[Proposal] sAVAX collateral onboarding

[bitn8]

@Max_Resnick

Bridge Risk: The bridge (wormhole in this case) is what keeps the version of sAVAX on Terra, wasAVAX, in 1:1 correspondence with sAVAX on Avalanche. If the bridge is compromised, this ratio may slip. Such a slip would cause Anchor to misprice the value of assets it actually holds, wasAVAX, relative to the true price of native sAVAX. This may cause protocol insolvency in the short term.

Thanks so much for the assessment; I wholeheartedly agree on points 1 and 2 that led us to a similar consensus. We are proud to have Risk Harbors Ozone backing us on this!

As to point 3, we are seriously looking at bridge risk and working with the team and Wormhole to implement solvency checks, similar to what Thorchain has done, as well as possible liquidity caps and other bridge safety measures. The reason we continue to trust wormholes is that they also take security so seriously by backing security breaches with refunds. We are in good hands in this regard!

Looking forward to adding more assets and building a sustainable anchor!

[bitn8]

@jc69

i notice the oracle feeder is a different wallet from the bAsset oracle wallet; just want to confirm that this is intended

yes, this was intended.

@jc69

sAVAX doesn’t seem to follow the bAsset pattern; can we get some background on how anchor will get the sAVAX rewards when locked as collateral?

@bitn8

sAVAX is actually being added without a taking a staking percent. The contracts would have to changed to do that and it would have to have it’s own interest rate, which it won’t. This model is more competitive to the new market space where liquid staking derivatives can be added to any saving and lending protocol very easily. Eventually, bLUNA, which is old and dated will be placed with stLUNA and lunaX. The theory is that borrowing against a LSD will demand a higher borrow rate because of higher utility. You could theorize that it would be something close to the base market borrow rate on the asset plus the staking rate.

[paletas]

Wait, so sAVAX won’t be contributing any yield to the protocol yields (outside of borrowing rates) and will share the same borrowing rate as the remainder collateral? That would make it the ultimate choice for collateral as it’s far cheaper… I understand we’re pushing towards the V2 model, but this is forcing the V2 model in on one collateral type only…

Did I understand this correctly? If so, this should have been clearly stated in the proposal and feels like it wasn’t an oversight.

[bitn8]

Correct you understood it correctly. No one will probably borrow sAVAX at the market rate of staking returns plus base borrow rate. Slowly bLUNA and bETH will be phased out as well and the dynamic borrow rate will be adjusted. If Anchor wants to compete it can’t have some of the highest borrow rates. This why we need a dynamic earn rate as well. We have reached the top of the competition and now need to act accordingly. Nothing is stopping lending protocols taking liquid staking derivatives at this point that are autocompounding. Look at Edge: it already has LunaX as borrow collateral, many more will follow. Anchor needs to be ahead of the curve.

[paletas]

I understand and agree with the statement and direction for the protocol, but this change feels like it’s been pushed in hiding and that I don’t agree with.

This isn’t the proposal for dynamic rates or Anchor V2, this is the onboarding of a new collateral type that was in every way similar to all the others, or at least I don’t remember reading any mention of it being different and assumed it would be equal, it’s my opinion that it should have been clearly stated and wouldn’t have made any difference in the outcome of the vote.

But it does matter as it won’t contribute at all to the yield reserve drain until borrowers start closing down sAVAX collaterized loans.

[bitn8]

Agreed it probably could have been better communicated. Since there were no contract changes that would be needed to make the rate different from the bLUNA bETH rate we assumed users would know we weren’t collecting the staking returns this was clearly mentioned as the new v2 and prop didn’t mention a new rate to take staking returns as part of the sAVAX whitelist. But again to your point, I appreciate you helping clarify this and make it more clear here.

@paletas

But it does matter as it won’t contribute at all to the yield reserve drain until borrowers start closing down sAVAX collaterized loans.

I don’t think I agree with this statement, as it’s going to add more collateral and borrowing to all the UST sitting idle not generating any return right now. Adding more collateral doesn’t drain the YR it at least at min helps buffer it.

[codeNoodle]

@bitn8

I don’t think I agree with this statement, as it’s going to add more collateral and borrowing to all the UST sitting idle not generating any return right now. Adding more collateral doesn’t drain the YR it at least at min helps buffer it.

@codenoodle

Current Anchor Borrow APR is 13.1%, sAVAX earns 7.2%. That is 7.2% of all of their deposited collateral. If their LTV was the average on Anchor right now (~48.7%), then that gives Anchor an overall gain of 13.6% and a 9.7% loss for their position, making it a net positive for the protocol.

gain = 6.4% APR from the borrowed UST (13.1%x48.7%) plus 7.2% from collateral
loss = 20% earn rate on their 48.7% borrowed UST amount, assuming they put all of it in Earn

If sAVAX is not generating any staking yield for Anchor, then every sAVAX depositor could take out the average borrowed amount on Anchor and deposit that into Anchor Earn, thus draining the Yield reserves at ~3.3% APR. Of course that assumes that 100% of borrowed UST goes straight into Earn and the other rates are the same as measured above. 30%+ of borrowed UST would then need to not be deposited into Anchor Earn for the yield reserve drain to break even in this case. Unless I am forgetting something important?

[bat]

Yeah great. waiting for next chain Polygon

[KashHurley]

Do we have any information on how ANC buybacks would work with this and other auto compounding staking derivatives going forward? If nothing heads to the yield reserve to reduce complexity, then wouldn’t ANC governance rewards be MUCH lower, this may hinder veAnc’s appeal also.
Just a thought

[heejn75]

Yes. I agree with you.

[TheThadP]

@codeNoodle

Current Anchor Borrow APR is 13.1%, sAVAX earns 7.2%. That is 7.2% of all of their deposited collateral. If their LTV was the average on Anchor right now (~48.7%), then that gives Anchor an overall gain of 13.6% and a 9.7% loss for their position, making it a net positive for the protocol.

Its pretty safe to assume any collateral deposited into Anchor and the UST generated will be deposited into the vault for aUST. While yes the addition of collateral is good for anchor in isolation, I think we have to accept the realtiy this will continue to drain the treasury until the dynamic earn rate is implemented.