1kx Tokenomics Proposal

30

Hi Mike,

The 1kx proposal maintains that the DAO (or RPL holders generally) should try to delegate more share to “protocol aligned NOs”. I asked how should the DAO determine “protocol aligned”? To summarize, your response was:

  1. “We can’t, but we can try (swiss cheese of several methods)…” ok, so what methods will you try?

  2. “already some options available (eg, Gitcoin passport)”… I don’t see how that would help, nothing stops a centralized entity from sock-puppeting this

  3. “They have more to gain by self-identifying”… Ok, so that is centralizing then (self-identified centralized entities benefit the most)? Or they don’t have more to gain by self-identifying, in which case they do have incentive to sock puppet…

  4. “Their ability to deceive pDAO into delegating to them depends on their ability to bypass the identification methods we implement”… What methods?

The entire 1kx proposal thesis (and concerns around RPIP49) are that a delegation system would be better since it benefits “decentralized/protocol aligned NOs”, but critical first steps have to be:

  1. Identifying those actors

  2. Filtering out pretenders/fake actors

Please provide some examples of how the 1kx proposal could implement those steps, beyond “swiss cheese of many methods” and “eg, Gitcoin passport”.

There is also the challenge that even if you accomplish those steps and it is somehow possible to know what RPL delegators “should do”, there is no way to “force” RPL holders to choose what they “should do” over financial incentives to do something on the contrary.

Skipping ahead to elsewhere in your response…

This is a non-issue. We can just make a channel for delegation requests and warn/ban anyone who requests delegation outside of this channel

It is more nuanced than that. Right now there is no incentive to sock puppet discord accounts and pretend to be multiple anons. If you instead incentivize this behavior as accounts try to “build social capital” for attracting delegations – you could cause more headache and noise than actual signal/substance in the discord. Centralized entities could attempt to build credibility through sock puppeting many anon accounts (all outside of actually requesting delegation in a “special channel”).

It seems your view is that, if centralised entities are >x% of the validator set, pDAO will activate the self-defence mechanism and begin reducing no_share until they leave

I never said “if centralized entities are >x% of the validator set”, I’m not sure where you got that idea. “Decentralization” isn’t necessarily: centralized entities are <x% of a set, I think an alternative method of quantifying decentralization is counting the total number of decentralized participants. Again, back to the best example of a decentralized network I can think of (Ethereum). Ethereum is estimated to only have ~6.5% solo stakers (Solo stakers: The backbone of Ethereum — Rated blog), but this “small percent” is a long tail of node operators that are critically important to censorship resistance and the overall health of the network. A single solo staker swings way above it’s weight class in it’s benefits to Ethereum compared to just 32 more ETH through Coinbase.

A simple example is that I think a network with 6.5% of it’s stake split among ~10,000 independent solo stakers (Ethereum - How Many Solo Stakers? — GLCstaked) out of ~14,000 validating nodes (https://www.theblock.co/post/285262/ethereum-one-million-validators), would be a much more robust “decentralized” network than a hypothetical network with 50% of it’s stake split among 50 independent solo staking nodes out of 70 validating nodes, even though 50% is > 6.5%.

I find it remarkable that if there are ~10,000 independent solo staking nodes on Ethereum, Rocket Pool accounts for ~1/5th of that node operator set, and ~1/7th of the total validating node operator set on Ethereum, even though Rocket Pool only makes up ~2.2% of staked ETH (https://dune.com/hildobby/eth2-staking).

Through the RPIP49 proposal, by lowering the barrier to entry we can hope to bring more home stakers online, which provides great benefits to Ethereum as more home staker nETH can lift even more staked pETH.

On the point about RPL, I disagree with it being used as a barometer for decentralization, but I’ve heard 1kx and NodeSet (previously) try to make that assertion. My point in the post you linked was that: if I did accept that assertion, (“ETH-Only = bad guys”, and “RPL = good guys”), under UARS we could lower no_share to zero, and all the revenue goes to “good guys” and none goes to “bad guys” (and “bad guys” have zero governance weighting). I disagree with that assertion though (the most Ethereum aligned guys – solo stakers – may prefer ETH-Only), so I don’t think that is what we should do. I responded further in my google sheet response to NodeSet, but their response of “coinbase has 90% of validators” doesn’t make sense to me… in the hypothetical example of lowering commission to zero and giving all revenue to staked RPL, why would coinbase continue to run RP validators if they provide no benefit over solo staking (and add more risk) – especially if there are other permissionless options available offering better yields such as Lido CSM?

Our proposal doesn’t claim or attempt to differentiate between centralized entities vs home stakers (since this currently seems impossible). If you meet the capital requirements, you may participate in the system (permissionless). The RPIP49 proposal doesn’t show partiality by subsidizing capital requirements (unlike 1kx proposal through in protocol delegations), so no one has any protocol supported advantages in the ability to grow their validator set (credibly neutral). The fear is that providing protocol supported advantages benefits centralized entities more than a neutral system (mostly due to the sybil resistance problem), and I still haven’t heard a response that addresses that concern.

As I described in my last response to you, the outcome of anyone performing this “attack” is increased validator count, increased rETH rewards, increased buy pressure for RPL, and increased RPL yield for pDAO. If you disagree with my response please say why and we can discuss it

The point was the same “centralized actors” you fear will “take over” the protocol under RPIP49 can still find themselves in the same (eth only) position in the 1kx proposal, but home stakers have less resources/ability to jump through these hoops. So yes: “increased rETH rewards, increased buy pressure for RPL, and increased RPL yield for pDAO”… But this is also accomplished by RPIP49 lowering no_share, which is much simpler and doesn’t include the friction of the 1kx proposal that acts as a barrier to home stakers. So at the end of the day the 1kx proposal seems net worse for centralization on this topic (just one example).

It seems you believe that once the tent is wide open, we will see more home stakers than centralized entities. If so, do you have data to support this hypothesis?

Again… I never said “more home stakers than centralized entities”, I don’t know where you keep getting this idea. We already discussed this over discord: Discord

As knoshua wrote: ‘you keep setting the standard at “guarantee that centralization is not possible" when the argument is that it’s more likely under 1kx’

If you are looking for a design where economies of scale have no impact, then I don’t think that is possible. The question then becomes: is a neutral system better than an opinionated system (through delegations)? My response is: a neutral system is better since any opinionated system I can think of only exacerbates centralization concerns. This is the fundamental question I ended my last response with, and I haven’t heard an answer from 1kx on it yet (that addresses the concerns).

Are you accounting for the fact that the first n validators do not require upfront collateral?

Could you propose an actual “n” value? If “n” is too small, it doesn’t provide much benefit to home stakers (too much friction still). If “n” is too big, the incentive to sock-puppet only grows higher. I don’t think you can find a meaningful “n”, and I’m not sure you’d ever be able to verify it’s effectiveness since it’s impossible to know when sock-puppets start. (back to my main concern).

[NOs setting their own lower commissions] is what will encourage competition between NOs, which helps rETH APY, and allows pDAO to provide incentives for protocol-aligned stakers.

My concern is it leads to a race to the bottom that prices out home stakers and accelerates centralization (subsidized validator set growth as well).

Responding to some other points you had, including your collapsible section

Extreme centralization risk of RPIP-49

Samus: I was surprised to read your post and find such a strong emphasis of “centralization concerns” around rpl.rehab.
Mike: Why? I have been very clear about my fears that UARS will lead to centralisation.

I was just pointing out the gap in time between your initial engagement: (June 17th) https://dao.rocketpool.net/t/tokenomics-rework-update-1-new-explainers/3014/9 …

And the first “centralization” concern you brought up (July 11thJuly 9th) that quickly became the highest priority for you to discuss by your second post (July 13th)… I appreciate the engagement, but the quick turnaround and emphasis on the topic surprised me so I wanted to focus on that topic in my response.

It seems you are mixing up parts 1 and 2 of our post. Part 1 describes why we believe it is the best direction for RP, and we remain optimistic about this approach. Part 2 of our post describes why we think RPIP-49 would ultimately lead to negative outcomes for RP. We believe RPIP-49 will lead to significant reduction of RPL price and the centralization of the validator set. So yes, we are very pessimistic about RPIP-49

The first 4/7 paragraphs of your original post on this thread focus on pessimism around RPIP49 rework, which is what I quoted in bullet points in my previous reply (I guess you wouldn’t call that “Part 1”?)…

We have already added some research to help quantify the “Will RPIP-49 lead to unstaking” question in the (steel man doc)[Steelman Arguments Against Tokenomics Rework - Google Docs]

I added a response to the doc, including a more thorough analysis of the excel sheet posted… There were some mistakes from 1kx due to outdated information, and generally I disagree with the logic/data presented.

You did not comment further in the thread, so I am not sure if Wander’s explanation of the risks changed your thinking at all here.

Wander posted as a reply in the main thread, and then linked to the other thread you mentioned. I responded briefly first in NodeSet’s (“I haven’t had time yet but I plan to read through the report and respond specifically to where I disagree. But I wanted to go ahead and respond to this since the question was also coming up in the discord…”), but in much more detail in the main thread:
https://dao.rocketpool.net/t/2024-tokenomics-rework-drafts/2847/41?u=samus

The same concern was also raised by other community members around the same time…which have not been addressed.

I included a thorough response to rocknet as well as NodeSet in the sheet I linked above (and never got a response from either parties)

It’s ok for people to not respond further, and it’s ok if you disagree with my responses, but it’s incorrect to say “we haven’t considered them”. One example was that these concerns helped lead to the Express Queue mechanism I proposed, to assist existing NOs and small NOs in ensuring they have the ability to migrate/join even with large ETH-Only NOs coming online and potentially limited rETH demand.

Did you ask yourself the same questions about RPIP-49? The answers also depend on multiple moving variables.

I think there was a misunderstanding with your Q&A…

My point was RPIP49 UARS variables are all Universal: all borrowed ETH revenue pays the same no_share commission, all borrowed ETH revenue pays the same to surplus share, all borrowed ETH revenue pays the same to voter share, all vote eligible RPL earns the same APY, and rETH commission is the sum of these Universal variables.

Under 1kx: every NO can have different commission, delegate_share/extra_reth_share, and recollateralisation_share (it may be activated on some and not others depending on your collateralization). None of these variables universally apply to all NOs. Similarly, under 1kx because RPL rewards are not socialized, the APY your RPL earns depends on the node you stake with (more RPL staked on the same node causes less rewards per RPL since it’s the same pie split among less RPL)

This is what I meant by tons of moving variables at the same time (every node is different, like different stakewise vaults), requiring awareness of pvp dynamics (unlike universal variables where this doesn’t exist).

Please clarify your intent with this statement, because it could be interpreted as a repeated attempt to insinuate our proposal goes against permissionlessness

You had suggested that 1kx proposal allows for “automatic” delegation to home stakers Discord as though there was some kind of “automatic method” for determining home stakers vs centralized actors. You then also suggested why not just “trust the pDAO” to make the best decision on who to delegate to (which subsidizes that delegates validator set growth) Discord

It was that context that I commented I don’t think we should trust anyone to pick winners and losers of who has an easier/harder time growing their validator set (lower cost of capital for the “winners” who don’t need to bring their own RPL). The permissionless ethos I was referencing is where everyone is treated equally just by bringing the same capital requirements, instead of the protocol picking and choosing. I understand with the 1kx proposal you could still “permissionlessly” bring more RPL capital, but different capital requirements for some vs others can lead to a similar effect as a permissioned system like Lido (permissioned entities have smaller capital requirements and can therefore grow their validator sets more quickly than home stakers can). Perhaps “credibly neutral” would have captured the idea better than “permissionless ethos”.

Do you mean to imply that a proposal which attempts to be benefit smaller NOs over large ones is “picking who is right and wrong for Rocket Pool”? You have expressed a desire to “subsidise the little guy”. RPIP-42 does too:

The problem again comes back to you have no way to know who is actually a small NO or who is large entity pretending to be a small NO. There is a limit to how much we can “subsidize” the little guy, since if you try to subsidize too much, then large NOs can just game the system. The bond curve was the best we could do (we take on more MEV risk for the little NO, but they still get to join and earn the same commission just on less borrowed ETH for the first two validators. Importantly, large NOs in this system don’t have any financial incentives to sock puppet). The express queue also benefits existing NOs and small NOs, but the benefit is marginal… There is no difference in capital requirements, just potentially shorter wait times in the queue. With the express queue it is also intentionally limited to 2 tickets (only enough for base validators), so this also helps remove financial incentives to sock puppet.

The 1kx proposal does not have these constraints on sock puppet incentives, so the attempts to “help the little guy” become ineffective at best (little guy blends in with big guy), and detrimental at worst (little guy has greater frictions/barriers while the big guy can more easily jump through hoops).

Questions about RPIP-49

  1. What criteria would you use to decide whether or not to activate the self-defence mechanism to prevent centralised takeover? i.e. how would you identify centralised entities in order to know whether they are >x% of the validator set?

As I described before, we don’t claim (and never have) to identify when centralized entities are >x% of the validator set. We will have to respond to supply/demand dynamics and if rETH demand consistently outpaces supply we may have to lower commission to balance out this dynamic. The surplus revenue could then either go to rETH or to RPL. I have general concerns about MVI compressing margins for solo/home stakers, and similarly if we lower commission too much this also compresses margins for solo/home stakers. But I don’t see how the alternative proposal under 1kx would provide any better solutions (back to the neutrality vs opinionated discussion, where opinionated seems to only amplify centralization concerns from my perspective).

  1. If you were unable…
  1. Do you have any data…

Skipping this question since the premise is misguided.

  1. Apart from the “reduce the no_share” self-defence mechanism, what provisions does RPIP-49 include to help reduce the likelihood of centralised takeover of the validator set?

The express queue should ensure that at least our entire current NOs, and small NOs have access to protocol ETH above: new, large NOs (which may or may not be centralized entities). The best defense we have is by attempting to attract the maximum number of decentralized participants. I think it is also important to compare the RPIP-49 rework to other options and ask:

  • Will RP be more decentralized under any alternative ideas/proposals? (I think the answer is no)
  • Will Ethereum be healthier and more decentralized with RP under RPIP-49? (I think the answer is yes)
  1. As a NO, how long would you remain with the protocol if no_share was set to 0%?

If the remaining 14% of revenue was going to voter share, then I would remain with the protocol indefinitely. (This premise is misguided though).

  1. It seems you think our proposal will harm permissionlessness. Under our proposal, from whom do you think NOs will need to request permission before launching a validator?

You could view “permissionless” as black and white (it is or it isn’t), in which case I would consider your proposal permissionless. If you instead consider it as a spectrum (how high are the requirements, and do they apply equally to all participants):

  • The most “permissioned system” has the highest requirements (meeting KYC requirements, proving institutional grade services/expertise, etc) – to filter out who to give “permission” to, and who to exclude.

  • The most “permissionless system” has the lowest requirements for everyone (Ethereum, Rocket Pool currently and under RPIP49) – everyone has the same capital requirements, if you bring the capital you may participate.

Then I would consider the 1kx proposal closer to “permissioned” than RPIP49 since the protocol may grant “permission” for some delegates to have much lower capital requirements compared to others - maybe those who identify themselves through Gitcoin Passport or other unspecified means (KYC? Interview with NodeSet – you mentioned this idea in the past? Some other “identification” mechanism?). Previously in this post you mentioned Rocket Pool should apply some identification mechanisms, but you didn’t say exactly what those would be (and yes, building a foundation of the base layer protocol on identification mechanisms does seem to go against a permissionless ethos).

  1. Related to my questions above, do you believe there is a difference between “no_share that is profitable based on the NO’s break-even point” and “no_share that Valdorff thinks each group will accept according to their level of Ethereum alignment”? If so, which do you think will be the primary motivating factor for the majority of NOs?

I do not claim to know motivations, or the no_share that the majority of NOs are willing to accept. I do agree with Valdorff that every NO might be different. Solo staking will still be an option after RPIP49 and I expect many NOs to continue to solo stake regardless of boosted commission RP provides, simply for ethos reasons. Similarly, node operators may continue to stake through Rocket Pool for ethos reasons, even if they would be less profitable than under a higher commission. Separately, I think we can listen to the market to broadly understand if we have balanced supply/demand or not.

  1. If UARS were to stabilise in the zone where no_share is profitable for centralised entities but nonprofitable for home stakers, and an increasing portion of the new validators were from centralised entities, what mechanisms in RPIP-49 would you use to attempt to reverse this trend?

Until it is possible to differentiate these actors I don’t see how it is possible to even know when or if this has happened. If it does become possible someday, then we can direct more rewards to the right actors (but until then, I think neutrality is the best we can do).

  1. Imagine a centralised entity launches a significant number of validators and pDAO decides to begin reducing no_share, as you described here. Onchain activity and social media show a large number of home staker types exiting the protocol, but the centralised entity has not left. In this scenario, which of the following do you personally believe would be the best course of action? a) continue reducing no_share, b) begin increasing no_share, or c) leave no_share unchanged

This premise is misguided, so responding within the constraints you put forward doesn’t make sense.

As I wrote in the link you described:

It sounds like the assumption is: “By keeping the RPL requirement, we can prevent centralization since centralized actors won’t buy RPL”
Even if we take that assumption as true for a minute, my response would be… (lower no_share, give to voter_share)

So under the assumptions in that scenario, only “good guys” can have RPL, so the incentives would reward “good guys” and not “bad guys”. However, I disagree with the entire assumption and premise in the first place.

4 Likes