Any marginal emission to Lifinity has very likely more positive social externalities than to borrow-lends.
Other protocols could employ a similar strategy (through POL or DAO-voting on gauges). That’s actually in the very best interest of Marinade DAO!
Lifinity liquidity is pulled when markets are volatile, which is an unfortunate coincidence since that’s exactly when liquidity is most-needed to avoid bad liquidations on lending protocols. mSOL holders who borrow have a higher risk of wrongful liquidation than when liquidity is on a classic AMM.
This is inaccurate. Lifinity never “pulls liquidity”; it adjusts prices more quickly and intelligently than other DEXs. Quoting myself from a post on Marinade’s Discord (since the same point was being raised):
I understand that Solend cares most about liquidity during liquidations, which tend to occur when price is volatile, but that’s also when Lifinity tends to be ahead of the curve in terms of correctly pricing assets. Other DEXs will tend to offer “better” prices for liquidations (but not for the other way – Lifinity will likely have the best price for the other direction), but that’s because they have no efficient way of pricing their assets and mitigating IL. So Lifinity may not offer good liquidity for liquidations, but that is merely a single use case for trading. For example, some people want to buy the dip or take profit rather than trade with the momentum (as liquidators do), and Lifinity will often offer them the best price.
Sorry but not sorry for avoiding IL 
But its not doing zero volume nor is this ever likely.
The volume it does is by in large when others have no offer so its providing an essential service of tightening spreads, hence attracting more people to hold mSOL over SOL. I know this from my own personal experience.
Please view a draft of the proposal, and hear comments before an edited version goes to vote.
I propose the following:
This would remove the current Lifinity liquidity gauge where deposits are limited to only Lifinity, and prevent the creation of future closed off or permissioned liquidity gauges.
“One of Marinade core value is decentralization, including the spreading of MNDE ownership around the ecosystem, so we should not allow Liquidy Mining Gauges for large pools that are private by design of by lack of community participation, because that leads to a feedback loop of MNDE concentration rather than decentralization”
This was eloquently put by Lucio, and I think serves as a good summary of the “why” of this proposal.
This proposal’s goal is to restore liquidity gauges to it’s original purpose:
“MNDE Liquidity mining serves 2 functions in Marinade:
And does not serve as a targeted attack towards Lifinity. I highly suggest opting for a volume-based or other goal setting for Lifinity’s mSOL-USDC pool to receive additional MNDE. This removes the potential of a feedback loop (Lifinity constantly receiving MNDE and relocking it for more MNDE rewards, leading to exponential growth) and focuses on the value-add of the mSOL/USDC pool.
This aligns mDAO and Lifinity’s interests better than the current model. I note that this was what Lifinity initially proposed, and should have no problems pivoting towards now. Under the current model, Lifinity’s gauge receives 10% of MNDE incentives, regardless of volume or results.
I am not arguing that Lifinity is -ev for MarinadeDAO. I am for volume based goals, and a way of quantifiable goals (volume, spread, execution) to be presented and assessed, instead of Lifinity utilizing its MNDE (that it got as a grant) to drive more MNDE to themselves.
The removal of the Lifinity gauge should see more incentives driven towards other AMMs (Orca and Raydium) and other DeFi platforms. While also keeping (hopefully) the liquidity/value add that Lifinity provides. (Once again, I highly suggest a pivot to volume goals instead of shutting the pool and dumping MNDE).
Under a more aligned model, mDAO benefits from Lifinity in a measurable manner, while also encouraging other DeFi protocols to participate. This should see more mSOL TVL growth, while maintaining the benefits Lifinity brings. More MNDE has to come from somewhere else to reward Lifinity, but mDAO won’t be overpaying for Lifinity’s liquidity as much as they are now and discouraging more protocols from acquiring MNDE.
There are a few potential risks with the proposal:
Under the proposal, we have Hao from Hedge/Nazare, and Rooter/Nope fromthe Solend team commenting. It would be ideal to see more DeFi protocols commenting here to ensure that the change is seen favorably. Lifinity has been an active participant in mDAO and would benefit from a pivot in reward mechanism. For the sake of keeping this proposal focused, I suggest we bring up the Lifinity’s volume goals in a separate proposal.
Edit disclaimer: the following post was sent before I saw the proposal above. Will create a new post with my reply to the proposal draft
–
Seems like discussion is starting to go around in circles and lots of arguments and bs (“avoiding IL” LMFAO) has been exchanged and repeated, either because people are straight up not understanding underlying concepts or are putting their personal interests over mDAOs because of their exposure to LFTNY.
I am yet to see any reason -yet any compelling reason- why a vehicle to exponentially grow someones MNDE for free needs to exists, which will just end up making MNDE liquidity gauges virtually useless.
It is obvious a lot of voters weren’t aware of the implications in the original proposal, which let to this -frankly- mistake of passing the original proposal. mDAO needs to decide whether and how this mistake can be corrected.
While also keeping (hopefully) the liquidity/value add that Lifinity provides.
Just so there’s no confusion later on, if the gauge is removed Lifinity will consider itself absolved of its duties (such as providing liquidity) since the terms will have been violated. It goes without saying that a one-time payment for perpetual liquidity is unreasonable.
It would be great if a volume-based compensation method was added to this proposal. Last time we suggested that it was rejected, so we don’t know what is workable and what isn’t. Perhaps team members can tell us what is feasible and what isn’t?
Wouldn’t this apply to any LP tho? Once someone deposits into the mSOL-USDC pool they will get MNDE emissions regardless of whether there was any volume going through that pool.
The emissions are to attract LP to the pool and compensate the depositor for IL risk. Lifinity is holding mSOL for the pool and is subject to IL risks regardless of volume.
Someone mentioned that a LFNTY holder doesn’t care if the mSOL-USDC pool gets wrecked. Of course they care, that pool is part of the operating income of the protocol and any reduction in value will directly affect the value of the protocol and the rewards distributed to LFNTY holders.
Very fair.
Can you help me here?
We can calculate how much MNDE Lifinity has received beyond the initial 2M MNDE, and based on Lifinity’s old volume proposal, how much “volume” has that bought.
I can add to the initial proposal that Marinade will pay out additional MNDE incentives, if Lifinity brings in more volume (- the MNDE Lifinity has already received) while waiting for a new proposal or something to be drafted and implemented.
Example, if Lifinity has received 300K MNDE, based on the old rate of “・Marinade compensates Lifinity with 0.05% of the total volume generated on mSOL-USDC with a cap of $100M per month”, Marinade has already paid for X volume, so Lifinity will receive more MNDE when >X, until the new proposal is ready and finalized.
I want to get clarity on something.
Suppose Lifinity were to open its mSOL-USDC pool so that anyone could deposit liquidity and earn MNDE rewards. Lifinity would then function like any other protocol with respect to the gauges.
Would this solve every problem brought up?
Do you mean calculate how much MNDE in $ we were paid for the volume we’ve generated until now to determine the % of volume we’ve been paid historically?
I think what first needs to be figured out is where that MNDE comes from. It was on this basis that the original proposal was rejected (don’t want to have an ongoing payment separate from the LM gauges).
I think it would, yes (personal opinion here). Other protocols are also free to acquire mSOL/XXX and provide liquidity to receive MNDE incentives, so even if Lifinity were to be the main provider of liquidity, I feel it would solve the issue of having “closed pools” concentrating the MNDE incentives towards one wallet that can use this full power to create this self-reinforcing loop.
I’d like to thank everyone for staying (almost) civil and debating this issue in the open, it’s a real pleasure to see governance decisions being discussed like this.
The whole point to give Lifinity Liquidity Mining rewards is so that I can increase Lifinity’s governance influence Marinade so they can better direct the growth of the protocol vs the current community models that were discussed before.
I am saying the prior use of Marinade was limiting its potential so I want to change that and Lifinity is a good ally to help move Marinade to the next level.
I… dont even know what to say. This is outright malicious towards mDAO, and I believe (and hope) @Durden doesnt agree with this.
thanks @Durden for the response,
I mean calculating how many MNDE lifinity would have received under the volume-based compensation, and Lifinity repays MNDE, or generates volume till the MNDE received from liquidity gauges is the same as what Lifinity would have gotten from volume-based compensation is the same.
Or if Lifinity would have gotten more from volume-based compensation, receive the difference.
I believe (and hope) @Durden doesnt agree with this.
Tbh not even sure what the sentence is trying to say.
Thanks for rephrasing, the meaning is clear to me now 
That of course makes sense and happy to do that if we move to a volume based model. But again probably more important to determine where the ongoing emissions would come from first.
This fails to take into account the MNDE rewards that were generated from MNDE token holders that voted to increase the gauge weighting of Lifinity’s mSOL/USDC pool.
Are you saying that those rewards should be subject to some sort of penalty? Seems counterintuitive to tell a community to use the gauges, and then when they buy/lock MNDE punish them for doing so.
Even if you oppose the current gauge setup, and think a change is necessary, you shouldn’t punish Lifinity, or its token holders (who are its LPs), by reducing future rewards to pay for the “excess” rewards that Lifinity’s pool has already received.
He’s trying to imply use of governance to advance other protocols you have a stake in is “bad” and that you aren’t “bad” like that evil lollipop.
Very fair! Let’s modify it to only include MNDE received as a direct result of Lifinity’s voting.
I suggest we include the number into the original proposal and vote on it! If mDAO wants to remove lifinity gauge, mDAO needs to provide a reasonable alternative. Moi favor
I would like to add my part on Lifinity’s incentives to pump the trade volume. As mentioned above, Lifinity pools currently charge a 0.08% fee. I looked at their docs for a bit, and found the formula to LP’s share of the fees, max(0, 1 - ((P+L)/T)^i).
Given the current mSOL-USDC liquidity of ~930k (P+L) and target liquidity of 500k (T), that gives us a (P+L)/T coefficient of 1.86, which means the incentivisation factor (i) would need to be -ve for LPs to get a share of the fees.
Given how opaque Lifinity is, I’m not really sure about what the value of i is, or how much liquidity comes from Flares (which always gets an 85% share)
In the worst case scenario which most of the trading fees goes to Lifinity itself, I can see that they have an incentive to wash trade (i.e. artificially pump trading volume) in order to boost their metrics. I would go as far as to suggest that Marinade halt all emissions to Lifinity and/or any POL protocols until the numbers are publicly verifiable.
A lot has already been said, so I don’t feel like adding my 5 cents to it.
What I’d like to highlight though, is, that I’m strongly opposing a vote that would disallow “private pools” before or unless including a new agreement with Lifinity (which would also require their holders approval before).
We voted for that agreement, and even though @durden - thanks a lot!!! - is very supportive here, just breaking terms feels absolutely wrong in my opinion.
Wow long thread, but trying to give some more color to my thoughts in response to this and what else have been said.
Under the proposal, we have Hao from Hedge/Nazare, and Rooter/Nope fromthe Solend team commenting. It would be ideal to see more DeFi protocols commenting here to ensure that the change is seen favorably.
Some thoughts from my side partially in response to previous comments on this thread:
I believe the reality of the current setup can lead to behavior that is unbeneficial to Marinade, e.g. Solend, Orca, etc. are incentivized to withhold MNDE rewards granted to them through the gauge from their users to compete with Lifinity’s voting power. I absolutely buy that this result was not a malicious intent by Lifinity, but that doesn’t change what the situation is.
At the same time I agree that It’s unreasonable to expect closed AMMs to provide liquidity in perpetuity without ongoing incentives / compensation.
However, figuring out what an appropriate compensation is, is not easy. E.g. a volume based sort of compensation could incentivize a closed AMM to just washtrade in order to create that volume (not implying they would do that, just highlighting how incentivizes might be misaligned)
I agree with Nope and Rooter, from Marinade’s perspective it’s a question of where/how can I get what we want in the cheapest way. That also needs to be kept in mind when discussing what a “fair” compensation for a closed AMM’s service might be.
Some stray not fully thought through idea, but what if ongoing “compensation” for closed AMMs would be billed based on a certain metric and the cost per unit is calculated based on how much MNDE it costs on other DEXes to achieve the same unit.
I’m aware coming up with a sensible metric is challenging and needs a lot of thought as well, but this might be a framework to approach this