Thread by @blockworksres

AMMs are at the root of everything elegant and useful in DeFi.

That's why @CurveFinance put LLAMMA at the center of its lending protocol.

@smyyguy breaks down this novel mechanism👇

LLAMMA and crvUSD are complex and require a lot of explaining, so we are breaking this into two threads:

1. LLAMMA soft liquidation today
2. crvUSD monetary policy tomorrow

1/

Nearly all DeFi lending today is overcollateralized, where protocols rely on "hard liquidations" to remain solvent when the value of the collateral declines.

2/

While the design protects protocols from bad debt, it forces borrowers to realize losses as collateral cannot be recovered post-liquidation, even if the value recovers.

Curve set out to improve this, derisking the process for the protocol and improving borrower UX.

3/

Insert LLAMMA, the lending-liquidating AMM algorithm.

By replacing a traditional lending pool with an AMM, Curve introduces "soft liquidation" that 𝐠𝐫𝐚𝐝𝐮𝐚𝐥𝐥𝐲 liquidates collateral over a 𝐩𝐫𝐢𝐜𝐞 𝐫𝐚𝐧𝐠𝐞 instead of a sudden liquidation at a specific price.

4/

To open a loan, borrowers deposit collateral into a LLAMMA and receive debt in Curve’s stablecoin, crvUSD.

The collateral is spread across a user-specific set of “bands," creating the “liquidation range.”

The # of bands controls the spread of collateral across the range.

5/

Loan A has fewer bands, so it has a more dense range that begins at a lower ETH price (1675) than Loan B (1750).

Also, the lower bound of Loan A’s range (1575) is at a higher ETH price than that of Loan B (1550), meaning it will reach full soft liquidation before Loan B.

6/

The price offered by a LLAMMA pool does not strictly rely on the balance of the assets in the pool. An oracle price influences the AMM price.

For example, if the oracle price increases/decreases by 1%, then the LLAMMA price will increase/decrease by 𝐚𝐭 𝐥𝐞𝐚𝐬𝐭 1%.

7/

The design creates arbitrage opportunities to incentivize third parties to interact with the pool, which in turn operates the soft liquidation process.

The use of an oracle is required, and the protocol currently leverages a Chainlink-based solution... for now 😉

8/

To remain solvent during collateral price declines, LLAMMA pools need to soft liquidate collateral for crvUSD.

The LLAMMA price decreases more rapidly than the oracle price, so Curve’s lending pool sells collateral for less than the current market price.

9/

Arbitrageurs can buy collateral from the LLAMMA with crvUSD and sell it on the market, capturing the spread between the market price and the LLAMMA price.

The arb transaction converts the collateral into crvUSD, soft liquidating borrowers with collateral in the active band.

10/

When the price of the collateral later increases, LLAMMA deliquidates borrowers by incentivizing arbitrageurs to sell collateral back into the pool.

As the oracle price increases, the LLAMMA price increases at a faster rate, so LLAMMA buys collateral above the market price.

11/

Arbitrageurs now buy collateral from the market and sell it to the LLAMMA pool, capturing the spread between the LLAMMA price and the market price.

The arb transaction converts the LLAMMA’s crvUSD back into collateral, returning capital to borrowers.

12/

The animation below walks through the lifecycle of the soft liquidation process for a crvUSD loan against ETH collateral.

13/

While arbitrageurs play a critical role in maintaining the health of the protocol, each transaction erodes borrower collateral.

In other words, borrowers must pay a small price for the potential ability to recover their collateral after being soft liquidated.

Blockworks Research subscribers have access to the full report that explains how LLAMMA functions in significantly more detail.

Remember, this is just Twitter.

blockworksresearch.com/research/curve…

Tomorrow's thread will walk through the protocol's monetary policy and discuss stability pools, peg keepers, and the interest rate calculation.