【In-Depth Research Report】Liquity: Rising Star in the Stablecoin Market
By Li Yuxuan, Mint Ventures
Updated on July 8th, 2021
1. Key points
LUSD has the best mechanism among current over-collateralized stablecoins. Although it has been launched for less than 3 months, the effectiveness of its mechanism has been fully proven. The subsequent development primarily tests LUSD’S capabilities of operation and promotion.
The stability of the Liquity system is not conducive to increasing the token value of LQTY, as LQTY benefits more from the liquidity elasticity caused by the transition of market cycles.
LQTY has gradually returned to normal valuation from a state of extreme undervalued, and its future revenue may further decline.
2. Basic information
2.1. Business introduction and analysis
Liquity is a decentralized currency protocol, allowing users to lend LUSD (a stablecoin always pegged to USD) using ETH as collateral.
Through the three-fold liquidation mechanism of stability pool — trove redistribution — recovery model, Liquity only requires a minimum collateral ratio of 110%, which greatly improves the capital utilization rate and maintains the protocol’s stability well.
The main use cases of Liquity are as follows:
- Open trove: Use ETH as collateral to borrow LUSD, with a minimum collateral ratio of 110%.
- Provide LUSD to the stability pool (SP) in exchange for liquidation yields and Liquity’s original incentive token LQTY.
- Stake LQTY to earn the fees paid by other users for borrowing or redeeming LUSD.
- Use LUSD to redeem ETH.
The core mechanism is as follows:
a. Price stabilization mechanism
LUSD is a stablecoin pegged to 1 USD. For any stablecoin pegged to fiat, the price stability mechanism is the core. The core of LUSD’s price stabilization mechanism is that the system allows users to stake ETH for LUSD minting at the price of 1 USD at any time, or payback or redeem LUSD at the price of 1 USD to recover ETH. (Note: payback refers that users who have the debt of LUSD can return LUSD to the system in exchange for the collateral ETH; redemption means that any user can exchange ETH with LUSD from those users with the lowest collateral ratio at any time at the price of 1 USD/LUSD).
This means:
- When the price of LUSD on the market is higher than 1 USD, users can mint new LUSDs and sell them to the market for profit (the minting cost of LUSD is only 1 USD). Of course, the above statement is not mandatory, but there are also many limiting factors, such as the liquidation fee of 200 LUSD reserved for new troves, as well as the minting fee and redemption fee (between 0.5% and 5% controlled by the algorithm, which will be detailed below). However, the minimum collateral ratio is limited to 110% at the time of minting, and LUSD is always priced at 1 LUSD = 1 USD (in ETH) when minting LUSD. Therefore, when the LUSD price is higher than 1.1, any user can obtain it by opening a new trove and immediately sell it to the market for profit. This can be said to be the upper limit of price control reserved by the system.
- When the price of LUSD on the market is lower than 1 USD, users who have the debt of LUSD will have an incentive to buy LUSD from the market to payback the system (because their LUSD minting cost is 1 USD). Of course, there are also extreme situations. For example, under extreme conditions, due to the desperately short of market liquidity the borrower may be unable to operate, so the system reserves a lower price limit through “redemption”. Any user can buy LUSD from the market at a price lower than 1 USD at any time, while redeeming ETH worth of 1 USD from the system at the same time.
The above mechanism is what Robert Lauko, founder of Liquity, called “hard peg”: the price of LUSD is controlled between [1-redemption rate, 1.1] USD by opening arbitrage opportunities to the entire market.
In addition, he believes that, because the Liquity system allows users to mint LUSD at the price of 1 USD at any time, or burn LUSD at the price of 1 USD, so there will be a “soft peg”. That is, the system always thinks that 1 LUSD = 1 USD, Therefore, the user will also form a Schelling point (people’s natural preference in the absence of communication in game theory) of 1 LUSD = 1 USD in the long-term game.
Under the effect of the above mechanism, the upper limit of the price hard peg is 1.1 USD. However, if the market knows that the price of LUSD will not be higher than 1.1 USD and the reasonable value is 1 USD, the price of LUSD will be more difficult to approach 1.1 USD. For example, for LUSD with a price of 1.09 USD, buying 1.09 USD of LUSD means a maximum profit margin of 1% and a possible loss risk of 9% at any time. The following figure is a good illustration of the overall price stabilization mechanism of LUSD:
In fact, this stabilization mechanism is very effective. The figure below is the LUSD/USDc price calculated based on Curve’s LUSD/3pool (Curve’s LUSD/3pool is the exchange with the largest liquidity of LUSD now).
Based on the above figure, Liquity has been launched for less than 3 months, and the price of ETH has fallen by more than 50% during this period with a single day dropping more than 40%. However, the price fluctuation of LUSD has been maintained between 0.97 and 1.03, and most of the time between 0.99 and 1.02. This shows the stability of LUSD.
b. Liquidation mechanism
A brief overview of Liquity’s liquidation mechanism is as follows:
Under normal circumstances, when a user is liquidated (i.e., the collateral ratio < 110%), the stability pool will act as the counterparty of the liquidation. When the LUSD in the stability pool is insufficient to cover the liquidated trove, the trove redistribution mechanism will be activated. Finally, when the over-collateralization rate of the entire system is less than 150%, the system will enter the recovery mode.
Among them, the stability pool is the first and most commonly used layer in the liquidation mechanism. The trove redistribution and recovery mode is more of a protection mechanism for protocol security in extreme cases.
Stability pool
The stability pool mechanism is as follows:
When a user’s collateral ratio is less than 110%, anyone can perform liquidation. The liquidator can receive a liquidation compensation of 200 LUSD and a reward of 0.5% of the liquidated ETH. The remaining 99.5% of the ETH will be distributed to all users according to the proportion of their LUSDs in the stability pool, and the LUSD will be burned proportionally. Assuming that a user is liquidated when his or her collateral ratio is just below 110%, then the ratio of the value of ETH obtained by all users in the stability pool to the value of LUSD burned is 1.09945:1. That is to say, when a user of the stability pool sells the ETH obtained by liquidation immediately after the liquidation, his or her earning is close to 10%. This is the main reason why Liquity quickly achieved more than $1 billion TVL after launch.
In fact, it is not without risks to participate in the stability pool, because depositing in the stability pool means “agreeing to buy ETH during the period of price decline”. When the price continues to fall (more than 10%), the ETH not withdrawn from the stability pool in time may also suffer losses. Therefore, the project party encourages users to deposit LUSD into the stability pool by staking LUSD to mine LQTY.
Trove redistribution
When the LUSD in the stability pool is completely used up by liquidation, Liquity will automatically move to the second stage of liquidation. In this stage, the system will distribute the ETH to be liquidated and the LUSD owed to the system to all existing troves, redistributing according to collateral ratio in proportion. In other words, the higher the collateral ratio of a trove, the more debt (LUSD) and collateral (ETH) it receives “from the liquidated position”. Through the mechanism of trove redistribution, the system will be ensured without chain liquidation.
The official example is shown in the figure below:
Now Trove D’s collateral ratio is 108%, triggering a liquidation. Due to insufficient LUSD in the stability pool, the collateral 4.3 ETH and 8000 LUSD debt in the Trove D will be distributed to other Troves, and Trove D’s net loss is about 7% (600/8600=7%). In the remaining three troves of ABC, the more sufficient collateral ratio is, the more collateral and LUSD will be distributed. Trove C is distributed the most collateral (7 ETH) and the largest amount of debt (4480 LUSD), so its collateral ratio declines from 233% to 180% with a net gain of $336.
It can be seen that in the entire process of trove redistribution, since the collateral and total debt amount have not changed, the overall collateral ratio of the system has not change. The collateral and debts of liquidated troves have been updates in the form of lowering down the collateral ratio of all troves.
It is worth mentioning that since the launch of Liquity, there has never been any use of “trove redistribution”, even for the plunge on May 19.
Recovery mode
When the overall collateral ratio of the system is under 150%, the system will enter recovery mode. The purpose of the recovery mode is to quickly increase the overall collateral ratio of the system to over 150% to reduce the system risk.
In the recovery mode, troves with a collateral ratio less than 150% may be liquidated. Since liquidation can receive a gas compensation of 200 LUSD and a 0.5% liquidation reward, there will be pure arbitrage users to perform liquidation to help the system improve the overall collateral ratio.
The liquidation under the recovery mode is complicated, but in general, Liquity sets a 10% upper limit for the losses that users suffer in liquidation. According to the relationship between the collateral ratio of current position and the overall collateral ratio of the system, as well as the adequacy of LUSD in the system stability pool, liquidation behaviors will be different:
In addition, the system will prevent transactions that would further reduce the overall collateral ratio. In other words, in the recovery mode, an opened position is not allowed to lend more LUSD by reducing the collateral ratio. New LUSD can only be issued by adjusting an existing position to increase its collateral ratio (that is, supplementing collateral), or by opening a new position with a collateral ratio >=150%.
In the plunge on May 19, we witnessed the recovery mode. The official published a more detailed analysis article after the event, you can click on view at： https://medium.com/Liquity/how-Liquity-handled-its-first-big-stress-test-160f20d5b18f.
In addition, let’s discuss the issue of “Will Justin Sun’s 600,000 ETH queued for liquidation make ETH to fall below 1000U?” at the time. In fact, because Justin Sun’s positions in Liquity accounted for a high proportion of Liquity’s total positions, reaching more than 40%. So Justin Sun’s personal “position collateral ratio” is actually very close to the “overall collateral ratio of the system”. The time duration for the system to go into the recovery mode is very short, so it is difficult for the afterward data statistics to reconstruct the scenario at that time, and we have no way to know the relationship between Justin Sun’s personal collateral ratio and the system collateral ratio. In order to comprehensively examine the mechanism of Liquity protocol, let’s consider all possible scenarios:
1. Justin Sun’s collateral ratio is lower than 150% but still higher than the overall system collateral ratio, which corresponds to the 6th scenario in the above table. So Justin Sun will not be liquidated.
2. Justin Sun’s collateral ratio is lower than the overall system collateral ratio, however, due to the rapid decrease of LUSD in the stability pool under the recovery mode, the LUSD in the stability pool is not enough to pay Justin Sun’s debts. This corresponds to the 5th scenario in the above table 5, so Justin Sun will not be liquidated.
3. Justin Sun’s collateral ratio is lower than the overall system collateral ratio, and the current LUSD in the stability pool is sufficient to pay Justin Sun’s debts, which corresponds to the scenario of 4 in the above table. In this case, the value of ETH equals to the LUSD owed by Justin Sun*1.1, which will be evenly distributed to all people in the stability pool to “thank them for paying the debt for Justin Sun”. At the same time, Justin Sun’s ETH part above 110% collateral ratio is still his own, which means Justin Sun will have a net loss of 10% in USD (at the time of liquidation).
The above scenario 3 is the scenario that everyone is most worried about. However, it should be noted that Justin Sun’s ETH is not obtained by one single liquidator, but by all users in the stability pool. All users in the stability pool are themselves ETH believers (that’s why they stake ETH to generate LUSD), they are not willing to sell after getting a 10% discount of ETH, and the high gas on Ethereum at the time will also prevent some small accounts from doing claim + swap operations. Therefore, even if the above scenario 3 occurs, it will not have too much impact on the ETH price.
There is no doubt that the plunge on May 19 has proved Liquity’s stability supported by its low collateral ratio.
Stability pool is the core factor of Liquity’s 110% collateral ratio and guarantee of stability. As the white paper of the project said: “Since the acquirers are known in advance, there is no need to find a buyer for a collateral buyout on the spot when a position becomes undercollateralized. This advantage allows for a considerable reduction in the collateral ratio, while keeping stability high.” Especially compared with similar projects, this kind of stability is particularly valuable. We will elaborate in detail in the section of 3.2 “Competition Landscape”.
c. Supply control mechanism
Liquity’s loans have no interest, and its control over supply and demand comes from rate changes in minting cost and redemption fee.
The minting cost rate and the redemption fee rate will be adjusted according to the time cycle and redemption amount for each redemption. Simply put, when there is no redemption in the system, the minting fee and redemption fee will be reduced (the minimum 0.5%), and these fees will increase with the increase of redemption amount.
It needs to be pointed out again that “redemption” here does not mean “payback”. Redemption means that any user can use LUSD to exchange ETH from those users with the lowest collateral ratio at any time at the price of 1 USD/LUSD. That is, when LUSD >= 1 USD, redemption is unprofitable. Therefore, there is no redemption demand for the protocol under normal conditions. According to historical data, most of the time the protocol has no redemption, except for the extreme market period on May 19.
Relatively speaking, the supply and demand control mechanism of Liquity is more “passive”: it does not take effect when the system expands, however it reduces the impact on the system through punitive measures during contraction.
More demand for LUSD will not increase the minting fee and redemption fee of LUSD. However, Liquity’s mechanism will not take effect until the LUSD in the system is so oversupplied that the LUSD price falls below 1$ and a redemption occurs. The primary purpose of this mechanism is to suppress large-scale redemptions (the rise of redemption fee will further reduce the profitable redemption price, thereby reducing the possibility of continuous redemptions) to maintain the stability of the system.
At the same time, the minting fee will increase simultaneously with the rise of redemption fee to suppress a large amount of borrowings after large-scale redemptions. A cooling-off period is used to ensure the long-term stability of the system and resist the systemic risk.
In addition, since Liquity does not charge interest instead a one-time fee at the time of minting, the Liquity protocol encourages users to hold LUSD for a long time. For users who borrow once within half a month, the actural interest is equivalent to 12% per year which is higher than the loan interest of most lending protocols, so the rational choice is to hold LUSD for a long time. This will reduce the short-term speculation in the use of LUSD to a certain extent, which will also affect the demand for LUSD to a certain degree.
d. LUSD use cases
For stablecoins, the stability mechanism is the foundation, and more important is the expansion of use cases. It is almost impossible for stablecoins without use cases to maintain anchoring.
The figure below shows the distribution of all LUSDs since the launch of LUSD (percentage):
We can see that the stability pool has always been the largest use case for LUSD, about 90% of LUSD distribution. In addition, in the first 6 weeks after the launch of Liquity, the official ETH-LUSD LP on Uniswap v2 had LQTY incentives (the pink part in the picture). At the same time, LUSD was launched on the 3crv pool of Curve at the end of April. Curve’s stablecoin exchange is a basic use case for stablecoin on Ethereum. At present, the stablecoin transaction volume of LUSD ranks sixth in the stablecoin pool of Curve, with a daily transaction volume of about US $3 million.
Curve’s 3crv pool is the main external use case for LUSD now. Currently, LUSD in Curve accounts for about 10% of the total circulation of LUSD. In addition, in terms of use case expansion, Liquity’s actions are very frequent, including: Yearn launched a strategy for crvLUSD, integrated with the application of DeFiSaver in the aggregation layer, promoted the integration of Element Finance with crvLUSD, launched LQTY jar on Pickle Finance, launched LQTY-BNT’s impermanent loss pool on Bancor, and will soon launch the integration with instadapp, etc.
However, from the distribution of LUSD, only 10% of LUSD is outside the stability pool, and most of LUSD is still in the Liquity system. If we don’t regard LUSD in the stability pool as circulation, the current ETH with more than US $2 billion has only generated less than 100 million circulating LUSD. It seems that LUSD with the goal to be a stablecoin only circulates within the Liquity protocol, so the sustainability of the entire system will be questionable.
Robert Lauko, founder of Liquity, also responded to this question in an article published on July 2. His core views include:
1. The LUSD in the stability pool is not locked. The current result is a rational choice of users under a large number of LQTY incentives. In the future LQTY token incentive will fade, and users will change their behaviors accordingly.
2. Based on data analysis, more than 3/4 of the minting users do not mint LUSD just for gaining mining profits from the stability pool.
The full article can be viewed at ( Robert Lauko’s frankness is impressive) https://medium.com/Liquity/addressing-concerns-about-Liquity-2af3e1d0a946
Honestly, for a stablecoin project launched only for 3 months, it seems a bit too harsh to discuss its insufficient use cases. Moreover, chainlink has not launched a price oracle service for LUSD, so the promotion of LUSD still has some constraints. However, the importance of LUSD use case promotion cannot be overstated for a stablecoin market with high switching costs and scale effects, and it will also be the main part of the team’s future work.
2.2. History of development
- The project was launched at the end of 2019.
- In May 2020, Liquity released the first version of its white paper, which established the core framework of the entire project (including 110% collateral ratio, stability pool, recovery mode, redemption system, and interest-free lending).
- The development version of UI was officially launched in June 2020.
- The seed financing led by Polychain Capital was completed in September 2020.
- An updated version of the white paper was released in February 2021. The new white paper focused on liquidation and stability: the recovery mode had been redesigned and the incentives for liquidators had been modified.
- The audit conducted by Coinspect and trail of baits was completed in March 2021.
- The Series A financing led by Pantera Capital was completed on March 29, 2021.
- Liquity was successfully launched on the ETH mainnet on April 5, 2021. All functions were normally launched except for the redemption function (from the perspective of system stability, redemption is not allowed in the first 14 days).
- After the project was launched, it quickly attracted more than 1 million Ethereum due to its refined design of the system, with more than US $4 billion of deposit at its peak. Since then, due to the price plunge of ETH itself and the liquidation, the total deposit amount of Liquity dropped to about US $2 billion.
The current core data of Liquity are as follows:
2.3. Team, investors and partners
a.Team
Robert Lauko: Founder & CEO，graduated from the University of Zurich with a doctorate in law. With many years of legal affairs and lawyer experience, he was an assistant researcher at Dfinity before he founded Liquity.
Rick Pardoe: Co-founder & Core Development, Rick’s experience is very interesting. He has a B.A. in physics and a M.A. in economics. Before working on blockchain, he worked in SEO and sales for a long time. In 2017, he started to develop blockchain field and created the website: http ://ETHdevs.com/
Kolten Bergeron: Head of Growth, former ecosystem and community development manager of Stellar Development Foundation.
The company LinkedIn shows that it has 10 team members now, most of whom are developers.
At present, Liquity has 3 consultants, namely:
Ashleigh Schap: also head of growth at Uniswap and previously worked at MakerDAO.
Yulin Liu: PhD in Economics from the University of Zurich, formerly an economist at Dfinity, He is now an associate professor of economics at Huazhong University of Science and Technology (this information comes from the official website), Dr. Liu has co-authored a large number of academic papers on cryptocurrency. Dr. Yulin Liu did the initial macroeconomic model simulation for Liquity, which provided a basis for the stability of LUSD under the fluctuation of ETH.
Cédric Waldburger: the original investor in Liquity.
b.Investors
Liquity has three rounds of financing, including pre-seed funding, seed funding and series A funding.
The pre-seed funding was invested by Tomahawk.VC, where Cédric Waldburger works. The specific amount and time of funding were not disclosed.
In September 2020, it completed the seed funding of US $2.4 million led by Polychain Capital. The investors included a_capital, Lemniscap, 1kx, DFINITY Ecosystem Fund, Robot Ventures (Robert Leshner, founder of Compound), and Alex Pack.
In March 2021, it completed the series A funding of US $6 million led by Pantera Capital. The investors included Nima Capital, Alameda Research, Greenfield One, IOSG Ventures, AngelDAO, as well as angel investors Bo Shen, Meltem Demirors, David Hoffmann, Calvin Liu and George LambETH, etc. Previous investors such as Tomahawk.VC, 1kx, and Lemniscap also made additional investments.
C.Partners
In addition to the above investors and various DeFi protocols, the partners also include Gauntlet Network. Gauntlet Network is a well-known consulting company in the DeFi field. The head DeFi protocols such as Maker, Aave, Compound, and public chains such as Stellar and Mina all invited it to deliver reports. Gauntlet Network has made a comprehensive and adequate review of the Liquity protocol and conducted detailed parameter simulation tests (the full report can be viewed here: https://Liquity-report.gauntlet.network/). The key conclusions are as follows:
- Under various ETH fluctuation levels, the risk of system bankruptcy is the minimum under the current parameters (the minimum collateral ratio is 110%, and the system collateral ratio is 150%). In addition, the decrease of the system collateral ratio will significantly reduce the possibility of the system entering the recovery mode, while it will not significantly increase the risk of the protocol.
- Without LQTY token rewards, the collateral incentive mechanism of stability pool may not be very robust. When the ETH volatility is low, there is very little liquidation, so users have little profit in staking in the stability pool. However, when the ETH volatility is high, the stability pool is more likely to be exhausted. When the ETH volatility remains within the historical record level, the return of stability pool is still positive, and it actually will increase as the volatility increases. It is expected that LQTY token rewards can reduce the instability risk of staking incentives during the initial stage.
- Network congestion and rising GAS prices have little impact on liquidation incentive and bankruptcy risk.
- The price stability of LUSD may be affected by gas price and average transaction volume. If the average position size is small, when the gas price is high, users may not have enough incentive to redeem their LUSD to provide the required level of price stability. At the same size, LUSD shows similar stability characteristics to the Dai of MakerDao.
Gauntlet also suggested that Liquity can adjust its system collateral ratio to 130% and the minimum collateral ratio to 120% in the future.
3. Business analysis
3.1. Industry analysis
Liquity is in the industry of stablecoin or over-collateralized stablecoin.
Since the establishment of Realcoin in July 2014, the predecessor of Tether, stablecoins have developed rapidly. Both USDT and USDC have ranked among the top 10 of cryptocurrency market capitalization, and BUSD was once ranked in the top 10. In 2021 alone, the total market capitalization of stablecoins has increased from US $28 billion at the beginning of the year to US $110.1 billion U.S. dollars today (as of July 14), faster than the overall growth rate of the cryptocurrency market capitalization during the same period.
The current stablecoin market is mainly divided into two groups. One is the custodial stablecoin represented by USDT/USDC, and the other is the non-custodial stablecoin based on the logic of synthetic assets on the chain. Specifically, based on the collateral ratio of synthetic assets, non-custodial stablecoins can be divided into over-collateralized stablecoins and non-full-collateralized stablecoins (including partially collateralized stablecoins and un-collateralized stablecoins).
Among the top ten stablecoins by market capitalization, USDT, USDC, BUSD, TUSD, PAX, and HUSD are all custodial stablecoins. In addition, DAI, UST, LUSD and USDN are non-custodial stablecoins, and all are over-collateralized stablecoins except for the special USDN.
Custodial stablecoins can be regarded as the mapping of fiat currencies (mostly USD) on the chain. If we do not discuss the trust in the fiat currency itself, for the custodial stalecoin, its credit mainly comes from the trust in its issuer, and the core is whether it can ensure that each stablecoin on the chain corresponds to one US dollar in the real world. Therefore, issuers of custodial stablecoins generally ask accounting firms to certify the corresponding U.S. dollar reserves. For example, for USDC, its accounting service company Grant Thornton issues a monthly certification report for the U.S. dollar reserves of USDC tokens in circulation; BUSD and PAX are certified monthly by the accounting agency Withum; USDT, which has been criticized before, also issued its first certification data on its reserves in May this year (although its quality is still not as good as that of USDC and BUSD).
For non-custodial stablecoins, their credit mainly comes from their codes.
For the over-collateralized stablecoins represented by DAI and LUSD, their codes specify the minimum collateral ratio and underlying assets. Users’ confidence in them comes from trustworthy underlying assets and reasonable and verified token mechanisms. After several years of exploration by DAI, the core mechanism of over-collateralized stablecoins has been verified to be effective with the recognition by the market.
Among the non-full-collateralized stablecoins, it has experienced the first wave of upsurge triggered by AMPL and YAM in the DeFi summer of 2020, the second upsurge caused by ESD and BAS at the end of 2020, and the third wage of upsurge triggered by FEI, FLOAT and OHM during March to April of 2021. In general, for the entire track of non-full-collateralized stablecoins, its core price stabilization mechanism has not yet been fully proven, since users are more involved in it as speculative products rather than stablecoins.
Since 2020, the following factors have increased the demand for stablecoins:
- The popularity of CEX derivatives (such as perpetual contracts) has grown sharply.
- The position of Bitcoin as the base currency in spot transactions was replaced by stablecoins.
- The outbreak of DeFi and the upsurge of liquidity mining tokens.
- The overall demand for cryptocurrencies increases, especially the entry of institutions.
The stablecoin market is a market with very obvious first-mover advantages and network effects. If latecomers do not have a very obvious advantage over existing products, users (including B-end users and C-end users) are not willing to switch their stablecoins. This is why the replacement of USDT by USDC has been slow (which has accelerated recently), despite the fact that USDT has always had great flaws in the core trust mechanism of stablecoins. Even in the next stage of market development, whether and how new players enter the game that can truly change the game may be determined only by “what problems the old players have”.
The high switching cost is the main reason for the first mover advantage of the stablecoin market. For CEX, whether it is spot trading or derivatives trading, most of its current trading volume comes from USDT trading pairs. The habits formed over time will also make users become “USDT standard”. For liquidity mining, its switching cost will also be very high once the project party has determined the stablecoin paired with the project party’s token.
The most fundamental reason why USDC and BUSD are catching up with USDT is that the credit problem of USDT itself has not been solved. On top of this, USDC mainly seized the DeFi boom and new opportunities of institutional entry. BUSD added the new scenario of zero transaction fee and mining on BSC in the Binance exchange, so it has a rapid catch-up of market shares. However, it should be noted that the most basic derivatives trading and spot trading scenarios of USDT are not greatly affected, which is enough to reveal the high switching cost of the stablecoin system.
3.2. Competition landscape
LUSD is an over-collateralized stablecoin. Currently, non-custodial stablecoins with a market capitalization higher than LUSD include DAI and UST. UST is not an over-collateralized stablecoin, its rapid development is mainly due to its good performance in the expansion of use cases, and it had a serious price de-anchorage due to the plunge on May 19. So in this article we will only compare LUSD and DAI.
a.LUSD vs. DAI
You can’t discuss LUSD without talking about DAI. The role of Maker and DAI in cryptocurrencies is unquestionable: Maker is one of the oldest protocols in the cryptocurrency field, and DAI is the most successful non-custodial stablecoin. In addition, MakerDAO has also led the trend of DAO.
As a new over-collateralized stablecoin, Liquity has a lot of settings aimed at certain pain points of Maker. This is the aggressive attitude that latecomers should have. Next, we will look at the differences between LUSD and DAI in detail from the aspects of price stability mechanism, liquidation mechanism, supply and demand control mechanism, capital efficiency, and degree of decentralization.
Price stabilization mechanism
In fact, DAI does not specifically design a price stabilization mechanism. The core rule of price stability is the Dai can only be mint at the price of DAI = 1 USD.
Therefore, when the price of DAI in the market exceeds US $1, arbitrageurs can produce DAI by opening a new position in Maker to pull the price of DAI to close to US $1 to some extent (this is not completely called arbitrage, because a collateral is required to open a position to generate DAI. If you want to recover the collateral, you still need to pay the DAI back to the system.) When the price of DAI is less than US $1, theoretically some users who already have DAI debt will buy DAI to pay off the debt.
It is not mandatory powter to maintain DAI to return to US $1 when it is higher or lower than US $1, and participants will be limited to those who already have DAI debt.
With the 110% collateral ratio and redemption function, Liquity fixes the price range of LUSD at [1- redemption rate, 1.1], and when the price exceeds this range, any user can carry out arbitrage. That is to say, anyone can participate this price stabilization mechanism, which makes the price stability of LUSD stronger.
Liquidation mechanism
Maker’s liquidation adopts an auction mechanism. When the collateral ratio of a user vault is lower than the minimum collateral ratio, anyone can use DAI to participate in the auction of collateral, and the auction price will gradually increase from 0 DAI until no new bids are made. Maker’s auction mechanism uses discounted collateral to encourage more liquidators to participate in the auction, so as to collect more DAI through the game between liquidators and reduce the system risk. For the protocol, the most desirable scenario is that the liquidator’s bid + gas fee is close to the price of the collateral.
During the sharp drop on March 12, 2020, the drastic fluctuation of the market led to gas soaring and network congestion, due to the auction validity period set only 10 minutes. Then the liquidity of DAI was exhausted, which made it difficult for borrowers and liquidators to obtain DAI to payback or participate in the auction. The minimum number of liquidators (robots) involved in the auction was 2, so there were quite a few ETH auctions were completed with a price of 0 DAI, which in turn resulted in a considerable loss for MKR holders (which was not compensated until November 2020).
Although Makerdao adjusted its auction mechanism after March 12 and extended the total auction duration to 6 hours, preventing insufficient liquidation bids. However, during the 6-hour auction duration, the prices of collateral (ETH) and auction token (DAI) can create more uncertainty. To some extent, it will further undermine the willingness of liquidators to participate in the auctions.
The liquidation mechanism is at the heart of the stability of any lending system. The liquidation mechanism should be designed with extreme situations primarily. In this regard, Liquity is fully prepared. On the one hand, since the initial version of its white paper, Liquity’s design has been around “stability under low over-collateralization ratio”. On the other hand, at the beginning of the project, it asked the consultant Dr. Liu Yulin to simulate the system stability with the historical ETH price data, and later it asked Gauntlet Network to conduct a detailed review. Finally, Liquity came up with the complete set of plans. It had an outstanding performance during the big test that the collateral plummeted by 40% within one day after only one month of launch.
Compared with Maker, Liquity’s liquidation mechanism has the following advantages:
- The incentives for liquidators (robots) are more direct and efficient. Gas compensation of 200 LUSD + reward of 0.5% liquidated position, while the cost is only 1 operation gas. Therefore, this makes liquidation obviously profitable. In addition, liquidation does not require any principal, and theoretically everyone can operate it. However, the liquidation of DAI essentially uses DAI to buy ETH at a discount. In this way, both the ETH-based and stablecoin-based liquidators need to have a token exchange process (ETH-based liquidators needs to obtain DAI first if they want to participate in the liquidation, and the stablecoin-based liquidators need to exchange the ETH back into stablecoins after they obtain ETH for the liquidation). Liquity’s liquidation mechanism will encourage more people to prepare to participate in the liquidation, and more liquidators mean that the system is safer. Because if a user’s collateral ratio falls below 110%, he or she will be liquidated immediately, and the system will not be exposed to the danger (the collateral ratio is below 100%).
- The liquidation is immediate. This means that there is no need for auctions, no need to consider block jams in extreme cases, and there will be no dilemma of Maker’s 0 DAI auction at March 12. With the prior consent of all stability pool users, the liquidation can happen immediately.
- The beneficiaries of liquidation are all users in the entire pool. This will have two benefits: First, because borrowing LUSD requires collateralization of ETH, users who hold LUSD are likely to be long-term investors of ETH. In terms of willingness, a considerable number of Liquity’s liquidation beneficiaries are not inclined to sell ETH (which is actually the reason why Liquity can obtain such a high TVL), so it is not easy to put more pressure on the price of ETH. Second, even if some users in the stability pool want to sell ETH in a sharp-drop market, the inevitably high gas fee (claim+swap) on Ethereum will actually hinder a considerable number of users from selling ETH. Therefore, from the perspective of ETH stability, Liquity is more suitable to be the largest central bank than Maker.
To sum up, Liquity has obvious advantages in terms of the liquidation mechanism over Maker, and it is the best in all over-collateralization protocols. In addition, Liquity will be more suitable to be the largest central bank than Maker from the perspective of ETH stability.
Supply and demand control mechanism
The supply and demand of DAI are adjusted based on the changes of its deposit and loan interest rates, and these adjustments are based on the judgment of MakerDAO members.
When there is excess demand for DAI, MakerDAO will usually raise the loan interest rate and lower the deposit interest rate, thereby reducing the increment of DAI and promoting the payback of DAI in circulation to bring the DAI price down. When the demand for DAI is low, MakerDAO will do the opposite to increase the demand for DAI (for example, during the cold winter of cryptocurrency from January to May 2019, the demand for DAI was low and the price was below US $1 for a long time. The deposit interest rate of DAI was raised from 0.5% to 19.5% in order to boost the demand.).
Compared with the supply and demand control mechanism of LUSD, Maker’s approach is more proactive. Based on the judgment of MakerDAO members, it guides market demand, much like the style of central banks around the world.
In terms of supply and demand control mechanism, Liquity chooses a completely different solution from Maker, although the latter solution may be easier to be understood by most people.
Capital efficiency
First of all, it needs to be clear that security and capital efficiency are fundamentally incompatible for any user who participates in over-collateralization to generate stablecoins (including over-collateralized lending). As a stablecoin protocol, Liquity has the best trade-off between security and capital efficiency among all over-collateralized protocols. The ingenious way it deals with this problem is that, it allows efficiency to be prioritized from the mechanism level, meanwhile it provides opportunities for security-first users to obtain more benefits.
The collateral ratio since the launch of Liquity has been below 300% from the long-term run, and even lower than 200% for a long time. The over-collateralization ratio of Maker’s ETH-A VAULT has been above 300% for a long time. From this perspective, Liquity’s capital efficiency is indeed much higher than that of Maker.
Just as the problem of excessive LUSD in the stability pool mentioned earlier, if we assume that the overcollateralization ratio of Liquity = the value of the collateral ETH/the actual LUSD in circulation, the overcollateralization ratio of Liquity will always be over 1000%. In fact, even if all LQTY token incentives end, some users with low risk preference will still deposit LUSD in the stability pool, and the system mechanism is designed to encourage this behavior. Therefore, we cannot accurately estimate the actual over-collateralization ratio of Liquity under this condition.
In general, Liquity has an obvious advantage over Maker in terms of capital efficiency in the most general sense, but this advantage is based on the fact that part of LSD will never go into circulation.
Degree of decentralization
At present, the DAI generated by USDC collateral has exceeded 60% of the total amount of DAI. This raises questions about the degree of decentralization of DAI: When 65% of DAI is generated by the collateral of centralized tokens (WBTC is also a centralized token), how can we can say that Dai is a decentralized stablecoin?
It should be pointed out that the current collateral composition of DAI is actually determined by the market. Before November 2019, DAI had only used ETH as collateral. Then it decided to expand the source of collateral, due to the influence of ETH’s volatility on system stability and scalability considerations. In fact, this consideration is meaningful, because it encountered with the dilemma of March 12 just 4 months after the expansion of the collateral. The challenge on March 12 had a huge impact on the stability of the Maker protocol. Under this background, MakerDAO formally proposed to add USDC as collateral for DAI. In the development process, since the interest rate paid to collateral USDC to issue DAI is very low, and DAI is widely used by DeFi projects, a significant number of users were willing to use USDC as collateral to generate DAI, resulting in this seemingly weird situation.
The author believes that it is reasonable to criticize the degree of decentralization of DAI based on this factor. MakerDAO was not unaware of this, but it gave up partial decentralization for the sake of protocol security, which is a more important reason. This decision seemed very reasonable at the time.
Now Liquity only allows ETH as collateral. According to Liquity’s non-governance framework, system collateral cannot be changed, so Liquity will only use ETH as collateral. Personally, the most important reason why Liquity is brave to make this decision is that the value of ETH has been recognized by more and more people in the past year. If we go back to March 2020, there is a question mark for whether Liquity dares to make such a decision. In addition, based on the concept of Web3.0, Liquity’s front-end is not operated by the core team of Liquity. Instead developers are encouraged to develop the front-end of Liquity. Developers are free to set up the kickback ratio of the front-end. There are currently more than 20 officially certified front-ends.
Overall, LUSD is much more decentralized than DAI.
Governance
The Maker protocols are governed in a decentralized way by their governance token MKR. MakerDAO can vote on protocol parameters (such as stability fee, debt ceiling, minimum collateral ratio) and important ecosystem issues (such as funding workgroups, grants, etc.).
MakerDAO is the pioneer of DAO in DeFi. They adopt the principle of “one token, one vote” to carry out governance work. Although there are several core participant groups to guide, the inefficiency of “populist” governance is still widely criticized. On the other hand, due to the complexity and high requirements of governance, the participation of ordinary token holders in governance is very low. MakerDAO’s decision-making is actually determined by several giant whales, which has also caused the concern in its community.
Liquity is the pioneer of non-governance or machine governance (LQTY has no governance function). The system parameters are either immutable or “algorithmic”. In Liquity, the system parameters such as the allowed collateral type and minimum collateral ratio parameters cannot be changed. The lending and redemption fees of Liquity are only determined by mathematics, without opportunity for human intervention.
On the issue of governance, Liquity shows trust in the founding team (rules and algorithms are determined by the founding team) and distrust of artificial governance. Even on the oracle switching, they have established a mechanism to automatically switch the oracle from Chainlink to Tellor.
As the veteran of DeFi, which was launched in 2018 and has survived to this day, DAI and Maker have contributed a lot to DeFi and the cryptocurrency as a whole, and their mechanisms are very successful in general. DAI is still an undisputed king of non-custodial stablecoins. After all, the overall design of Liquity draws on the development experience of DeFi during the past three years. In terms of the overall mechanism of stablecoins, especially the price stability mechanism and liquidation mechanism, the author argues that LUSD and Liquity have obvious advantages over DAI and Maker.
However, DAI still has a strong first-mover advantage over LUSD. As we mentioned earlier, “advantage” is not enough for the latecomers in the stablecoin market to replace existing players. Instead, it requires “absolute advantage” or even it can help existing players to achieve some “weak spots”. Now it seems that Liquity does show a certain “absolute advantage”, but DAI does not show some “weak spots”. The March 12 incident may be one, but there was no Liquity at the time, and DAI actually survived the May 19 incident smoothly after being upgraded under the governance of MakerDAO.
All in all, although Liquity has mechanism advantages, it will still test Liquity’s operation and promotion capabilities, if LUSD wants to take a share of DAI and even completely replace DAI.
3.3. Token analysis
a.Token distribution
The total volume of LQTY is 100,000,000 (one hundred million), which was released together with the launch of Liquity mainnet on April 5, 2021. As of July 8, 2021, the circulation of LQTY is 6.31 million (data from DuneAnalytics), and the circulating market capitalization is 33.5 million.
- 35.3% of LQTY is distributed to the Liquity community, of which 32,000,000 LQTY are distributed to the LQTY reward pool. These tokens are used to incentivize deposits in the stability pool, and the cumulative issuance follows the function: 32,000,000* (1–0.5 ^ years). That is, 16 million will be distributed in the first year, and 8 million for the second year. In addition, 1,333,333 LQTY will be distributed to the LP in the LUSD-ETH Uniswap pool within six weeks after the protocol was launched. Moreover, 2,000,000 LQTY has been set aside for the community reserve fund.
- 23.7% is allocated to current (and future) employees and consultants of Liquity AG. This part of LQTY has a one-year lock-up period, 1/4 will be unlocked after one year, and 1/36 will be unlocked every month thereafter.
- 33.9% is distributed to investors. 33,902,679 LQTY is distributed to early investors of Liquity, with a 1-year lock-up period.
- 6.1% is reserved for Liquity AG Fund, with a 1-year lock-up period.
- 1% is distributed to service providers, with 1-year lock-up period.
Within 36 months from the launch, the volume of tokens in circulation is shown in the figure below:
It can be seen that 12 months after the launch, namely April 2022, there will be a peak of LQTY unlocked, and the total circulation of tokens will increase to 75 million.
b.Token economy
As mentioned in 2.1.c “Supply and demand control mechanism”, the only use case of LQTY is to capture the minting fee and redemption fee of the protocol, without governance function. If we consider only from the perspective of LQTY holders, we can find that the value of LQTY does not depend on how much LUSD is in circulation, but on how much new LUSD is issued and how much new redemptions are made. The rising in the redemption of LUSD will increase the minting and redemption fees of LUSD at the same time. Therefore, the stage where LQTY benefits most is when the entire system is not healthy (experiencing contraction).
We have calculated the relationship between the total amount of LUSD and the main value captured by LQTY (total fee revenue of LUSD) since the launch, as shown in the figure below.
From the picture above, we can see:
- In the process which the supply of LUSD increases from 500 million to nearly 1.5 billion, the cumulative fee incurred is nearly US $4 million.
- Since then, the fee of Liquity system increases slowly from late April to mid-May, when the lockup volume of Liquity system is the highest.
- After the sharp drop on May 19, LQTY’s speed of capturing fees has accelerated.
- In June when the LUSD system is gradually stable, the value that LQTY can capture is not much, As the supply continues to increase, the yields of LQTY staking have decreased significantly. We calculate the total yields of LQTY on a daily basis, and convert the APR of daily yields according to the price of LQTY and the volume of staking, as shown in the following figure:
It can be seen that throughout June, the gain of LQTY was very small, and the annualized rate was far lower than the growth rate of LQTY’s inflation.
As the value capture token of the stablecoin system, LQTY does not benefit from the stability of the system, but from the fluctuations and even depression of the system. In addition, LQTY does not have the governance value commonly for DeFi tokens, so high TVL does not create value for LQTY.
We can even roughly get the following relationship between the status of the Liquity system and the value of LQTY:
Therefore, starting from the logic with the support of the above data, we have got a “weird” conclusion about the token economy of LQTY: the stability of the system is not conducive to (or even harmful) to the improvement of LQTY token value. LQTY benefits more from the liquidity elasticity caused by the transition of market cycles.
3.4. Risks
a. Oracle risk
The oracle risk is an inherent risk of any DeFi project. Liquity uses Chainlink’s oracle as the preferred solution and Teller as an alternative. In order to prevent problems with the oracle, they also designed a program to automatically switch oracle machines. Since only the price of ETH is generated through oracle in the core transaction process of Liquity, there will be less dimensions to be attached. Therefore, the oracle risk of Liquity is less than that of other DeFi projects.
b. Risks of insufficient promotion of LUSD
The stablecoin market has a strong first-mover advantage. If LUSD cannot be effectively promoted, it may not be able to make a real voice in the market.
c. Substitution risks of new projects
LUSD is a stablecoin anchored to the U.S. dollar, which is essentially an extension of the U.S. dollar. Starting with BTC, the whole market has always had the demand for a “stablecoin against inflation”. The recent money over issuing of the Federal Reserve has exacerbated this demand, but there is no product available to meet it. If a new stablecoin project can meet this demand, it could affect Liquity’s growth.
d. System upgrade risk
Since the main functions in the Liquity contract are non-upgradeable, this certainly shows the confidence of the team. However, if there are some unknown problems that need to upgrade some core functions of the system, how to organize and upgrade will be a test for the team’s ability.
4. Valuation
4.1. Five core questions
What business cycle is the project in? Maturity stage or the early and middle stage of development?
The project is in the early stage of operation since it has only been launched for three months.
Does the project have a solid competitive advantage? Where does it come from?
The competitive advantage of the project mainly comes from its verified core logic. Behind this there are many thoughts of the team on the design of stablecoin. From the actual situation, this competitive advantage is relatively solid.
Is the medium and long-term investment logic of the project clear? Is it in line with the industry trend?
The medium and long-term investment logic of the project is clear, aligned with the industry trend. The goal is to become a better decentralized stablecoin.
What are the main variables in the operation of the project? Are these variables easy to quantify and measure?
Given that the core logic of the project has been basically confirmed, the main variables in the operation of the project mainly come from the promotion of LUSD use cases. This variable is easy to measure, and it can be validated by the data such as the total volume and distribution of LUSD.
What is the management and governance of the project? What is the level of DAO?
The governance method of the project is the team’s preset rules + algorithm governance. Its token LQTY has no governance function at all, so there is no DAO.
4.2. Valuation assessment
a.Vertical valuation assessment
It should be pointed out that the launch time of Liquity protocol is too short and the protocol has not entered a stable state in most of the time. So the vertical valuation assessment will change dramatically, and its early valuation is not a good reference for future valuation.
Liquity’s protocol revenue is all captured by LQTY tokens, so its P/S (full-diluted market cap /annualized protocol revenue) and P/E (full-diluted market cap/annualized protocol earning) are essentially the same. Since its launch, the full-diluted market cap of LQTY tokens and the trend of P/S are shown in the figure:
Here the “P” refers to the full-diluted market cap, and the circulating volume of LQTY is still less than 7% until now. So when we convert it into the circulating market cap, we can find that the circulating P/E of LQTY is less than 1 in most of the past time, and LQTY is absolutely underestimated. With the recent stabilization of the Liquity system, the minting fee revenue has gradually decreased. With the gradual increase of the circulation volume, its circulating P/E has gradually increased, although the current circulating P/E is still less than 10.
Overall, from a vertical comparison, the valuation of LQTY tokens has gradually shifted from underestimation to reasonable.
b.Horizontal valuation assessment
We select Maker’s governance token MKR which is the biggest competitor of Liquity to compare it with LQTY for valuation assessment.
Since all of Maker’s protocol revenue is also earned by MKR, its P/E and P/S have the same calculation method as LQTY. Let’s directly compare the P/E of MKR and LQTY:
Similarly, here the “P” also refers to the full-diluted market cap. After the conversion based on the circulation rate of LQTY, the circulating P/E of LQTY gradually rises from less than 1 to close to 10, while MKR is already in full circulation, with its circulating PE between 20 and 30.
In addition, it is also worth pointing out that, MKR has governance value compared to LQTY. Although the governance value cannot be accurately estimated, it must be real (proved by a series of pure governance tokens, such as UNI / YFI / MIR / BAL). Given this, if the governance value of MKR is removed, its valuation will be closer to that of LQTY.
Since the system revenue of Liquity and Maker are all captured by the circulating tokens, it makes sense to use the circulating P/E to compare the valuations of the two. From the perspective of circulating P/E, the current valuation of LQTY is still lower than that of MKR, but the two have gradually become close.
In the long run, as the Liquity system has entered a stable stage and LQTY’s ability to capture value will be weaker than the fluctuation stage, LQTY’s future revenue capability is expected to keep decline. The cumulative revenue of the system also shows this trend:
In general, from the perspective of the circulating market cap, the valuation of LQTY tokens has gradually returned to normal from an extremely underestimated state. In the future, the valuation of LQTY may further return to normal driven by the decline of revenue.
4.3. Summary of core logic
LUSD has the most advanced and excellent mechanism (better than DAI) among the over-collateralized stablecoins. Although it has been launched for less than 3 months, its mechanism effectiveness has been fully proven. The future development mainly tests the operation and promotion capabilities of LUSD;
The stability of the Liquity system is not conducive to improving the token value of LQTY, and LQTY benefits more from the liquidity elasticity caused by the transition of market cycles.
LQTY’s valuation has gradually returned to normal from a state of extreme undervalued, and its future revenue may further decline.
5. Reference
https://www.chainnews.com/articles/480992856731.htm
https://medium.com/Liquity/on-price-stability-of-Liquity-64ce8420f753
https://duneanalytics.com/dani/Liquity
https://www.chainnews.com/articles/020914946311.htm
https://blog.Makerdao.com/the-market-collapse-of-march-12-2020-how-it-impacted-Makerdao/
*If there are obvious factual, understanding or data errors in the above content, please give me feedback and I will revise it.
