[AIP-2] Transmuter Farming
Welp, guess this is an AIP now.
scoopy
unorthodox is fine with transparency.
I support the idea of using the transmuter to safely create more cashflow, but I would personally make it so that all revenue and farming proceeds go by default to the DAO treasury.
I am neutral on buyback programs, but I think they should come from the treasury and be voted on seperatly.
Also I think upgrading the transmuter would be a great opportunity to include other upgrades to the transmuter:
For example, around peg-protection of alUSD:
-emergency mode: a mode the team can trigger if there is some type of emergency that allows for ordely distribution of the DAI in transmuter if there ever was an event that made alUSD backed by less than 1 DAI.
-rate-limitation : max amount of DAI exchanged out of the transmuter per day
Paired together, this assures there can't be a run on the bank of the DAI in transmuter by arbitrageurs if there ever was a peg-breaking event.
state I support breaking this into two AIPs with seperate forum discussion groups
- One is the farming of Transmuter Dai, and then sending all the revenue to the DAO Treasury. In addition discussion around other Transmuter upgrades. There is a Lot to talk about regarding this and how the farming is implemented and deserves its own working group
- Second is any Buy Back program for ALCX using treasury funds. This also has a lot of implementation details that can easily need its own space.
And these two ideas don't need to launch together either, which is more reason to handle them seperately.
- Edited
In general I like this proposal but it’s not without risk. For one thing, this doesn’t guarantee a liquid alUSD-DAI transmuter. If a large quantity of people mint alUSD then all of a sudden the ratio becomes less than 1:1. This would ordinarily not be a problem since we can just withdraw from whatever source of yield we’re using, but it becomes a problem if the transmuter excess is lost due to a hack or some exploit found in the source of yield.
To remedy this, why not change the buy back mechanism to a dividend for ALCX holders. ALCX holders could serve as lenders of first resort, earning a portion the yield that transmuter excess earns. In the case of the yield source losing funds, staked ALCX could be sold to cover the deficit. This would also allow us to earn more yield since we could safely go under the 1:1 ratio in the transmuter and always have a backstop. I do like the idea of using transmuter yield in part to fund dev salaries, but I don’t think we need to use all of the yield for this purpose
- Edited
travisformayor if this happens there should be some kind of guarantee that no one can vote with the ALCX tokens in the treasury
Regarding the concern for the increased smart contract risk with farming funds in the Transmuter: I support the AIP specifying the farm source be a low risk choice, and keeping the % farmed really low at launch with plans to increase it over time as confidence+audits is established.
ButlerAndTheThirdStringers While ALCX bought back could also serve as a backstop, it would provide a downside risk to holding ALCX without any upside. A dividend would provide upside to holders to counteract the downside of serving as a lender of first resort. And again we could still use a large part of the yield to fund devs
- Edited
ButlerAndTheThirdStringers The stakedDai will still be liquid and avalibile to the transmuter during normal operations. It can be unwound in occasional replenish operations when ratios get low, or it can be called on demand if necessary. Given we are talking about a worst case scenario, ie the platform the Dai is earning yeild on is hacked, and not normal oprations I don't think we need to design anything too involved as protection. If we keep the amount of Dai 'at work' low enough compared to the idle Dai, then transmuter solvency wouldn't be an issue even in a worst case scenario.
We can always add on extra protection mechanisums (like the ones being discussed now) later if the desicion was made to increase the ratio and earn yeild on more of the transmuter fund. We could have a whole AIP discussion around just protocol fund protection at that point, including these ideas, opt-out insurance, etc etc. We should aim for the simplest possible version 1 for this current proposal, even if it means we use a lower % of the transmuter for yeild earning at first.
Awesome writeup @RogueItachi . A few thoughts:
1) i think while the protocol is still young it makes sense to be risk averse and maintain a strong backstop to the alUSD peg. starting with closer to 25% of the transmutable capital deployed seems safer, and can be raised over time, potentially even to 90+ percent.
2) reducing yearn dependency is a great idea. i think vanilla aave lending would be the highest, easiest yield we could get on dai right now, sitting around 10%. the technical integration would also be pretty straightforward.
3) I'm strongly in favor of sending the generated yield directly to the treasury and using it to hire more devs. i'm not sold on the buybacks just yet, as I think they will be absolutely dwarfed by current farming inflation rates, so the money would be better spent on actively building and improving the protocol. in 2 or 3 years as inflation tails off, it definitely makes sense to allocate more capital to ALCX holder dividends or ALCX buybacks.
I plan to start exploring the technical implementation this weekend. Always open to suggestions and input.
- Edited
travisformayor It would be planning for a worst case scenario, and if governance selects a "safer" yield mechanism on only a small percentage of the Transmuter excess then I see where you're coming from.
Despite that, I think a buy back without a burn is introducing too many negative price pressure's onto ALCX. We would not only have a steady inflation, but would be paying devs in ALCX which they would presumably need to sell in order to benefit from. Introducing this kind of downward pressure on the price results in an asset that nobody wants to hold, which is worse for devs, holders, and arguably the protocol as a whole since it disincentivizes participation in governance.
I'm not completely opposed to a buy back, but I think it needs more thought. A burn is an easy remedy, but people have mixed opinions on the practical effects this has on the price. Another suggestion might be to completely replace the inflation of ALCX through farming with a portion of the bought back ALCX. This would at least curb the inflation of supply, although we'd still have a negative pressure on the price due to devs needing to sell in order to pay themselves. You need some kind of positive pressure on ALCX to counteract this, otherwise the value of governance over the protocol slowly decreases over time. The only mechanisms I know for accomplishing this are a dividend or burn
- Edited
All really good points. ButlerAndTheThirdStringers This plus what thetechnocratic added reinforces for me the need to split buy back and transmuter farming discussions into two seperate topics. I started another forum post on buy backs so we have a place for it.
This conversation seems mostly focused on the transmuter farming part of the original proposal, which I think is also the most time sensitive part of the conversation given that excess Dai is just sitting there, so I would suggest we continue with that part here.
travisformayor good idea. generating revenue from the stale transmuter funds and what to do w/ that revenue are 2 different topics. thanks for splitting.