Stablecoins and their so-called “stability” have been hot topics since the massive blow up of UST and Luna, leaving many investors questioning whether they can feel safe holding any sort of USD pegged asset. Understandably, the trust in algo-stables has died down greatly. Tron’s USDD hovered below $0.98 for 12 days in June and even USN (Near’s native stablecoin) will transition to a v2 where it is fully collateralized by USDT (Coinmarketcap, 2022; Decentral Bank, 2022).
Fiat-backed USDC and USDC, on top of being the largest and most popular, are the most stable iteration of stablecoins. However, these stablecoins are issued by centralized authorities, which is counterintuitive with DeFi’s desire to be completely permissionless and decentralized. While MakerDAO’s DAI is often touted as the overcollateralized and decentralized solution, most of its backing is in USDC (de Best, 2022). Liquity and its stablecoin, LUSD, aim to be the solution to both stability and complete decentralization.
Liquity is a decentralized borrowing protocol that allows users to draw interest-free loans against Ethereum as collateral (Liquity Docs, 2021). Unlike money markets such as AAVE or Compound, users do not loan out their collateral in order to borrow. Instead, the assets are simply locked in smart contracts until repayment. To begin, a user opens a “trove,” which acts as a siloed vault for borrowing the protocol’s native stablecoin, LUSD. Users borrow LUSD as debt against their collateral. For each trove, the collateral ratio (collateral value/debt value) must stay above 110%, meaning for each $110 of ETH deposited, individuals can borrow a maximum $100 of LUSD. With a circulating supply of over 177m LUSD, the stablecoin is currently the 14th biggest stablecoin in crypto by market cap (Coinmarketcap, 2022). The only stipulation for borrowing is that, in order to open a trove, users must borrow more than 2,000 LUSD. Individuals looking for exposure to LUSD below the $2,000 threshold can purchase LUSD from popular centralized or decentralized exchanges.
Since Liquity has no governance, it cannot charge a voted-upon variable interest rate for borrowing. Instead, the protocol charges an algorithmically determined borrowing fee each time a user draws down LUSD from their trove. This fee is charged as a portion of the LUSD loan taken out and fluctuates between a minimum of 0.5% and a cap of 5%. The rate increases with every redemption and decays over time as long as no redemptions take place. Redemptions are different from loan repayments because they have no fees associated. Redemptions are a peg stability mechanism, enabling arbitrageurs to exchange 1 LUSD for $1 worth of ETH. The redeemers also face the same fee percentage as borrowers when they redeem, except their fee is taken out from the ETH they receive back.
The borrowing fee’s dependence on redemption volumes directly helps to disincentivize borrowing after large redemptions. Since redemptions imply that LUSD is trading below peg, increasing the cost to borrow is the most effective method for reducing supply expansion to stabilize the market price.
As a result, this one-time fee mechanism favors users who wish to take out long-term loans, given they can hold LUSD for as long as they want without incurring any marginal interest fees. Over a five-year horizon, LUSD borrowers may face just one 0.5% flat borrowing fee, while an individual who borrows USDC on AAVE would face a 10% total interest charge, given an average 2% annual interest rate.
Liquity is the most capital-efficient borrowing protocol given its low minimum collateral ratio requirement of 110%. Liquity’s biggest competitor, MakerDAO, offers multiple vaults for minting their stablecoin, DAI, but requires higher minimum collateral ratios, ranging from 170% down to 130% depending on the interest rate charged (Oasis Borrow, 2022).
Theoretically, low collateral minimums set up users' deposits to accrue bad debt if the collateral takes a nosedive. In crypto, there is nothing worse for a stablecoin issuer than facing insolvency because they have more liabilities than assets. However, Liquity’s automatic liquidation engine allows it to maintain a healthy protocol while also providing one of the cheapest forms of leverage within all of DeFi.
Liquity relies on two unique protocol designs to ensure automatic liquidations whenever troves’ collateral value drops below 110%: Stability Pool and Liquidation Reserve Charges.
When troves need to be liquidated, the ***stability pool ***acts as a first line of defense to cover the debt position. Users can deposit their LUSD into the stability pool, which uses the deposits to pay the debt of liquidated troves in return for the ETH collateral. Over time, the LUSD deposits are burned with each liquidated loan. This enables depositors to earn a greater nominal value because the troves are liquidated at just under the 110% collateral ratio, leaving a 10% spread to the stability pool. In addition to the 10% liquidation profit received in ETH, depositors also earn LQTY, the native token that can be staked for capturing, borrowing, and redemption fees generated by the protocol. Having this liquidity backstop for liquidations is great for earning users yield. More importantly, the liquidity backstop is far more efficient than lengthier liquidation auctions that competing protocols rely on because users are not required to bid to liquidate debt positions. Under the unlikely circumstance that the stability pool is drained and a trove needs to be liquidated, each borrower will receive a portion of the liquidated collateral and increase their debt position as a part of the redistribution process.
Offering rates between 7-20% APY on average, the stability pool is the most popular destination for LUSD, with 48.5% of the entire LUSD supply deposited there (Dune Dashboards, 2022; Etherscan, 2022). Understandably, the potential for 10%+ yields on a stablecoin, compared to 2% on competing stablecoins via AAVE, has drawn many users to mint LUSD for the sole purpose of depositing it right back into the stability pool.
Whenever a trove is opened, there is a 200 LUSD liquidation reserve charge added to the user’s debt position to cover gas costs in the event of a potential liquidation. If a user avoids liquidation and completely pays back the loan, they receive the funds back after closing out the trove. The locked funds for gas costs ensure that liquidations can be completed immediately, even in the case of high costs and at zero cost to the stability pool depositors.
Liquity is no exception to the inherent risk in DeFi protocols. Although the automatic liquidations prevent the protocol from amassing bad debt, users are at risk of losing 10% of their deposits if their trove is liquidated. Additionally, if the protocol’s overall collateral ratio drops below 150%, any transactions that would further decrease the total collateral ratio are blocked and any individual trove with a collateral ratio below 150% are at risk to liquidation. Even under this circumstance, the liquidation loss is capped at 10%, similarly to a trove that’s liquidated at a 110% threshold. The remaining collateral is liquidated, but saved and claimable by the individual who was liquidated. Additionally, when LUSD is redeemed for ETH (not the repayment of a loan), the trove with the lowest collateral ratio loses part of its ETH collateral while seeing its debt position decrease. While technically net-neutral, the depositors can end up with more LUSD and less ETH than initially desired. Thus, no user wants to have the lowest collateral ratio, encouraging users to keep their collateral ratios well above the minimum 110%.
Given that LUSD is collateralized by a volatile asset like ETH, it is understandable that its soft peg to $1 oscillates more frequently than fiat-backed stablecoins. However, unlike other decentralized stablecoins, which often face the issue of trading below $1, LUSD has more recently been trading above $1. In general, the stablecoin maintains peg through its natural price floor and ceiling. The ability to redeem 1 LUSD for $1 worth of ETH (minus the redemption fee) maintains a price floor of $0.95 to $0.995, and the minimum collateralization ratio of 110% acts as a price ceiling, given that no rational agent would value LUSD greater than $1.10. This insinuates that LUSD is worth more than its backing.
Highlighted in black are the times of substantial price deviation from peg in addition to the most recent trend since April, which has been valuing LUSD above peg. One possible explanation for its market value above $1 is the requirement to borrow 2,000 LUSD at a minimum to open a trove. If any individual wants exposure to the stability pool yield with less than $2,000 in value, they must buy LUSD on the open market, making it trade at a slight premium.
Aside from the efficient mechanics, Liquity’s adherence to the principles of maximum decentralization offers an added unique selling point. The team is offering valuable competition to the idea of a truly decentralized stablecoin. With no active governance, Liquity’s code is completely immutable and can never be changed. Liquity takes an approach that removes the ability to ever propose future changes. Liquity doesn’t even run their own front end, relying on third parties to host the back end services in return for a small fee charged on the user’s end to avoid any points of potential centralization. While competitors like MakerDAO have active governance forums, with recent proposals for venturing into real world lending, Liquity simply wants to be a protocol that can last exactly as it is for decades to come. Whether this is an uphill battle against communities that can mobilize and pass net positive proposals or a safety mechanism to prevent any potential unforeseen problems with future changes is yet to be seen. All that is known currently is that Liquity works well, remains decentralized, and won’t be changing anytime soon.