UPD: Summary of the discussed models (exact number below) which one should be pushed as YIP:
Substreight 50k Model - Decaying V2
Neither (more discussion)
This is a request for comment on a monetary model for $YFI. It will often refer to Daryl’s original post on emission.
Thoughts and Intentions:
The emission schedules being discussed in (B) and (yfi_whale) models on the respective google sheet are slightly too low, especially given the high initial emission rates and radical decay. We have the opportunity to create a monetary policy which is a bit more long term and prudent, while serving the interests of early stakeholders.
I would like to put forth and discuss a clear monetary policy to be integrated into proposal A. This is now listed as “50k Model” on the google sheet.
After initially forcing several different ideas into the same binary proposal, I’ve decided to break it up into three distinct proposals.
Proposal A For: Adopt 50k Model inflation schedule and set the YFI hard cap at 50,000 YFI tokens after 10 years. Against: Do not adopt 50k Model schedule.
Great work @Substreight. Continuously developing the emission model based on feedback from the community and the figure of 50,000 looks so tempting and in my opinion cap with 50,000 after 10 years is a long term view with perfect low emission.
But I am also supportive to @yfi_whale proposal with 50% initial inflation to bootstrap network and reward for early participants.
I appreciate the work on the model. The Substreight model has IMO the best numbers for the first 2 years, but I’d rather not finish as high as 50k. D and E are much too high inflation.
However I question if this type of model is the best approach to suit YFI. This model is approaching this as a “monetary model”, which works for something like Bitcoin that is primarily a Store of Value. However YFI is more analogous to a growth company in a growth industry.
I think a better approach would be a 2 year plan of issuance (for example 3k/year + 2k for Andre/Dev Fund, so 8k total over 2 years). Mint function ownership could maybe be transferred to a timelock governance contract and further issuance could be reevaluated closer to that time. This gives the market medium term assurance while keeping open flexibility to compete with new protocols as they come online.
I strongly agree with this and find that falling back onto the emissions models of Bitcoin, Synthetix, or some other base layer protocol is short sighted. YFI emission is the key to a successful ecosystem, but just focusing on minimal issuance or a hard cap severely limits the flexibility that yEarn has for the future?
Embedded in this conversation are massive questions about the incentivization of future product releases and the payment for future work to make those products real. What do we want to incentivize now and what do we think we will want to incentivize in the future? Is this some immutable protocol that will find a steady state and not change or is an organic entity that will thrive through good governance and useful product offerings?
Why decide on a highly inflationary early monetary policy now before we know what the landscape is going to be in 3 months? Can we compartmentalize these questions to both ease the market and allow flexibility? Right now, it seems the most important things that need to be incentivized are good governance and paying for Andre to keep working on the project, so creating and maintaining rewards for those items are paramount. Keeping but greatly reducing rewards to the yCRV pool also seems like a good idea to not have a mass exodus from that pool, but we could also just commit to back paying liquidity providers who are in during the lull.
Without answer these larger questions, no monetary policy will be credible and will most likely require changes in the future.
The only thing I know for certain, is that we WONT have the perfect long term monetary policy figured out in the first proposal so I would opt for the lowest inflation model. We WONT know what other rewards and incentive programs we will want to start up in the future just like Synthetix started 5 new incentive programs they had not planned for when they initially set out to do this. Really, not much incentives are needed at all since ypool had a lot of traction before YFI.
If we want to anchor on some numbers, I would suggest we set inflation that aims for 30-50% reward yield for the short term as that is very attractive and competitive with other yields
I would opt for what has the lowest inflation and we can work our inflation up if needed. I would vote for the model that has the least aggressive start - that would be Model A if looking at 1Y YFI supply
In terms of pools, we should direct it to only one 80 YFI / 20 yCRV pool as that would be most capital efficient. It accomplishes all the goals of incentivizing yCRV usage, YFI liquidity and potentially governance all at once
I totally agree with this sentiment and the more generalized idea to try and keep our options open, as I’ve expressed in this thread and others.
However it seems that there is a portion of the community who are eager to make a decision so that the Minting keys can be burned. I do understand this from a security perspective, even though overall I think the increased flexibility is worth it.
@rewkang do you (or anyone else in favor of keeping our options open) have solid ideas on ways to mitigate the security risks (as well as YFI Inflation Uncertainty to a lesser extent) without burning the multi-sig keys? I think a solid proposal on that front could help bridge the two sides.
Interested in hearing ideas on this as well. Really appreciate all the points. @rgalloway@rewkang@Sinsecato I agree that there should be more room left open to adapt to uncertainty in the future.
One option that was discussed last night was to set a linear inflation schedule and split 50% of rewards to LPs and 50% to the Multising//DAO which could be leveraged for more informed decisions as new ideas and opportunities arise. (andre payouts, dev fund, new incentive schemes, etc). What if this were set to a simple 3k/YFI per year in the beginning, allowing 1.5k YFI to LPs and 1.5k to Multisig//DAO?
The market stability and security a hard cap or very low long-tail emission provides is hard to match. Imo we should work towards a round hard cap sooner than later.
These are good points and it is important to keep our options open while the monetary policy is further developed. I worry that voting on, and accepting, a monetary policy will shift the narrative and it will become significantly harder to make changes to it.
Instead of picking a monetary policy now, could we not publicly signal that even though new issuance of $YFI will pause, yCRV minters and governance participants will be back-paid a pro-rated yield in $YFI tokens once a monetary policy is chosen? This model is conceptually similar to the assumed distribution of Curve.fi’s governance tokens, which will be paid out to liquidity providers who have been active since the project’s inception. Signalling this publicly would help prevent a a feared mass exodus of liquidity from the y.curve pool while the community continues to discuss monetary policy and emission schedules.
With that said, I have been writing up my thoughts about fairness and the potential benefits of higher inflation that I think should at least be considered when deciding upon a monetary policy.
On Fairness and the Potential Benefits of a High Issuance Rate
In light of all the competing proposals for Emissions schedules and monetary policies, I wanted to write a bit about the idea of fairness. Namely, how should we approach fairness when determining a monetary policy? This is a big question but should be considered as it could define how the project is viewed to outsiders.
Many in the crypto space have taken notice of YFI because of the way that it was released. It has a working product, no pre-mine or pre-sale, no VC money, and everything was announced publicly. I agree with these sentiments, but do want to highlight the now privileged position that this puts everyone who had the opportunity to farm for the last week. Even if this distribution method was fair by most standards, if we now drastically lower the rewards for new entrants into the space we should be aware that it could be viewed as a set of privileged actors skewing the playing field in their favor.
It could be argued that continuing the rewards at their current emission speed is the “fairest” method for everyone but those who recently bought YFI with the expectation that there would only be 30k or 50k total supply. Continuing with high inflation may have a long term negative effect on the price of the YFI token, but community members and farmers who are committed to the project can continue to provide liquidity and participate in governance and be minimally diluted.
These are compelling arguments, but I would like to make a 2 counterarguments that I have not yet seen brought up.
One way to think about these issues is to frame them in terms of the total value or marketcap of the YFI token in the long term. If one determines a valuation for the YFI token based on the rights to future cash flows from the yEarn ecosystem, then the value of all YFI tokens combined should be the same independent of the total supply.
If the yEarn ecosystem earn $1,000,000 in revenues a year that is returned to token holders, it doesn’t matter if there are 50k tokens, 50million, or 50. The rewards are split between all of them. A lower emission rate privileges those who got in early. A higher rate benefits later entrants as all stakeholders need to continue contributing to the ecosystem in order to keep up with issuance. I would argue that a higher emission rate is inherently incentive aligning.
Higher issuance could also prevent governance capture, especially if mechanisms are put in place to incentivize governance participation. In a high inflation model, early token holders who stop being active in the ecosystem will eventually get their voting power diluted. I recommend looking into the model that the dxDAO has implemented as a case study for how this has been done before. I am not implying that their model be adopted as it works quite differently, but they have made a commitment to be generous in giving out REP (their pure governance token) to active members so that non active members are quickly diluted.
It is important that we take into consideration the future views of people who are not so deep in DeFi and didn’t have the opportunity to get in “at the ground floor.” Even if early entrants have taken significant risks to acquire YFI, getting our piece of the pie and then turning around and telling everyone else that we are now only giving away the crumbs is not a good look and it could sour public sentiment of the project.
Interested in hearing other people’s thoughts on this. happy to break it out into a separate thread if that will keep this conversation more focused.
Is burning the keys not going to be an option then. Once a schedule is decided? I don’t like any schedule without burning the keys. Otherwise we’ll be debating this same story in 2,3,5,8 years when ‘easy money’ drys up. Don’t people work hard for their capital any more? I’m here on just a hand full of YFI. And I had NO second though about putting time and effort away from my kids and family to mock up designs. My point is. If your invested and HAVE a skill that can help your investment, I’ve every faith that people will step forward to make it happen. As for remuneration. I do this because I love to design and I’d love more than anything to see my designs live on a project like this, but knowing many won’t have this view, I strongly suggest we have a tread to discuss diverting a % of protocol yield to a multi sig wallet that either accumulated in yCRV or YFI (through automated open market purchases) or both. And to use this % mechanism (dialling that up and down as needed) rather than playing with minting.
So my vote will require a burning of the keys and the addition of a mechanism to dial up or down the % of protocol yield to a multi sig wallet for use in future endeavours.
I’m completely against this idea. The reason we have crypto is so we don’t have to trust what people will do in the future. We can look at the code and know what will happen. Certainty in economic policy is paramount.
I 100% understand where you are coming from, but I think it is worth zooming out a bit and considering all our assumptions and then reviewing how they apply in this situation.
There are a few options that need to be chosen between and tradeoffs for all of them.
Option 1: pick a monetary policy in the next day and never change it
This is the most pure crypto option in the spirit of Bitcoin. It has a successful track record in working for Bitcoin. Much of the impetus around this model is the notion that when people are able to change a monetary policy, they do so for bad or poor reasons and in the end make the system worst for having changed it. If this is the option we go with then we either get it right the first time or try again after the whole system has fallen apart. This trades one form of risk for another, and this is really true for all options.
Option 2: Pick a potentially final monetary policy now with the explicit expectation that it may be tweaked going forward
I think this describes the rationale for this option well:
If we go this path then there should be some level of understanding as to what the scope of potential future changes is. Are we just adjusting curve shapes and percentages to different pools or uses while maintaining the same max supply? Is it possible to increase the max supply? create a tail emission in perpetuity?
Option 3: Pick a clearly temporary monetary policy that everyone knows will be changed in the future
So I lied above when I quoted @rewkang as an example of option 2, as it could also be considered as an example of option 3
We are now moving pretty far away from option 1 and I can hear the Bitcoiners grinding their teeth. The reason to adopt this strategy is it could make it easier to actually make the requires changes to the protocol when unexpected events occur or when we have a more “final” monetary policy.
option 4: no more issuance until a policy is decided upon in the near future
This is what I proposed above. It trades off a risk that liquidity will flee without a hard promise of a future monetary policy with the risk that we pick something, get it wrong, and then can’t change it.
There are probably other options that I haven’t covered
@Mr_Sadim 's point above is only held if we adopt option 1. Otherwise we have varying levels of trust in the governance structure that has been created.
Questions for any model proposed
Is this model applicable and best for yEarn? If we pick a policy based off another project (i.e. Bitcoin), will it work as well here? Is the ability, or inability to change or adjust parameters a net positive for yEarn? Is absolute scarcity as a feature or meme a key value driver for yEarn?
I would argue that the value of YFI comes through the usage of the yEarn system first and foremost and not just as some token that people should hoard. As I mentioned in my post above about fairness and inflation, if the real value of YFI comes from future cash flows and revenue from yEarn, and and open and inclusive fair governance model, then the Bitcoin model may not be applicable. Is this project about trusting in only code and Austrian economics, or is it about DeFi, Yield Farming, and hyper inclusive and fair governance? How far do we want to push the idea that the community can govern well and that we can remain flexible in the face of a rapidly changing crypto landscape?
This is a very different time compared to bitcoin launch. YFI arguably has had a very fair launch and the current token holders are one of the most in savvy and in the loop people in crypto today. The forum activity and the quality of discourse attests to that.