$89 Million Liquidated on Crypto Lending Platform Compound BySherlock GomesPRO INVESTOR Updated: 30 November 2020 LoanScan suggests that around $89 million was liquidated on the crypto lending platform Compound in the past 24 hours after the price of DAI was increased on Coinbase Pro. Third larger COMP farmer liquidates According to Nansen CEO Alex Svanevik, the $89 million liquidation includes $46 million from the third larger COMP farmer. He said, “My understanding is that the DAI price on Coinbase was driven up to a premium of around 30%. Compound’s oracle uses Coinbase for pricing data. This caused liquidations as the value of the loans exceeded collateralization-ratio thresholds.” He added that Compound worked in the way it should ideally have. However, there will be questions about the oracle. What is Compound? Compound is a crypto lending platform that lets users lend crypto assets to each other. It has grown to become the third-largest DeFi platform and has over $1.55 billion locked in its smart contracts. Users need to put collateral for their lending whose value is higher than the amount of borrowing. The loans are issued on the Ethereum blockchain. If the Ethereum network considers that the assets have been undercollateralized, it liquidates the loan. The blockchain doesn’t know the current price of the coins. Therefore, it uses oracles to fetch live prices. A malicious actor may have manipulated DAI price on Coinbase Pro, which tricked the blockchain into believing that the price of the coin has gone up to $1.30. The blockchain then liquidated the loans. According to Svanevik, a $100 loan could have gotten liquidated because the price of DAI was fixed at $1 (it is a stablecoin). With prices rising to $1.30, the loan is now valued at $130. If the user has collateralized their positions at up to $125, then it will get liquidated on the blockchain. Chainlink co-founder Sergey Nazarov criticized the use of centralized oracles that depend on pricing from a single exchange for any activity. He said that this usage is exposing users to multiple risks.