PUP-11: WAGMI Inflation

Attributes

  • Author(s): @adam
  • Parameter: RelaysToTokensMultiplier
  • Current Value: .01 POKT per relay
  • New Value: Variable

Summary/Abstract

I’d like to encourage sustainable inflation that protects rewards against a decrease in relays and sensibly reduces rewards as Pocket Network grows. The proposal would implement an annual inflation target set by a DAO proposal, which would provide a predictable amount of inflation for users of Pocket Network. The first inflation target will be set by a different proposal. I have provided some suggestions below for numbers, but I would encourage you to debate what the numbers should be if you support the proposed scheme.

Note: I am a Pocket Core team member and Partner at NachoNodes (with Nymd on Discord). I run nodes and stand to benefit from extremely high inflation. The ideas in this proposal are my own and don’t necessarily reflect the views of others on the core team or Pocket Network, Inc.

Motivation

At present, we are experiencing high levels of inflation with no end in sight caused by the current RelaysToTokensMultiplier (I refer to this interchangeably as the mint rate). The graph below shows the relationship between the Relays per Day on the network and the annual inflation rate. There is a simple linear relationship between the two variables. The more relays, the more inflation grows on the network. This model has done a good job to get us where we needed to be.

First, I’d like to address the advantages of high inflation as there are certainly pros to keeping rewards at the same mint rate:

  • Very attractive rewards for new and existing participants

  • High revenue stats attracts more attention via leaderboards like the Web3Index

  • Excitement in the marketplace driven by attractive node rewards

  • Near term buying pressure from new node runners

  • Ability to bootstrap new chains more easily

The inflation level was great for bootstrapping to this point. Continuing to rely upon inflation instead of growing the value of the overall network is counterproductive.

I would argue that the cons of excessive inflation outweigh the pros at this point in our lifecycle. Here are a few that move the needle from my perspective:

  • Grow my slice mentality where users don’t necessarily care about token price, instead they care about selling immediately as to not be holding the bag. To explain further, if the inflation rate exceeds the increase in the rate at which the market price increases, POKT tokens will decrease in value. Taking this a step further, because node runners will realize that token prices won’t increase, sell pressure from wanting to exit the position immediately to get the best possible price will create a negative feedback loop on token price.

  • Constant sell pressure from fiat-based hardware costs (15K nodes * $75/node = $1.125M/month)

  • Major network underutilization leading to unnecessary hardware costs (i.e. we only need a certain amount of nodes to serve a certain amount of relays)

  • Perception of long-term unsustainability (hype around high rewards will eventually get outweighed by FUD of unsustainability)

  • Exponential growth in the number of nodes (if scalability ever becomes an issue with RC - 0.7)

While this level of inflation has pros and cons, we should consider the long-term impacts and perceptions from new participants to the network. I would posit that it’s more valuable to grow the value of the whole network rather than my slice of the pie. If we frame value generation from a network-wide perspective, then we should be doing everything we can to grow the value of the network rather than a few individuals.

At the present, users are motivated to keep up with or exceed inflation rates on the network which are leading to an arms race in the number of nodes. While this is the intended outcome of the economics in the early days, one could argue that we have accomplished our mission of providing a base-layer of infrastructure for Pocket Network. Hyper-inflation is no longer required.

In this regard, peaking at the right time matters. As we close in on critical (Copper Launch, wPOKT, etc.) events we don’t want the best times behind us. If we have inflated too much too soon, the market may not look so kindly upon the token. Instead, if we peak at the right time, it will add extra fuel to the fire and invite the market to enjoy the rewards of Pocket Network. We, as an ecosystem, don’t want to inflate ourselves to the point where there is no token price growth left to achieve because of inflation.

Rationale

I propose that there is a better way to manage inflation. My suggestion revolves around the idea of creating a targeted annual inflation rate or what I’m calling the Weighted Annual Gross Max Inflation (WAGMI) rate. The WAGMI rate would set an annual inflation figure, in POKT, from which the RelaysToTokensMultiplier would be adjusted dynamically. In practice, the WAGMI would be a figure set at the beginning of the period, from which we would set the initial RelaysToTokensMultiplier based on a given WAGMI target. In subsequent periods, the RelaysToTokensMultiplier would be adjusted per the curve established by the WAGMI.

An example WAGMI Mint Rate curve is provided below.

Source: Node Rewards Modeling. WAGMI = 50%

It becomes quite clear that as relays increase, the Mint Rate exponentially decays. What’s interesting is that from a total perspective the inflation remains constant. The WAGMI could be adjusted to ensure healthy, stable rewards are present. Further, this protects against a downside scenario where rewards decrease below [100M] relays/day. If relays/day decreases beyond that point, the mint rate increases to compensate for the decrease in relays, ensuring that there is a stable form of income for node runners on RelayChainIDs where relays do exist. Note that in the chart above we’ve set the mint rate for 100M relays to be 0.01, which is the current value. This means in practice that node runners should continue to earn what they’re currently earning as a baseline; this proposal just prevents hyperinflation beyond this point.

The execution of this activity would be delegated to the Foundation Directors. I suggest a semi-random schedule based on the trailing 30-day average relay/day so that the system is less likely to be gamed. Foundation Directors would adjust the RelaysToTokensMultiplier according to the trailing 30-day average to meet the WAGMI target as set by a separate proposal.

Dissenting Opinions

Perpetual fair launch - We should punish those that don’t participate. High inflation is a great way to do that.

This is true, but low to medium inflation also continues to punish people that don’t participate.

Give it to the DAO - We should not decrease inflation, but instead give it to the DAO. Hyper-inflation designed to maximize the DAO treasury helps all the holders of POKT.

While I’d be in favor of increasing the DAO allocation parameter as they are diluted as they don’t run nodes, we shouldn’t rely upon the DAO allocation parameter as a primary economic strategy whenever we want to solve economic issues. Even by increasing the treasury, we still end up with the problem of diluting all holders of POKT. To maximize the per token value of POKT, we need to avoid dilution in the first place. Simply moving the tokens around doesn’t eliminate the presence of the tokens and the pricing impact of those tokens. Further, the DAO allocation parameter isn’t a fix for larger economic problems because it merely hides the symptoms, rather than being the cure.

Targeting a fixed inflation rate, regardless of relay count, decouples the incentive for people to grow the relay count.

This is true. Instead, we’d be incentivized to grow the value of the network as a whole. Also, if we introduce the app stake burn, growing the relay count means growing the buying pressure from apps, which reintroduces the direct incentive to grow the relay count.

We can counter the selling pressure from fiat-based hardware costs by introducing the app stake burn so that the mint rate is directly matched by relay buy pressure.

While this is true, there are other trade-offs to introducing app stake burn at this time, and it shouldn’t be relied upon as the only strategy for countering sell pressure. It could be paired with this WAGMI proposal for a double-whammy - lowering the selling pressure while increasing the buying pressure.

Analyst(s)

@adam, Core Team Member, NachoNodes Partner, man with cats, pikpokt on discord

Copyright

Copyright and related rights waived via CC0.

9 Likes

I like this proposal.

The WAGMI rate calibrated correctly will serve as a backstop for hyperinflation while preserving the current APR which made the current growth and the level of awareness possible.

How this rate will evolve and what the process for its calibration should look like is another question but this proposal is a hard yes from us.

WAGMI.

8 Likes

I agree that inflation needs to be addressed sooner rather than later. with relays above 100M a day, inflation is close to 200% apr. If relays continue to climb, which we all hope they do, inflation will get out of control.

What we want to avoid are large cratering events in which price gets absolutely nuked due to too many free tokens printed, causing panic and mass unstaking of nodes in a reflexive cycle.

Just because price is not reacting to high inflation now does not mean we should not prepare for it. otc is illiquid and we do not know how markets will behave with liquidity.

8 Likes

Thanks for taking the time to get this up Adam. I want to get some initial thoughts down now to help others think through potential options as I flesh out a more detailed response.

  1. As an overall viewpoint, we want to reward our most active and best node runners, especially this early on as we build a core community around the protocol. By decreasing inflation too early we optimize more for speculation rather than active work done.

  2. I would like to explore framing this proposal as payback on node rewards, as opposed to protocol wide inflation. Today it is around ~5 months per node. What does it look like at 6, 7, etc?

  3. With the above points, I don’t think the high inflation we have today is problematic due to the “Goldilocks zone” we are currently in. We’re in a unique expansion mode and should monitor on-chain data to extrapolate new node runner growth rates. In the grand scheme of things, I don’t think inflating to billions (ball park no math behind this) of POKT is negative long term for the project due to the size of the market we are in. Don’t want to lose the forest through the trees here.

  4. Adam noted the “grow my slice mentality” as a negative. I think we collectively should be able to fluidly shift between what the right mentality is given the current state of the protocol. Growing my slice means I’m incentivized to stake POKT and do work, which is a positive. Those who are free riding get diluted.

  5. I would like to explore instead of lowering the node reward, increasing the 10% DAO allocation instead. At our current stage and price of POKT, I don’t see downsides in increasing treasury capabilities for the long run. Instead of burning potentially useful capital I would prefer to give the DAO stronger allocation potential.

  6. While I agree with the concerns about creating a negative loop with too much inflation and price decline, I believe there is enough buffer today in the rewards and price of POKT with many node runners already having reached ROI on their initial stakes. I view this as something we need to monitor, not a short-medium term risk

These are just some off the cuff thoughts to help spark a bit more conversation and thought - will work on fleshing more of this out next week.

9 Likes

3 Likes

On the whole, I don’t disagree with many of your points. The timing of implementation of the WAGMI system would require care and could be adjusted as the market changes, which I think addresses many of the concerns.

To clarify on a few points:

While I agree that this is a positive of uncapped inflation, I would counter that setting an inflationary target does not necessarily mean reward minting decreases, it only ensures that we have certainty around what that inflation will be. These ideas aren’t mutually exclusive. If we set the WAGMI high enough, it could increase the current mint rate. Further, any inflation, no matter how small, has the same effect. I’m unsure of what percent moves the needle for people to act to avoid dilution, but I believe it could be as low as 20% annually.


I reference this above, again, not mutually exclusive with my points in this proposal. If the market views treasury tokens as non-circulating tokens, I think this is a strong and compelling option. If not, I’d be concerned about being a project with a high fully-diluted market cap and a high number of tokens. Right now, we’re in this Goldilocks zone where market cap is increasing due to adoption and momentum, and that’s outpacing inflation. As soon as that flips, we’re in for a world of hurt where we kick in the negative feedback loop referenced by @miner:

I would be open to an increase in the DAO allocation parameter.

I can see the camps forming already :slight_smile:

1 Like

Before I begin, I am not an economist, so I’m operating off high level theory here, not deep numbers. NFA DYOR etc. etc. :smiley:

I’m also undecided on this proposal currently, so I want to put my thoughts out there and see where it goes, so please take everything here as thinking out loud and asking question, not actual criticism.

First, I think I speak for a number of noderunners when I say that our strategy is currently based around what might be considered the “accumulation” phase of Pocket. We’ve studied the tokenomics docs countless times, listened to the 2020 economics summit, run 2-3 year projections on growth, etc. Pocket represents a significant investment in fiat for many of us from both a hardware and token stake perspective, and we’ve built out multiyear strategies around the established economic model.

So, any change to that model will likely cause some consternation, and much rebuilding of spreadsheets and formulas. That’s not the end of the world, especially if the change helps the network grow, and drives value in the underlying economics, but no matter what, there’s going to be some aspect of “Hey, that’s not what we were promised!”

Second,

I don’t think these two things are mutually exclusive. Inflation typically reduces the buying power of the currency in question, and yet we’ve seen the value of $POKT steadily increasing in secondary market buying in spite of the rapidly increasing supply, due to a commensurate increase in TVL. Freely circulating supply is not keeping up with new minting. What’s more, the volume of purchasing has increased in spite of the rising prices and rising inflation, signaling the likelihood of orders of magnitude higher demand in the open marketplace once DEX and CEX trading are supported.

Is this not growing the value of the overall network?

While I agree that inflation needs to be mitigated at some point, I’m not seeing the triggers that make this a priority right now; the economic model is working as intended, and we have already approved economic controls that can be implemented to control inflation when ready (e.g. “the tenthening”).

See my opening statement about not being an economist, and I apologize in advance for any glaring and obvious errors in the next part:

What are these assumptions based on? We know that “rational market” models don’t often align with reality, since the market is wildly irrational by design. I looked through the formulas being used to derive the values in the spreadsheet, and I can’t figure out the logic behind them. They seem to assume a fixed relationship between token supply and token price which is inverse to what we’re actually seeing in the markets, and I don’t see any accounting for TLV in the spreadsheet, which is part of what is driving the perception of scarcity, and pushing prices up. Since October, we’ve seen an inexorable march towards staked flipping unstaked in C0d3r’s charts:

How does the increasing TLV factor into your calculations?

Given my failure to understand how the assumptions in the first part have been made, this doesn’t land with me at all, since token prices HAVE been increasing in spite of the existing inflation, and perhaps even in response to it. Stonks go up, more people buy. Even if we saw a temporary downturn in price (like we had over the summer, for instance), it seems to me based on history that we might see a short lived exit phase (mid-July in the chart) that quickly stabilizes as cheap entry points bring in new noderunners replacing the ones that left:

image

The idea that noderunners would “realize token prices won’t increase” sounds like an opinion stated as fact, which makes axioms predicated on that realization not compelling.

Third, I agree completely that “sustainability” is a critical component of the network growth. Since 2013, I’ve heard that the rising price of Bitcoin was not sustainable, yet the Grandaddy token continues to set new all time high prices orders of magnitude from where it began. Sustainability models must account for increases in demand of both the token, and the underlying service, with a little bonus for FOMO effects and speculation.

Regarding the underlying service, Pocket seems well positioned to achieve multiples in growth of network relays, which will result in supply side demand, buy pressure on the token to get new nodes staked, and additional PR, all bullish signals for the market at large. Additionally, the current limited markets available for Pocket mean that the token hasn’t been exposed to the 100x demand of mainstream listings and speculation. The upside bull calculations could potentially 10x the current VWAP, and trigger the equilibrium calculations for a tenthening type event, a reduction in the node stake requirements, and other already existing pricing controls built into the protocol and administered by the DAO.

The flip side, of course, is the potential for a massive bear market. The current distribution of relays by chain has a high risk profile, so until that is stabilized with multiple 100MM per day relay providers, it is certainly plausible that the network could see a large enough reduction in demand that many noderunners exit their position, and apply downward pressure to the price on their way out.

But again, isn’t this literally working as expected? The tokenomics, and MIchael’s previous discussions on the topic indicate this is exactly the scenario the tokenomics were designed around; increased supply demand draws in more noderunners which drive the price up until the supply outstrips the demand and drops the price as whales accumulate, and as demand returns, the inverse occurs, with long holders profiting on the churn.

I guess having gotten this far and talking it out in my head, what you’re proposing isn’t truly a sustainability mechanism, but a volatility dampener. And in dampening volatility, we reduce some of the mechanisms which attract traders and speculative investors. Perhaps that’s desired, perhaps not. I’m certainly not inclined to see pricing in the teens again, but there are arguments to be made for shakeouts and accumulation phases in an open marketplace.

As for the relative merits of the underlying mechanism proposed, then yes, if we’re going to put a predictable inflation model in place, then this one seems to be well thought out, and who can resist a good crypto pun like WAGMI?

I’m just not convinced yet of the underlying need, some of the justifications for the proposal, and why this proposal is better than the existing controls available. I encourage all of the smarter folks than me in this community to shred my response and help me understand.

7 Likes

I don’t think speculators who don’t run nodes are necessarily free riding. They provide to the protocol in terms of capital. if this class of buyer all sold their pokt roi would be much less on nodes, because market cap would be much less.

The question is if some of these market participants can be converted to node runners by dilution. maybe, but i dont think totally alienating the holder only class of buyer is beneficial either.

It will be interesting to see how much of the currently unstaked pokt goes into regen farming, which seems much easier to hop in and out of.

Another thing i want to note is that having excessively high apr values can be confusing for new market participants and sometimes brings a negative connotation and air of uncertainty for people looking to enter. its less clear what the downside risks are when people have been farming at 500% apr for months, compared to a simple price chart, and this could hurt pokt’s mass adoption into the broader crypto market.

9 Likes

Copper Launch, bridge, WPOKT, DEX & CEX-listings will introduce POKT to the masses. Putting any ceiling on exponential growth before those specific events would be doing POKT a disservice. I agree that this conversation needs to begin now, and inflation does need to be tamed in time, but not until we really let the market speak for itself. In terms of fighting the soon-to-come FUD regarding inflation, the question is whether the transparency and communication around such a proposal will be enough to state: “we’re working on it”.

Can someone give Adam a gold-star or something for “WAGMI”? Best double entendre in crypto of all time!!

8 Likes

This is a very good take.

We need to have a good answer to “omg inflation wtf” but it doesn’t mean we need to actually implement this mechanism now.

5 Likes

After reviewing some of the feedback, I have a few thoughts to add to the conversation.

First off, thanks for the feedback. You guys & gals are the best. I’m glad this didn’t get immediately shot down. To @Jinx’s point, I’m making a lot of assumptions here, which is an unfortunately reality of the economic phase we’re in now. Thanks for not being too harsh.

One of my evolving thoughts is that I’m in favor of increasing the DAO allocation from 10% to something higher. I would argue that this takes those tokens out of circulation and thereby shouldn’t effect token price as much as other mechanisms. While I list this as a contrarian view in my proposal, I don’t think it should be overlooked as an investment the community should make for the greater good. We’re going to need more tokens to grow this thing - let’s all agree to make that happen. That’s a whole different proposal/discussion. Point is, let’s not rely on this as a cure all.

I think we’re circling up on a general sentiment that we need to have an answer to why inflation is so high and that we’ve got a plan to rein it in - even if we choose not to.

This was my attempt to limit the impact of inflation while providing effective incentives. Additionally, this proposal protects our downside better than the current system, which we may want to rely on in a downside scenario. @Jinx put this well:

I think @Rosati makes a good point about keeping everything stable for now. I think we can accomplish this with WAGMI. If we use the base-case scenario I’ve mapped out here, there is only a small effect on inflation at the moment. I think the important thing to note with WAGMI is that we should agree on the process, then the numbers. With WAGMI, we could actually increase inflation at the current relay rate. I’d be in favor of implementing this in a way with negligible impact today to avoid any sort of impacts on new participants.

I’m open to timeline of implementation. I don’t think it needs to be today, but I do think we need to have a sensible answer to this in the near future.

5 Likes

The elegance of your arguments is a thing of beauty, Adam.

I like that as a separate proposal as well, and I’m in favor of it.

Man, I’m sold. Quantifying a predictable number that keeps it high for now, with a clear process to adjust the number for some predictable period of time (and I think quarterly is better than annually from a management perspective, even if most quarters stay the same), is likely a much more stable control than a tenthening type event, which would have some pretty radical impacts, and be difficult to walk back.

If passed, I would not want to see this implemented before the beginning of Q2 '22. We have a LOT going on in Q1, and disrupting the current inflationary cycle would be detrimental to what will likely be a surge of profitability during that time. With that being said:

I support this proposal.

6 Likes

Amazing commentary here guys, after reading how the thoughts have evolved with a higher DAO allocation, I am coming around to being in support of this proposal.

Like Jinx and Michael, the biggest concern is the timing of this implementation. I would also push for implementation of WAGMI sometime after Q2 2022 (or even later) once Copper launch and bridge have been established.

4 Likes

Even if we begin WAGMI implementation in Q2, that’s enough runway for someone entering today to literally double their position through weekly compounding between now and then. I’m behind it especially given the plan is to implement through analysis and fine tuning along the way. No painting ourselves into a corner in other words. Good stuff and thank you for the responsible stewardship.

2 Likes

I am in favor of this proposal. This is something that should definitely be discussed and figured out ahead of time. Every aspect within the Pocket Ecosystem needs to be well balanced so that one side never unbalance the other to a way it’ll impact the project negatively. As long as adjustments can be made as the network expands and economic factors changes, we will be able to keep growing Pocket Network as a whole on a stable and healthy path. Having a healthy balance is the key to the success of Pocket Network and along with it, all of us.

3 Likes

Thanks, @adam for this well put together proposal. And :clap:for all the contributions in the thread too. Sadly, it seems there is no collective noun for octopuses, as this is an impressive insert collective noun here of Poktopuses.

While I also stand to benefit from high inflation from running nodes in the short term, there is a need to manage inflation in the medium to long term for all the very good reasons given by everyone in this thread. And as much as we all loove the returns now, I think we know it’s not sustainable at current levels.

I’m generally in favour of the proposal as we need to build consensus as a community on a sensible path forward. The process also needs to feel legitimate, even if there is some dissent around the actual target from time to time.

Before we get to numbers, I think there are a few things worth exploring that should help the future WAGMI committee decide on the quarterly (I agree with @Jinx that this feels about right in terms of update frequency) inflation target.

  1. What does this model look like at the extremes? Eg a 25x increase in relays? 5B relays per day is extremely punchy, but Pocket will be doing a lot more if it becomes the dominant RPC provider for all chains.

I do worry about this, although it could be caveated by education around the point that nodes who more readily support new relay markets that Pocket network supports have a quasi-equity stake in such markets as there will be less nodes competing for such rewards, and they will get rewarded well for actively supporting adoption of Pocket network endpoints that they run nodes for.

What are the timings for onboarding direct application staking of POKT? And to encouraging more buy pressure from the demand side?

  1. Is the WAGMI model unfair to newer long-tail markets? Do rewards get too low for such markets once the network as a whole is supporting large numbers of relays?

4.What other parameters should we think about?

  • The question from @o_rourke about the profitability of node runners is a good one, although I think this data point should be considered in line with the target number of node runners for the network. Eg. if there are 100,000 node runners in the network, do all of them need to be super profitable? How many node runners does the network need to service apps, provide redundancy, and cater for growth? If much lower than 100,000 nodes, rewards are probably too high

  • Profitability should also consider the $ price per POKT. If the price per POKT goes up 20x, then the inflation can - and arguably should - come down.

  • The participation rate is mentioned a lot in the docs - Participation Rate - Pocket Network - while I don’t think this is a fantastic parameter, it is worth watching to see the demand to stake POKT in the network to earn rewards as a node runner.

Really enjoying the discussion so far.

I think it’s fair to say that WAGMI

1 Like

I believe poktopi is our preferred nomenclature. :smiley:

2 Likes

I am new on the POKT community, and I do have on the present inflation rate my way to accumulate more POKT’s. It is the core of my strategy now. BUT, this rate of inflation looks like not sustainable in the medium run (I am not even talking about the long run).

Please, correct me if I am wrong, all assumptions I am making here are based on what I read here and there and comments from all the community.

Besides all the cons that high inflation generates, there might be another point that this proposal will cover. If the relay cost is pegged to the USD and not to POKT price, the WAGMI proposal will fit it perfectly when POKT prices increases. In the other hand, if inflation keeps at the same pace and prices increases (which I think is not doable, but lets just think it is) the cost per relay will increase and POKT will end up with same problem as ETH gas fees right? E.g. if POKT prices goes up to 5 usd, and the relay costs 0,01 POKT then the cost of relay is now 0,05 usd and not just 0,01 usd. Therefore, the network might lose ground to competitors.

Again, I can be completely wrong if there is already a mechanism that control it (POKT price variation x relay costs), since I am new here, I am learning by the day. So, apologies if what I just said is not applicable.

2 Likes

This issue has huge implications for Pocket Network’s future. No one wants WAGMI to become NGMI. Jinx makes the point that he is not an economist. I’d be curious to know the perspective of a “crypto economist.” Given the importance and complexity of the inflation question, why not commission a study by an outside expert who is not invested in the project? More fat for the community and DAO members to chew.

2 Likes