Grace

If any one of the top DeFi lending protocols were to lose just 1% of its TVL, you may lose 100% of your deposit if you’re last to exit. Those who exit early enough would lose nothing. DeFi lending protocols are not designed to handle any loss. They think they can completely eliminate risk when something as simple as collateral price going down too quickly or even debt price going up too quickly can cause bad debt which is fully absorbed by the last lenders to exit. This is a time bomb. It is unsustainable, unfair and heavily incentivizes bank runs. Given an infinite timeline, every lending protocol will lose some amount of money. If not soon then later. Because they know there’s 0 room for loss, they’re forced to be overly conservative in collateral selection, price feed options and risk parameterization. This prevents them from scaling and serving the long tail of collateral tokens. Yet, it never actually eliminates risk. Today, I’m about to publish my solution to this problem. It’s a maximally-resilient cross-margin lending protocol called @GraceProtocol that combines (A) the collateral and capital efficiency of cross-margin with (B) the safety guarantees of isolated margin and (C) the ability to absorb and proportionally distribute loss due to attacks, volatility or any other black swan scenario. It’ll be a long write-up that goes into the details of the complex risk management mechanisms that had to be invented to be able to make this possible. It took a lot of time and effort to get here and to write both the code and explanation so I hope you’ll find some time to go through it. If you’re interested, please turn on tweet notifications on @GraceProtocol to be notified as soon as the announcement is published later today.