During the last year DeFi has seen a boom in popularity, and one of the most successful sectors has been Automated Market Makers. Today there is around $20B in liquidity provided to AMMs, locked in assets known as LP Tokens, and the amount is increasing each month! As LP Tokens became increasingly popular, more and more users started demanding tools that allowed them to be used as collateral. Impermax and Alpha Homora are two of the pioneer platforms in the new sector of lending based on LP tokens. In this article we will see the key differences between the two and their expected performances.
The main difference between Alpha Homora and Impermax Finance is the architecture. Specifically, there’s a big difference in how lending funds are pooled. This has consequences for all the properties of the two applications that we will see in the rest of this article. While on Alpha Homora the lending pools are shared between all the LP tokens, Impermax has an isolated pairs architecture, which means that each pair has its own lending pools. This is a radical difference that has some important advantages and drawbacks.
Pros of isolated lending pools:
- A user can only borrow the underlying assets with a certain LP token. This allows Impermax to be much more efficient for reaching high leverage positions.
- It’s always possible to borrow all the underlying assets of a certain LP token.
- A lender can decide which pair to lend to in order to stay away from bad LP tokens.
- Isolated pools add safety. An attack on a certain pair can’t affect all the other pairs.
- Anyone can add a new pair. This makes Impermax more permissionless.
- A virtually unlimited number of tokens are supported for both borrowing and lending.
Cons of isolated lending pools:
- A user can only borrow the underlying assets with a certain LP token.
- Newly created pairs need more time to reach ideal economic incentives since borrowers need to wait for lenders to join the new lending pools.
Alpha Homora uses the classic loan-to-value model which establishes that a borrower with a certain collateral can take a loan of any assets available up to a certain value defined by a ratio. This ratio is pre-defined for each type of collateral.
One of the main innovations brought by Impermax is the dynamic collateralization model that allows optimal collateral efficiency for LP tokens. Impermax recognizes that the loan-to-value model becomes obsolete when the value of the collateral and the borrowed assets is somehow correlated. For LP tokens leveraging, this correlation is very high because the collateral (LP tokens) is backed by the same assets that the borrower is borrowing. The Impermax collateralization model takes this into consideration and calculates a dynamic loan-to-value that depends on which assets the user is borrowing and in which ratio. This collateralization model allows Impermax to increase the max leverage without increasing the liquidation risk.
A practical example
An high liquidity pair like ETH-WBTC on Alpha Homora could have parameters such as 60% work factor and 80% kill factor. These parameters mean that the user can take a loan worth at most 60% of the collateral that he has, and that the loan will be liquidated once its value goes over 80% of the collateral value. With these parameters, the user can enter in a 2.5x leveraged position by taking a flash loan of ETH worth 150% the collateral deposited, converting half of it to WBTC, and converting the newly obtained ETH and WBTC in the ETH-WBTC LP to be deposited as collateral to cover for the loan taken. The position will be liquidated once the value of the loan grows by 33% in comparison to the collateral (from 60% to 80%). This will happen once the value of ETH grows by 78% (1.33²) in comparison to WBTC. On Alpha Homora a user can enter in a 2.5x leverage with a 78% price swing tolerance.
Now let’s see how much leverage a user could achieve with the same price swing tolerance on Impermax. First of all, since Impermax doesn’t use the work factor and kill factor parameters, we need to find out the equivalent parameters to those. Impermax uses the parameter called “safety margin” which indicates how much a price can swing after a loan becomes liquidatable before it will not be covered by the collateral. On a “low volatility” pair like ETH-WBTC, a conservative parameter for the safety margin would be 1.5, which means that the price can swing 50%.
With these parameters, and considering a 78% price swing tolerance to liquidation, a user on Impermax can deposit his ETH-WBTC LP, take a flash loan worth 300% the value of the collateral of ETH and 300% of WBTC, convert them to ETH-WBTC LP and deposit it as collateral. By doing so the user has effectively entered in a 7x leveraged position and he won’t be liquidated until the WBTC-ETH price swings by more than 78%. This is possible because the loan taken by the user is considered optimally collateralized since it is composed 50% by ETH and 50% by WBTC, which happens to be the same ratio of ETH and WBTC contained in the ETH-WBTC LP.
In conclusion, with similar parameters and the same tolerance to liquidation, a user on Alpha Homora can enter in a 2.5x leveraged position while a user on Impermax can enter in a 7x leveraged position.