In response to community feedback on removing incentives for recursive borrowers, this proposal aims to align with the community’s strategic preference by updating the per market distribution of LQ rewards. For the first two months on mainnet set at 50-50 for suppliers and borrowers in both the ADA and DJED markets, if this proposal passes there will be an update to the per market distribution to set it to 100-0, reducing borrow side incentives to 0%. Suppliers in each market will receive 100% of LQ user distribution rewards.
UPDATE: The temp check to update collateral and interest rate parameters (Update collateral and interest rate parameters for ADA and DJED markets) would apply a “penalty” for suppliers with recursive leverage such that any supplier in the ADA market who has used their qADA to open an ADA loan would receive 0 LQ rewards (same for DJED suppliers who open DJED borrows backed by qDJED).
As observed in previous lending protocols using boosted yields to bootstrap liquidity we have seen recursive or wash borrowing drive market utilization to the far end of the curve. This poses serious liquidity risk to the protocol and is not a safe position for a newly launched liquidity protocol to be in for an extended period of time.
We are currently observing just over 40% market utilization in the ADA market, with some community members actively discussing recursive borrowing strategies to capture higher LQ rewards yields in Discord/Twitter. If this proposal passes we expect a reduction in the level of wash borrowing in this market as yields will be significantly higher for supplying in other markets with less supply than the ADA market (e.g. DJED, SHEN markets).
This proposal will apply to the per market distribution ratio starting at the vote end (if successfully voted in). This proposal introduces changes to both markets live on mainnet currently, and to upcoming future market launches (e.g. SHEN).
UPDATE: To @Tron point, the result of the vote will take effect following the closing of the vote, not retroactively. To clarify the only reason it was at the start of the month is due to the monthly epoch currently in use for rewards calculations. However I have confirmed with the developers we can have arbitrary length epochs and they do not need to remain a month long.
Do you support an update to the LQ distribution rate to remove incentives for recursive borrowing?
This change makes almost no difference to discourage wash borrowing.
With this approach, you can still wash borrow and re-supply to earn a larger share of the supply LQ rewards, and it would still be very profitable.
Wouldn’t it be better to disable supply/borrowing loops in the same market?
i.e.:
If you want to borrow Djed, you have to supply ADA (or any token other than ADA once we have more markets).
And if you want to borrow ADA, you have to supply Djed.
I’m happy either way, whether keeping borrow rewards or moving all to supply. That said, I agree with LapinMalin in that my calculations show a little bit lower total rewards for recursive borrowing than currently, but not enough to discourage folding to get a bigger slice of the supply.
However, that said, just the announcement alone that borrowing no longer earns LQ tokens may cause enough psychological reason for a lot of people to unfold and supply only. Not everyone loves spreadsheets and calculations, so they might not fully realize that the mathematical benefit still exists. Just a thought.
It looks like if you wash borrow up to around a factor of 3.7 times your original supply you will lose around 30ish percent of the rewards that you would have gotten in the model used in the first 2 months.
That is a big dent to be sure, but based on the currently available returns across Cardano Defi that I can find the amount that you will earn still dwarfs the returns from other protocols even if the price of LQ drops below 10 ADA.
Those rewards will shift to pure suppliers, so there may be an increase in free liquidity based on peoples defi habits. There are clearly people that don’t have the appetite for wash borrowing, and those people will be receiving the extra rewards.
I would wager this would be a net win for the protocol, and a simple mechanic to apply. Eliminating same pool wash borrowing may technically be more time consuming to track, so this could be a stop gap until the details on how much time that would take the team are available?
Maybe if you use an aquafarm NFT to amortize the loan side regressively. Instead of income agriculture, LQ would be used to pay off debt faster. I think it would be good for the protocol to encourage stability. In that case, perhaps recursive borrowing would be viable.
Removing borrowing using the same token as collateral would be more effective in discouraging wash borrowing as it would add more risk and make people use higher health factors. As the proposal is now, it wont discourage wash borrowing although it will make the rewards lower doing it.
Anyway I voted no just because of the retroactive effect. It should be known in advance if there are changes to the rewards structure so people can react if needed.
How does this point make sense if the rewards are proportional to supplied value?
The rewards for February 2023, March 2023 and the period of April 2023 until the votes is done will receive the LQ according to the original distribution system.
There is no mention of retroactivity, see here below.
This make absolutely no sense at all. No contract has ever been signed with another legal entity that would explicitly make a LiqwidDAO proposal voted on by the entire community not retroactively enforced, it’s also not even being proposed as being enforced retroactively. It surprises me enormously how clueless you are.
