PUP-13: Initial WAGMI Parameters

PUP-13 passed with 67% quorum and 100% support. Snapshot

This means the RelaysToTokensMultiplier parameter will now be adjusted in order to target a 100% network-wide inflation rate. The parameter change hasn’t gone into effect yet, I will update here when it has.

The parameter will also be adjusted again every 30 days from now on the following dates:

  • Mar 26th: 90%
  • Apr 25th: 80%
  • May 25th: 70%
  • Jun 24th: 60%
  • Jul 24th: 50%

I will update on these dates when the parameter change happens again.

To avoid confusion, could you please confirm that adjustments to RelaysToTokensMultiplier will continue to be made after July 24th on a semi-random schedule using a trailing 30 day average per PUP-11 in order to maintain network-wide inflation of 50% going forward. Also, could you please confirm we will be targeting a percentage the POKT supply at the time each adjustment is made, as opposed to a percentage of whatever the POKT supply was when PUP-11 passed - since POKT supply will increase significantly between adjustments.

For the avoidance of doubt, the WAGMI parameter changes will be calculated using:

  1. a 30-day trailing average of daily relays at the time of each adjustment
  2. the total supply at the time of the proposal passing (Feb 24, 2022, 6:37 GMT), which corresponds to block height 51909, which is 945,014,988.719332 POKT. We’ll call this the Total Supply Baseline.

If we were to use the total supply at the time of each adjustment, as opposed to the Total Supply Baseline, we would not actually be hitting the annual inflation targets. In a period of sideways relays, the total supply at the time of each adjustment will be continuously growing, which would mean the inflation target essentially compounding on itself. In this proposal’s spirit of stabilizing inflation, we will be using the Total Supply Baseline for all future adjustments until another proposal is passed or 1 year has elapsed.

The parameter is going to be adjusted now to target a 100% annual network-wide inflation rate based on a 30-day trailing average of 306M daily relays and the Total Supply Baseline. This means a RelaysToTokensMultiplier value of 0.008461, reduced from 0.01. I will update here when the parameter change has been executed.

The parameter will be adjusted again every 30 days from now on the following dates, to target the following inflation rates:
• Mar 26th: 90%
• Apr 25th: 80%
• May 25th: 70%
• Jun 24th: 60%
• Jul 24th: 50%

The calculation will be performed on these dates using a 30-day trailing average of daily relays and the Total Supply Baseline. If daily relays changes significantly in between these dates, we will make discretionary adjustments to smooth out the reduction of the parameter.

Once we have hit the 50% target, the parameter will continue to be adjusted according to this method to maintain 50% regardless of the relay count, until a new proposal is passed.

1 Like

Thank you for the clarification.

After following this for quite a while and seeing how this has played out I have to agree with some of the dissenters here. This feels a bit forced and rushed and poorly timed much like the exchange listings just over a month ago. As a comparison, wPOKT has been under development for over 3 months and outside third parties/experts were hired to assist with its development and implementation. While I understand that is substantially different from implementing an inflation adjustment curve I believe this comparison is worth noting. When others here expressed the need for a third party/expert/economist to weigh in on a drastic reduction in rewards affecting the entire community their suggestion was largely ignored and the proposal moved forward without consideration. The optics of this are unsavory at best. Without a neutral third party weighing in on these inflation adjustment parameters and their corresponding timeline they appear arbitrary. Dealing in abstracts with something as impactful as this is ill-advised. I believe this implementation should be halted until a study is performed by an outside third party with suggestions based on metrics and projections that can be shared with the community, debated openly, and then voted upon.

2 Likes

Development was outsourced because it’s not a core priority not because we needed neutral third-party expertise. I don’t think the comparison is relevant.

The magnitude of the reduction shouldn’t be calculated based on the last couple of months. If you zoom out, we’re looking at a return to the baseline rewards that most community members were happy with before.

For this to happen, you would need to submit another proposal to revert PUP-13, like this one did for PUP-4.

Also to clarify/make sure I have the numbers right.

The RelaysToTokensMultiplier will be adjusted as follows:

2/23/2022 - 0.01
2/24/2022 - 100% Inflation Rate = 0.008461
3/24/2022 - 90% = 0.0076149
4/24/2022 - 80% = 0.0067688
5/24/2022 - 70% = 0.0059227
6/24/2022 - 60% = 0.0050766
7/24/2022 - 50% = 0.0042305

Which will have the following effect on rewards:

2/23/2022 - 100%
2/24/2022 - 85% of Current Rewards (Immediate 15% reduction)
3/24/2022 - 76% of Current Rewards
4/24/2022 - 68% of Current Rewards
5/24/2022 - 59% of Current Rewards
6/24/2022 - 51% of Current Rewards
7/24/2022 - 42% of Current Rewards

Ending in a total of 58% reduction of POKT/Node/Day in July 2022 and continuing forward?

Please correct if any of this is wrong I want to be clear.

The RelaysToTokensMultiplier value will depend on the 30-day trailing average of daily relays. Assuming the 30-day trailing average remains constant at ~306M relays/day, your first calculations would be correct, though this can’t be translated to an estimate of per-node rewards.

Let’s say the 30-day trailing average were to drop to 100M relays/day, the 90% target parameter value would actually be 0.023302. So, while we’re reducing the upside, we’re also increasing the floor, meaning less downside risk.

WAGMI targets total inflation but we can’t predict from that what the % reduction/increase of per-node rewards will be because the RelaysToTokensMultiplier parameter value depends on the 30-day trailing average daily relay count and the resulting per-node rewards depends on the total node count.

Understood. Thank you for that explanation.

this is so interesting to watch. I have only been following the project for under 3 months now so it is tough to fight too fervently for a perspective. My immediate reaction to the infla debates that have been going on since PUP-11 though is that the inflation rate is ridiculously high ha. obviously attractive to new node runners who see 150% APY in a huge TAM and get tingly but, by any long term rational sense is so unsustainable. Fair, I am not showing models to prove this but, 150% inflation??? fun but if you understand the mechanics that increased demand will lead to more relays and higher infla it just gets silly.

That said, to many dissenters here, I do understand the perspective that it Feels bad to have the incentive design changed right after going to exchanges and more adoption. Perhaps the team should have done these in the reverse order, but I think that is just growing pains not playing games.

As a long-term solution I don’t think inflation targeting makes sense, it would be a game of guess-and-check. I understand the need for a stop-gap though so we don’t continue the hyperinfla. Two thoughts on longer-term incentive design below and would love to hear from the team what ideas are being thrown around.

A) on inflation: infla is the Pocket Network’s mechanism to pay node runners and as well as the cost to app stakers via dilution. The balancing mechanism to stop the price from going down (due to infla) would be demand growth or inorganic node growth (ie not compounding nodes). So the game here is to balance app demand with inflation. So theoretically a stable price relationship is where [(increase in app relay demand per session) / (relays/POKT)]/POKT in circulation = infla rate. Some issues: this assumes increasing demand and based on this relationship, if it were to fall then you would have deflation. Also this ignores inorganic node growth because it assumes that the market will find an equilibrium for number of nodes based on cost to run a node and infla. Certainly, if the growth side is strong so infla is high then it will incentivize for supply which should both push price higher and vice versa.

B) on balancing supply and demand: One of the levers in the incentive system is relays/POKT for app stakers. To balance supply growth via relays and cost to nodes measured in POKT price you could define the relays/POKT variable as a function of the price of POKT in a positive relationship. So as POKT px goes up, the number of relays/POKT goes up and vice versa on the way down. My intent on the incentive design is to balance supply and demand such that a higher POKT price in USD terms means node runners have to sell fewer POKT to fund their nodes (which cost in USD) which means less supply entering the market, so the cost to apps can go down. When price falls, there could in theory be a spiral lower as nodes need to sell more to fund operations. So by increasing the demand by increasing the relays/POKT for new apps you can reduce volatility and hopefully find an equilibrium in the market. Perhaps not relevant for the current growth stage in the market but it’s another tool that I think is reasonable incentive design in this economy. Is there an oracle problem here about verifying the price of POKT?

Tough job trying to create the right incentive structure in this new protocol to make it sustainable and beneficial for users (apps) and suppliers (node runners). Economics is the dismal science. I think hyperinflation is a dangerous game so I’m glad to see something has been done about it. Excited to see the team continue their work on this and hope the next four years are as or more successful than the last.

Then again the rewards reduction percentage is ridiculous, and the “considerate” timeframe just makes it a little bit more so.

If only you guys were as clear with maximum supply, and when this future token burn will be enforced, as how you’re for the inflation targets, it would be way easier to properly value the project and wait for another 75+% to “invest” in it.

There are many people who believe the current reward rate is ridiculously high, as is the fact that we’re reducing gradually over 5 months rather than immediately. You’re welcome to your opinion, just know that it’s in the minority.

We can’t predict when the burn can be instated (and thus what the total supply will be) because this is determined by the v1 development roadmap. If you’re running nodes these should be irrelevant to you anyway.

1 Like

The RelaysToTokensMultiplier parameter has now been updated to 0.008461 POKT per relay, which equates to a ~15% reduction in network-wide inflation.

As a reminder, the two WAGMI proposals (PUP-11 and PUP-13) were approved by the DAO and will manage inflation according to the following targets:

  • Mar 26th: 90%
  • Apr 25th: 80%
  • May 25th: 70%
  • Jun 24th: 60%
  • Jul 24th: 50%

Where the RelaysToTokensMultiplier value that achieves these inflation target will be calculated using:

  • a 30-day trailing average of daily relays at the time of each adjustment
  • the total supply at the time of the proposal passing (Feb 24, 2022, 6:37 GMT), which corresponds to block height 51909, which is 945,014,988.719332 POKT. We’ll call this the Total Supply Baseline.

If daily relays changes significantly in between the above dates, we will make discretionary adjustments to smooth out the reduction of the parameter.

This inflation management framework provides nodes with stable rewards on the upside and downside. If daily relays changes drastically in either direction, we are no longer at risk of a supply shock.

1 Like

I have closed this topic now. If people wish to debate PUP-13 or propose rescinding this DAO-approved decision, they are welcome to open up a new thread.

This thread is now only for updates on the execution of PUP-13 parameter changes.

2 Likes

Note: I already posted this update in the Discord #announcements channel but realize that not everyone is looking there. Not posting here too was an oversight and I’ll be sure to update here in the future.

The next WAGMI inflation reduction is upon us and the new annual network-wide inflation target being set is 90% . This equates to a reduction in the RelaysToTokensMultiplier from 0.008331 to 0.007498 .

This was calculated using a 30-day trailing average of ~311M relays/day and a total supply at the time of the WAGMI proposal passing of ~945M POKT. The next scheduled WAGMI reduction will be to an 80% target on Apr 25th .

Until then, due to the recent significant increase in relays, we anticipate making weekly adjustments anchoring around the 90% target with an updated 30-day trailing average each time. These discretionary adjustments will serve to smooth out fluctuations between the WAGMI reductions.

1 Like

Due to the recent spike in relays, we are making weekly adjustments around the 90% inflation target under WAGMI. These weekly adjustments hold everything constant while updating the 30-day trailing average daily relay count that is used to calculate the new RelaysToTokensMultiplier value.

The new value of the parameter is now 5776, corresponding with a 30-day trailing average of ~403M daily relays. This means 0.005776 POKT minted per relay.

We will continue to make weekly adjustments under the 90% inflation target, until we reduce the target to 80% on Apr 25th.

1 Like

On Friday we made another of our weekly adjustments.The new value of the parameter is 4847, corresponding with a 30-day trailing average of ~481M daily relays. This means 0.004847 POKT minted per relay.

We will continue to make weekly adjustments under the 90% inflation target, until we reduce the target to 80% on Apr 25th.

1 Like

The next WAGMI inflation reduction is upon us and the new annual network-wide inflation target being set is 80%. This equates to a reduction in the RelaysToTokensMultiplier to 2988, which means 0.002988 POKT minted per relay served.

This was calculated using a 30-day trailing average of ~693M relays/day and a total supply at the time of the WAGMI proposal passing of ~945M POKT.

The next scheduled WAGMI reduction will be to a 70% target on May 25th. Until then, if daily relays changes significantly, we anticipate making weekly adjustments anchoring around the 80% target with an updated 30-day trailing average each time. These discretionary adjustments would serve to smooth out fluctuations between the WAGMI reductions.

2 Likes

The next WAGMI inflation reduction is upon us and the new annual network-wide inflation target being set is 70%. This equates to a reduction in the RelaysToTokensMultiplier to 2109, which means 0.002109 POKT minted per relay served.

This was calculated using a 30-day trailing average of ~860M relays/day and a total supply at the time of the WAGMI proposal passing of ~945M POKT.

The next scheduled WAGMI reduction will be to a 60% target on June 24th.

3 Likes