Borrow mUSD against imUSD collateral

Summary:

Provide a service where imUSD can locked and borrowed against for mUSD on Polygon for a portion of their yield.

Abstract:

Users could either swap mUSD for FRAX and stake the the FRAX feeder pool or swap mUSD for FRAX and then swap FRAX for DAI/USDC/USDT on another AMM and then deposit that to mSave.

Motivation:

This increases supply of mUSD. This increases the awareness of the mUSD asset as a base asset in the ecosystem. It increases TVL which also draws attention. It also increases Governance Revenue from 1. more swaps 2. more TVL yield and 3. the additional borrowing fee.

Pros:

  • mUSD more ubiquitous
  • TVL increases
  • Governance revenue increases

Cons:

  • smart contract risk
  • the boosted yield over just lending to AAVE would dwindle
  • many users already staked to the FRAX/mUSD pool and they would be disadvantaged compared to other users who can earn both part of their save yield and the pool rewards.

Is the idea that we would allow users to borrow mUSD in Save with their imUSD?

1 Like

Yes. That is exactly the idea I am proposing here. My original hope was to increase TVL and Governance Revenue. Although there is the possibility that TVL would go up because users could convert mUSD to FRAX then to USDC/USDT/DAI on Sushi and then redeposit to mStable, I have since posting this realized two issues. 1. The boosted yield rate on mSavings from mUSD not in mSAVE would be driven down because users could instead place in mSave and borrow against it thus there would be very little mUSD in the ecosystem that was not borrowed against it (this already makes it dead in the water IMO), furthermore this would introduce constant downward pressure on the mUSD peg. 2. This one is even worse, but a user could place mUSD in mSave and start earning yield and then borrow mUSD against that and redeem the mUSD for a base asset effectively leeching off of the yield of other users (I hope this last point is clear because it is relevant for using Rari Fuse pools also). This has led me to the conclusion that mUSD not in mSave is the key for drawing in new users and organic TVL growth. The only two approaches I have come up with today are 1. protocol owned mUSD – rather than bonding as in Olympus pro – part of the Governance Fee could be diverted from Buybacks and put in mUSD/MTA, mUSD/ETH, and mUSD/mBTC pairs. 2. Guerrilla TVL recursive looping: I played around today on Polygon and boosted non-mSave mUSD supply by $1000 at a cost of $2.80 to myself (mUSD β†’ FRAX on mStable, FRAX β†’ DAI on Zapper: Sushi+DODO, DAI back into mSAVE); This could only be done to a certain extent where the arb opportunity wasn’t so great as to draw in oppositional arb reaction.

1 Like

Thanks @Jeshli - some very interesting concepts there. Let me reflect on this for a bit

1 Like

My current proposal would be to allow mSavers to access up to 80% of the amount that they have in mSave at a zero interest rate (or an interest tied to but less than the mSaving rate). But the catch is that they pay an upfront fee. That fee is not hard set but is the loss associated with transferring mUSD to USDT/USDC/DAI through the curve pool and then minting mUSD from the Basket Asset called through the curve pool. So for instance a user deposits $125 mUSD in mSave then can accrue $100 mUSD debt interest free and the user will receive what mUSD β†’ through Curve β†’ USDT β†’ through mStable β†’ mUSD yields. Probably about $99.5 mUSD.

Technically this could be possible with Rari Fuse Pool, supply imUSD and borrow mUSD. However, we should really evaluate whether this is secure. In the ecosystem, there were a few economic exploits that used yield-bearing tokens to manipulate the price. We should really consider and ensure that this would be safe.

Yes. I don’t think using Rari is a good idea. Nor do I think that issuing mUSD against imUSD is a good idea. If you see the point directly above, I feel that the most logical way to do it is to hand the user a stable coin in the form of DAI/USDT/USDC by issuing mUSD and transfering through the curve pool. The logic is that enabling borrowing of mUSD against imUSD would lead to redeeming while imUSD is locked, thus the user would be taking out the asset which generates yield while still making yield effectively leeching yield off of others.