common_spelling:

why would a 5% minimum commission help decentralization when most validators already have at least 5% commission?

something like making the top 15 validators subsidize the rest of the lower validators with <1% of their rewards until your choice of decentralization KPI is met would increase decentralization by making the bottom of the set more competitive with their commission rates.

…this was my point

Just my opinion, but I think small validators should try to gather delegations from other sources than lowering to 0% their commission just to attract new users, then increase later on without warnings. They should attract more delegators by offering value (BD, working with prospect protocols and chains, growing their community, be informative,…etc).

I don’t think either that big validators should be able to lower their commission just to attract even more voting power (definitely not improving decentralization). Also, it seems that we have issue running consumer chains because a part of the set operates at a loss… maybe they wouldn’t if they weren’t at 0% commission?

Tagu:

They should attract more delegators by offering value (BD, working with prospect protocols and chains, growing their community, be informative,…etc).

isn’t that the justification that validators give before voting to pay themselves from the community pool? Cryptocito, informal, notional, CosmicValidator, etc…*which is a reason that 5% minimum because you hope it will just inspire validators to do work for free is silly.

This is just one parameter, we could have both a healthy minimum commission rate AND work towards validators adding more values with other parameters (PSS, Power voting Tax, governance, building tools … etc).

Tbh, 0% commission makes no sense to me and I’m only a user/staker. Maybe 5% is too high and some mechanics could be applied to allow some validators to go lower IF their reach some criteria.

If that’s your point, I won’t disagree but it’s a different one than the original which was how the minimum hurts.

The minimum does only hurt. You say its justifiable harm because it doesnt hurt staker’s significantly and seem to be unaware the level to which 5% minimum reduces competition amongst validators. so, while you admit there is a non-zero level of harm that results from a 5% minimum, you provide no reason to think it would help.

Tagu:

This is just one parameter, we could have both a healthy minimum commission rate AND work towards validators adding more values with other parameters (PSS, Power voting Tax, governance, building tools … etc).

Tbh, 0% commission makes no sense to me and I’m only a user/staker. Maybe 5% is too high and some mechanics could be applied to allow some validators to go lower IF their reach some criteria.

5% minimum commission on other chains has done nothing to support the claim that implementing one does anything beneficial for decentralization nor validators adding value to the chain or whatever. there is no “race to 0% commission” 10 validators out of 180 validator’s have 0% commission.

The benefit has been laid by OP. Oppositely you’ve only seem to muster a reason why not, nothing of benefit, seems to m you have an outsized reason not to and thats or only concern. I haven’t seen anyone else here so fervent about reason wh not.

“how is it a race to zero? it seems far more like a race for skin in the game. validators with more self stake can set their commission lower.”

This exacerbates centralization of voting power and is terrible for network security. Besides, 5% minimum commission waa already approved in proposal 826 and goes into effect with the next upgrade, scheduled for February. Link below.

Mintscan

Interchain explorer and analytics powered by Cosmostation.

Really well thought out. Thank you. As a smaller validator who needs to weigh costs and time involved with supporting consumer chains, we love to see this.

Fyi, 5% minimum commission was approved last year and from what Informal Systems has said, goes into effect with the next upgrade. Link to proposal 826 below:

Mintscan

Interchain explorer and analytics powered by Cosmostation.

BlocksUnited:

“how is it a race to zero? it seems far more like a race for skin in the game. validators with more self stake can set their commission lower.”

This exacerbates centralization of voting power and is terrible for network security. Besides, 5% minimum commission waa already approved in proposal 826 and goes into effect with the next upgrade, scheduled for February. Link below.

no, validators being able to use aggregate votes of their delegators exacerbates centralization of voting power and is terrible for network security.

Yes, validators have voted to arbitrarily raise the minimum commission and pay themselves more for no rhyme or reason. That validators can vote with delegator’s votes to pay themselves more for no additional work illustrates the problem.