DeFi Lending 2.0: Silo Finance
— By Ibrahim Quabboua
DeFi is not asleep, isn’t it?
Today we will be focusing on one of the most interesting next-gen lending platforms.
Let’s dive in!
Some of the most effective exploits stem from price oracle-related vulnerabilities in borrow/lending protocols. By hackers manipulating oracles and the supply with the weight of a token in a certain pool.
What adds more risk to this is that users have no freedom of choice on what certain risk they are willing to expose themselves to. So when manipulation happens, all is gone.
This is where Silo Finance becomes very interesting as a base layer application, as the project aims to isolate risk to specific lending pairs through their use of “Silos”.
The concept is simple; instead of having all borrowable tokens deposited into one shared pool, the Silo team came up with an elegant design to split each token into its own pool or “Silo”. This isolates risk to a single borrow pair, as opposed to compounding that risk across one huge pool of capital with multiple assets and accepted collaterals. Each token is paired with $ETH in their Silo, and $ETH acts as the “Bridge Asset”, which enables the whole system and allows for exotic borrow liquidity.
The user has $USDC and wants to borrow $CVX, but there is no native lending market, and the User doesn’t want to sell $USDC exposure just to borrow $CVX.
Solution = User deposits $USDC into the $USDC/$ETH Silo and borrows $ETH against their $USDC, User “bridges” borrowed $ETH to the $CVX/$ETH Silo, User deposits “bridged” $ETH and borrows $CVX against it, User can now do whatever they want with that $CVX. Not only is the User able to get $CVX to borrow liquidity against their $USDC, but the risk is also isolated to single pair pools (Silos) and not shared with any other borrowable assets.
It is important to note that you are opening up two positions with this strategy, an $ETH borrow against $USDC, and a $CVX borrow against $ETH, and you will have to monitor two LTV ratios!
Using this tech, technically ANY collateral can be used to borrow any asset, given there is a Silo spun up for the desired asset. (wen $shiba ?)
I told you that silo is next-gen, to prove my point we will do a quick hack simulation.
An attacker convinces the oracle for the $USDC/$ETH Silo that $USDC is $100 USD instead of $1 USD. This gives the attacker the ability to borrow $ETH against $BNT at 100x the real value of the collateral, at which point the attacker can potentially borrow all of the $ETH liquidity in that Silo. This obviously means that both “bridged” $ETH and basic $ETH deposits in that Silo are not safe. $ETH depositors always assume risk.
Although $ETH depositors always take on the risk of loss in Silo, there is a huge unique benefit here:
The protocol lets $ETH depositors have autonomy over the risk profile they assume, as opposed to having handfuls of collaterals whitelisted for a single pool via governance. Since every Silo is a single $shitcoin/$ETH pair, $ETH depositors can pick and choose which collateral type would be hardest to manipulate and deposit their $ETH there, without having to worry about any other collateral being used as an attack vector.
Riskier pools will have less ETH, driving borrowing rate up which might attract some degen lenders.
With all the risk being focused on a certain pool, that will drive a huge volume that seeks safer but lower returns.
Everything with its price, you know.
My take on this is if such an idea succeeded, it will drive a hell of volume from the TradFi money, which is needed because we need exit liquidity for sure.
The market may tumble, and the spirit may get weak.
But DeFi will never ever be asleep.