It’s not hard per se, but requires some work, e.g. testing to verify that our adapters work correctly with new pools (even seemingly similar contracts can have subtle differences that break the adapter). We don’t have capacity for that right now
Thanks @Van0k
Added tokens to the proposal draft. Can you please double check it and let me know if I missed something?
Do we also need to vote on the list of the contracts/adapters btw?
Thanks so much for clarification, I’m an idiot and messed up the yearn article.
Actually was re-reading intern’s article and noticed the: yVaults for USDC, DAI, and WBTC, and ETH… my question is WHY. The borrowing APY would always likely be higher than whatever historically those vaults have been earning. Those specific Yearn vaults were always very experimental (in my opinion) as they pack strategies from 5+ protocols at a time. I would not add yVault for USDC, DAI, ETH, and WBTC. No point.
I mostly agree, but it is not just for farming:
If you want to go long on BTC or ETH you can buy yvWETH or yvBTC on a USDC or DAI CA to approximately cover the borrowing fees.
If you want to go short you could use yvUSDC or yvDAI on a WBTC or WETH CA for the same reason.
Should be on the choice of the users imho. But well, I wouldn’t consider much of that either. yvDAI is ok, but the rest is clearly below <1% while adding extra gas and extra risk …
2 Likes
Ah this is actually VERY interesting I didn’t think of that… damn!
LGTM, I would also mention that LDO and LQTY have no USD oracles and will always be price at zero (but are still tracked by Gearbox so they can be sold or sent to user on account closing)
@amantay I see yvCurve-stETH has higher LTs than steCRV for all but 1 one pool? I would’ve expected them to be the same?
Updated the list of pools/tokens.
Thank you fellow gearheads!
1 Like
Proposing to add wrapped Coinbase staked ETH (address 0xBe9895146f7AF43049ca1c1AE358B0541Ea49704) to the tokens AllowedList for Gearbox v2.
Coinbase is one of the most credible and trustworthy custodians who recently launched their liquid ETH staking derivative token that can be freely traded on-chain (currently at a discount of c. 5% to ETH). A number of strategies relevant for Lido’s staked ETH derivative, stETH, have been extensively discussed and marketed externally; most of these strategies, incl leveraged staking, appear similarly relevant to competing products.
This proposal assumes that the idiosyncratic risks associated with the token contract (contract upgradeability, pausing and blacklisting enabled) are manageable; the LTV thresholds could, for example, be set to more conservative levels to account for this higher degree of centralization and counterparty risks.
Actually, ye! But does it made sense to have it as an asset to trade into, if the farms / pools for it have not been made & audited yet? What do you foresee as strategies, indirect leveraged staking only?
from my point of view the main problem here is in low on-chain liquidity of cbETH. so in the case of depeg it can be stuck as bad debt…
2 Likes
Mr. @RV_ivangbi is soliciting input regarding using $CVX, $CRV $FXS etc. as collateral in his latest tweet storm so thought I’d throw my 2 cents in.
Would I use $MKR $LDO $CVX $FXS $SNX $SUSHI $YFI $COMP $CRV as collateral and lever up? Well, it depends on what I’m borrowing. If I’m borrowing ETH or wBTC; No. Exposure to two volatile assets at high leverage isn’t a risk I’m comfortable with. USDC; maybe, but only at low leverage, i.e. 2 or 3x max but only if there was high yielding stablecoin pools available at that low lev. If pools were < 10% APR then factoring in tx fees etc and risk, probably not.
However, if it was possible to borrow the same asset used as collateral, i.e. deposit CRV and borrow CRV at 4-6x then yes; certainly would use this. I’d be throwing the CRV into the cvxCRV pool (if it was whitelisted of course) and FXS in the cvxFXS pools etc. etc.
My opinions on each as a form of collateral:
$MKR - Don’t own any, nor intend to. Nil interest in it as collateral.
$LDO - Don’t own any, nor intend to. Nil interest in it as collateral.
$CVX - Own reasonable amount but it needs to be put to work. If returns can be > than bribes etc. then LFG.
$FXS - As above.
$SNX - Don’t own. Nil interest.
$SUSHI - This is nearly dead, no? I believed but was burned. No return to the sushi-fam unless there’s real change (and can the new fella do it? I don’t know; we’ll see). No interest.
$YFI - Don’t own. No interest as collateral.
$COMP - As above.
$CRV - Own but use in Gearbox needs to beat bribs and LP-ing returns and make up for potential opportunity of CRV for use in minting $crvUSD (if it’s in fact needed?).
Appreciating or value-deriving tokens would be better forms of collateral so whilst they remain volatile they’re trending in the right direction, i.e. up.
Just my thoughts.
2 Likes
Been discussing with @0xmikko and @apeir99n the topic last few days and we decided to make some minor changes. I cannot seem to edit my previous posts for some reason, so, below is the last updated version for the upcoming proposal. The changes are - wstETH pool instead of stETH, due to wrapped version is more applicable for integration and minor changes of some of the LTs:
Motivation
Gearbox V2 has passed external audits and is about ready for deployment on Mainnet.
V2 includes integrations with Lido, Curve, Convex, and Yearn protocols and allows new leveraged strategies for high-yield farming and trading. Those protocols have well-established processes, reliable infrastructure as well as strong community support.
The V2 strategies involve assets that can be leveraged using Curve, Convex, and Yearn:
All of the assets conform to main risk requirements: deep DEX liquidity, the existence of Chainlink price-feed for underlying tokens, and a viable tokenomic model.
DEX Liquidity can be tracked using the risk-committee report: Gearbox Credit Accounts
Chainlink price-feeds: Decentralized Data Feeds | Chainlink
To mitigate the risks of malicious exploitation, V2 will be gated on the first phase using the set of whitelisted addresses (Leverage Ninja mode). To bootstrap TVL, whitelisted participants have committed to a minimum $50k entry.
Given V2 is around the corner with a tentative ETA soon after the Merge, it’s crucial to consider all the associated risks and decide on the Allowed List policy for V2. The new parameters aim to give greater leverage ability for the less risky collateral assets as well as to reduce leverage for the more risky ones.
Proposal
Taking into account V1 experience and expectations from the upcoming V2, I propose changes as below:
- Create a new pool: wstETH 0x7f39C581F595B53c5cb19bD0b3f8dA6c935E2Ca0 with the same parameters as for the ETH pool. This would help to attract liquidity from stETH holders and also allow less risky sTETH borrowing and bigger leverage when using stETH as collateral.
The pool parameters have to be adjusted to satisfy the expected demand during gated Leverage Ninja mode.
- Set up Asset LTs as below for V2 assets:
| Token | USDC pool (%) | DAI pool (%) | WETH pool (%) | wstETH pool (%) | WBTC pool (%) |
|---|---|---|---|---|---|
| WETH | 85 | 85 | 94.5 | 90 | 85 |
| stETH | 82.5 | 82.5 | 90 | 94.5 | 82.5 |
| WBTC | 85 | 85 | 85 | 85 | 94.5 |
| USDC | 94.5 | 92 | 82.5 | 82.5 | 82.5 |
| DAI | 92 | 94.5 | 82.5 | 82.5 | 82.5 |
| USDT | 90 | 90 | 82.5 | 82.5 | 82.5 |
| sUSD | 90 | 90 | 82.5 | 82.5 | 82.5 |
| FRAX | 90 | 90 | 82.5 | 82.5 | 82.5 |
| gUSD | 90 | 90 | 82.5 | 82.5 | 82.5 |
| LUSD | 90 | 90 | 82.5 | 82.5 | 82.5 |
| steCRV | 82.5 | 82.5 | 90 | 90 | 82.5 |
| cvxsteCRV | 82.5 | 82.5 | 90 | 90 | 82.5 |
| stkcvxsteCRV | 82.5 | 82.5 | 90 | 90 | 82.5 |
| FRAX3CRV-f | 90 | 90 | 80 | 80 | 80 |
| cvxFRAX3CRV-f | 90 | 90 | 80 | 80 | 80 |
| stkcvxFRAX3CRV | 90 | 90 | 80 | 80 | 80 |
| 3Crv | 90 | 90 | 80 | 80 | 80 |
| cvx3Crv | 90 | 90 | 80 | 80 | 80 |
| stkcvx3Crv | 90 | 90 | 80 | 80 | 80 |
| LUSD3CRV-f | 90 | 90 | 80 | 80 | 80 |
| cvxLUSD3CRV-f | 90 | 90 | 80 | 80 | 80 |
| stkcvxLUSD3CRV | 90 | 90 | 80 | 80 | 80 |
| crvPlain3andSUSD | 90 | 90 | 80 | 80 | 80 |
| cvxcrvPlain3andSUSD | 90 | 90 | 80 | 80 | 80 |
| stkcvxcrvPlain3andSUSD | 90 | 90 | 80 | 80 | 80 |
| gusd3CRV | 90 | 90 | 80 | 80 | 80 |
| cvxgusd3CRV | 90 | 90 | 80 | 80 | 80 |
| stkcvxgusd3CRV | 90 | 90 | 80 | 80 | 80 |
| FraxUsdc | 90 | 90 | 80 | 80 | 80 |
| yvDAI | 90 | 90 | 80 | 80 | 80 |
| yvUSDC | 90 | 90 | 80 | 80 | 80 |
| yvWETH | 82.5 | 82.5 | 90 | 90 | 80 |
| yvWBTC | 82.5 | 82.5 | 80 | 80 | 90 |
| yvCurve-stETH | 82.5 | 82.5 | 90 | 90 | 82.5 |
| yvCurve-FRAX | 90 | 90 | 80 | 80 | 80 |
| CVX | 25 | 25 | 25 | 25 | 25 |
| FXS | 25 | 25 | 25 | 25 | 25 |
| LQTY | 0 | 0 | 0 | 0 | 0 |
| CRV | 25 | 25 | 25 | 25 | 25 |
| LDO | 0 | 0 | 0 | 0 | 0 |
| SNX | 25 | 25 | 25 | 25 | 25 |
LTs (Liquidation thresholds) above address the demand for higher leverage on base assets and stablecoins and reflect the absence of interest on the majority of the tokens in V1 so far. It is expected that CVX, FXS, LQTY, CRV, LDO, and SNX would be used as farming rewards rather than leveraging trading on them, therefore high LT values are unnecessary. This can be reconsidered later if there will be proven demand for higher leverage.
@amantay seems you missing FraxUsdc pool’s (both for Curve and Convex) address in the 1st table list…
1 Like