Frax Finance

Authors: Michael Henry

Summary: Propose the creation of a new Fraxlend pair on the Ethereum Mainnet: sfrxETH/sFRAX

Details:

• Fraxlend Pair Name 1: sfrxETH/sFRAX
• Chain: Ethereum Mainnet
• Max Loan-to-Value (LTV): 75%

Background and Motivation: sFRAX is the yield-generating primitive of the Frax stablecoin and can be thought of as Frax’s cost of capital. Therefore, when lending sFRAX via AMO Frax Finance can be assured of profitability regardless of how much money it deposits in the pool.

Positives:

• Guaranteed AMO Profitability: As long as there are no liquidations that take on bad debt, Frax AMO’s will need less managing because FRAX lent into the pool will always be earning more than Frax’s cost of capital
• More liquidity: For lenders, the sFRAX pools can be thought of as restaking lending pools that allow them to return additional yield on-top of the existing yield they’re already generating from sFRAX. As a result, I’d expect a decent portion of the $47m to be deposited as liquidity.
• Cheaper Borrowing: Lower borrowing rates because the borrower doesn’t need to subsidize the entire unlent portion of the pool. All unlent sFRAX is still appreciating in value.

Negatives:

• Reduced Fraxlend Revenue: Fraxlend won’t profit off of the intrinsic borrowing cost associated with sFrax, so if the lending rate is functionally 8% Fraxlend will only earn fees on 8 – sFRAX Rate = 2%.
• Higher Gas Fees: Borrowers will need to unwrap sFRAX for FRAX anytime they want to borrow money.
• Less Clear Borrowing Cost: The real borrowing cost is abstracted from the user but this can be resolved in the future with a minor UI change.
• No Frax AMO Support: Frax AMOs don’t directly support sFRAX but they are largely manually managed, so I don’t think the conversion should be problematic. If this is untrue, please let me know in the comments.
• Frax AMO impact on sFRAX yield: The sFRAX contract does not have functionality to ignore a portion of minted sFRAX when distributing profits, so all AMO Minted sFRAX would receive a portion of profits reducing the underlying sFRAX yield. This could be offset on a 1-week behind basis by adding all profits held by unlent sFRAX back to the sFRAX contract.

Additional Details: The Fraxlend Fees for sFRAX pools can be raised to account for the lower revenue if we want.

Liquidity Profile: sFRAX is always losslessly convertible to FRAX, so it shares it’s liquidity profile. The Max LTV is identical to that of the sfrxEth pool, so that shouldn’t be a concern.

Curve Details: To account for the underlying sFRAX yield, we’ll need to create a new significantly less step Curve. I would suggest a curve where the Borrow limit linearly increases from 0.5% to 4% until hitting 85% utilization, at which point if would mimic the behavior of the Variable 2 curve.

Closing Thoughts:

I believe that in the future a significant portion of debt will be denominated in term of the yield-generating primitive much like real-world loans are denominated in terms of SOFR, so we should try to get ahead of the curve.

Voting:

• For: Approve the creation of the sfrxETH/sFRAX Fraxlend pair on Ethereum Mainnet with new curve, and max LTV
• Against: Do not approve the creation of the new Fraxlend pair.