PUP-13: Initial WAGMI Parameters

Attributes

  • Author(s): @adam
  • Parameter: RelaysToTokensMultiplier
  • Current Value: .01
  • New Value: Variable, decreasing over time
  • Related Proposal: PUP-11: WAGMI Inflation

Summary/Abstract

The WAGMI proposal has garnered a significant amount of attention and support over the last month and a half as a mechanism for controlling inflation and providing sufficient rewards to nodes - in all scenarios.

I will not rehash what WAGMI is or how it works in this proposal, for that, please view that proposal directly. This proposal will govern the first several parameter changes of WAGMI if both are approved by the DAO. Below, I will attempt to detail a fair and simple means to gradually reduce token incentives over time.

If you haven’t read WAGMI, please don’t comment. If you are unfamiliar with the current economics, please read this: Economics - Pocket Network (all of it).

Motivation

Node runners, new and old, are seeking security in token incentives so they can plan their node operations.

Many are concerned about a death spiral that a single, significant reduction would bring. Further, many have risked much to stand up their new node businesses and we don’t want to cut them off at the knee.

Rationale

This plan provides for a gradual step down to token incentives to the RelaysToTokensMultiplier via WAGMI to accommodate all who are part of the Pocket ecosystem.

I propose the specific WAGMI rates to be:

  • Upon passing: Target Inflation Rate = 100.00%
  • Passing +30 days: WAGMI Target Inflation Rate = 90.00%
  • Passing +60 days: WAGMI Target Inflation Rate = 80.00%
  • Passing +90 days: WAGMI Target Inflation Rate = 70.00%
  • Passing +120 days: WAGMI Target Inflation Rate = 60.00%
  • Passing +150 days: WAGMI Target Inflation Rate = 50.00%

All else equal this would look like the following for per node rewards:

  • Upon passing: Target Inflation Rate = 84.85 POKT/Day
  • +30 days: Target Inflation Rate = 76.36 POKT/Day
  • +60 days: Target Inflation Rate = 67.88 POKT/Day
  • +90 days: Target Inflation Rate = 59.39 POKT/Day
  • +120 days: Target Inflation Rate = 50.91 POKT/Day
  • +150 days: Target Inflation Rate = 42.42 POKT/Day (or 15,486 POKT/year)

This would address a vast majority of the concerns raised by commenters on the last two proposals (PUP-11/PUP-12). After 150 days, I propose that we re-evaluate the landscape through a new proposal. For the avoidance of doubt, the WAGMI Target Inflation Rate of 50.00% would stay in place until a new proposal replaces that value.

Dissenting Opinions

TBD

Analyst(s)

@adam - Wizard of WAGMI, Stirrer of the Pot, Least Popular Proposer, number one source of economics related angst

This proposal is my opinion and does not represent the opinion or position of Pocket Network, Inc., Pocket Network Foundation, or the Pocket DAO.

Copyright

Copyright and related rights waived via CC0.

9 Likes

I think these are reasonable targets for inflation control, and address the many concerns that have been brought up along the way.

I support this proposal without reservation.

7 Likes

This is a great proposal and I support it wholeheartedly. It puts Pocket Network on the right path and I’m more optimistic than ever about the future. I staked my first fistful of POKT today after reading this proposal.

1 Like

Seems reasonable.

Let us also give this proposal some extra visibility on the official Twitter and beyond so everyone is prepared and the node market started to react.

4 Likes

I support this measured, gradual approach.

Maybe I’m missing something but wouldn’t Target Inflation Rate of 84.85/day per node = 30,970 = 200% per node?

Seems that all the inflation %s need to be doubled.

Good call. This proposal has been shared on the DAO Twitter account and further updates will be shared there. https://twitter.com/POKTdao/status/1490806301252894730?s=20&t=FZ46vzmSDWnZtkK9t-iZcg

1 Like

We need to be careful about confusing annual inflation with node returns. WAGMI rate is an annual network-wide inflation target and doesn’t directly translate to per node income.

WAGMI % * total supply / # of nodes = annualized per node rewards.

2 Likes

Could you also share to the main Pocket Network Twitter account? There may be others like me that are interested but didn’t realise there was a Dao Twitter account. It would allow for wider community feedback given 30k followers compared to 1k for the Dao account.

It has already been retweeted by the official account :+1:

2 Likes

I think this is too aggressive a cut to rewards unless it is balanced by an increase in relays. By limiting the inflation so severely we decrease the amount of tokens that are available for use by the network to stake relay nodes and new DAPPs.

I also don’t want to support this while inflation is being expressed as a number of POKT/node/day. It should be expressed as a multiplier (.01 etc.) per relay.

Put another way, the current Target Inflation Rate limits the network to a doubling of relay nodes each year.

The question I have for the group is this: Are we going to be okay if in 1 year we have only 50k nodes? And 100k nodes the next year?

Would love to see some numbers around this as I’d like to see if I’m missing something.

4 Likes

I think if we end up in the fortunate situation where there’s so much relay demand that there’s not enough POKT being minted for the amount of worker nodes needed, the team could just consider reducing the amount of POKT needed to stake a node. That would immediately allow existing node runners to expand their operations and meet demand.

Such a situation would also drive up the price of POKT significantly, so a reduction in minimum node stake would make sense , otherwise staking a node would become prohibitively expensive.

1 Like

IMO this would be a disaster. If the limits were lowered you would immediately have much of the network unstake to take advantage of the lower limits, potentially taking the network offline for 3 weeks. Token price would tank as the market is flooded with new inventory. The volatility of such a move is way less desirable than simply ramping down inflation in a less drastic manner than what is proposed.

IMO two things are important to the stability of the network:

  • Inflation only ever goes down.
  • The number of tokens needed to stake a node remains consistent.

Both are crucial for maintaining credibility for the project.

Better to manage this risk by ensuring sufficient token supply for effective network operation.

1 Like

I think that’s a highly pessimistic outlook. If you reduce the minimum stake by 5% you’d need to take 20 nodes down to spin up an additional one, which would take years to pay off in lost revenue. If you reduce the min stake by 5% every couple of months you could ensure an orderly transition. I think the community is also smart enough not to commit mass suicide.

1 Like

A 5-month ramp down is drastic? What would be less drastic in your eyes?

POKT inflation is already expressed on-chain as a multiplier per relay. Adam is just abstracting this into network-wide inflation figures for the purposes of this proposal. Is there any particular reason you have an issue with this?

This is not the problem that you think it is.

There are only ~2k on-chain app stakes and that is not going to change prior to v1.0. These app stakes are administered through the Portal which allows many real apps to share one app stake, so this is not a constraint for real user growth. Assuming all of these app stakes were simultaneously in sessions, needing 24 nodes each, we would only need 48k nodes, which is less than double the current node count. Given node growth in the last few months and the fact that this proposal has a 5-month implementation schedule, we’ll likely be much closer to that figure by the time we have reached the final target. Also, if we lowered the node per session count back to 5 (which it was until last week), we would only need 10k nodes.

The more important metric for growth is not on-chain app stakes but the volume of relays coming through the existing app stakes that the Portal administers. And, as we have repeatedly alluded to, the current nodes are massively underutilized in terms of relay throughput. I can’t provide an exact figure but that doesn’t mean it isn’t true.

Also, try to frame this proposal in the context of v1.0. Once we have migrated to v1.0, which I anticipate happening within 1 year of this WAGMI reduction being fully implemented 5 months from now, everything in this proposal is obsolete anyway since we will likely implement app stake burning if not other more robust deflationary mechanisms.

Edit to Add: Since posting this I have also learned that there is technically no limit to the number of sessions that a node can simultaneously be in, which amplifies the point I was making.

5 Likes

I understand the desire to get control of inflation. But the impetus seems to be based upon a yet to be realized large uptick in relays. Why not set a formula that accomplishes this reduction in inflation that is tied to upticks in relays? So if relays average above XX million per day for a 30 day period, you make the first reduction in target inflation rate and then continue to scale down rewards proportional to increases in relays.

That way we don’t have runaway inflation from a sudden doubling or tripling of relays, but we also don’t have investors that are currently contemplating buying POKT to stake shying away from the project because reward structure is being cut prematurely. Or people who bought in the past 30 days bailing out because they made the decision to invest in an early stage project based on high returns.

If the team knows there is about to be a sudden massive increase in relays from onboarding new chains or there is some other reason to do this now (like a strategic investor insisting upon it as a condition to investment), then the team should just put all cards on the table and then people can get on board (or not) with full information.

One other point to note - and I’m looking at this from an investor point of view - the cuts to inflation and how they impact an individual’s willingness to allocate capital to this project also needs to take into account the reward share model that most node providers have implemented. That 15,486 per year at 150 days is really only net approximately 9,300 POKT per day if you are using a provider with a 60:40 reward split.

2 Likes

This is essentially what WAGMI does, except WAGMI provides more downside protection in the case that relays fall again. WAGMI ensures it’ll dynamically adjust up or down, as relays go up or down, and provides stable income to nodes.

So you’re talking 60% APY rather than 100% APY. That seems reasonable to me. Also, extra incentive to run your own nodes and contribute to the decentralization of our network :slight_smile:

4 Likes

I am more inclined to 1000 tokens at 2 dollars than to 100000 tokens at 0,02. Having a reputation of a token that only goes down is hard to beat. On the contrary having a reputation of a project that has a thoughtful deflationary mechanism will attract more people. Also a gradual roll out of WAGMI parameters insures that those who staked earlier are rewarded.

2 Likes

I understand this is crypto, but a 50% decrease on rewards within just one year is huge, and that’s even without considering that many new node runners have already lost 50%+ of their POKT value with the sharp drop in price. I really fail to understand why the team is so reluctant to consider a simple burn that would equally reach the alleged target without drastically reducing rewards.

I may have missed a prior message so I don’t understand this point. Generally speaking, I’m not opposed to a burn, but our protocol is only designed with so many mechanisms.

Where do you suggest this burn coming from?