This article will explore the performance of lending, trading protocols, and decentralized stablecoin systems that were most affected by the USDC de-peg crisis, as well as potential trading opportunities.

Overview of the USDC De-Peg Crisis

USDC is a centralized stablecoin pegged to the US dollar and issued by Circle and Coinbase. When Silicon Valley Bank (SVB) filed for bankruptcy on March 11, 2023, Circle’s partial cash reserves were frozen at the bank, lowering market confidence in USDC and triggering substantial runs and selloffs. USDC’s price fell from $1 to $0.878, with significant price differences compared to other stablecoins such as DAI and BUSD. Following, USDC prices returned to near-normal levels on March 13 as market panic gradually subsided in response to the joint rescue plan of the Federal Reserve, Treasury Department, and FDIC.

Chart: Major events in the USDC De-Pegging Crisis and the price trends of related stablecoin tokens in the stablecoin race.

In order to address this crisis, Circle took the following actions:

Throughout the USDC crisis, it triggered panic and volatility in the cryptocurrency market, affecting investor confidence and trading activity. During this period, certain segments of the market, including centralized stablecoin markets, decentralized stablecoins, on-chain lending, and DEX, all faced risks:

Centralized stablecoin market: The USDC crisis may damage its position and reputation in the centralized stablecoin market, first raising doubts and panic selling of all stablecoins, but later providing an opportunity for other competitors (such as TUSD, USDP, etc.) to seize its market share. At the same time, the panic period created low-risk arbitrage opportunities for stablecoins like BUSD and USDP that were not at risk of breaking their peg.

Decentralized stablecoins: The USDC crisis affected decentralized stablecoins (such as DAI, FRAX, MIM) that used USDC as their reserve or collateral, resulting in significant deviations from their peg and exposing them to liquidation risks and arbitrage opportunities. At the same time, it may stimulate innovation and development of decentralized stablecoins (such as sUSD, LUSD, RAI) that do not rely on Fiat reserves or collateral.

On-chain lending: On-chain lending platforms that utilize USDC as a lending asset or collateral, such as Aave and Compound, were impacted by the USDC crisis, which resulted in interest rate fluctuations, USDT liquidity depletion, or liquidation events. At the same time, Compound, which defaults to a USDC price of $1, faces greater risks.

DEX: DEXs (such as Uniswap, Curve) that use USDC as a trading pair or liquidity pool asset may experience price slippage or arbitrage opportunities as a result of the USDC crisis. Simultaneously, it may push DEXs to improve their trading efficiency and flexibility to adapt to market changes.

Synthetix

Synthetix launched in 2018 as a synthetic asset protocol and gradually transitioned into a decentralized liquidity supply protocol built on Ethereum and Optimism. Users can generate the stablecoin sUSD by staking the governance token SNX. The current size of sUSD is approximately $55 million. The collateralization rate for SNX minting sUSD is 400%, and the liquidation threshold is 160%. Capital utilization is inefficient because of this relatively high collateralization rate. Additionally, due to SNX being the governance token of the protocol, its price fluctuations are significant. To address serious market risks and maintain stability, a higher collateralization rate is needed.

sUSD maintains its price peg through arbitrage mechanisms. If the market price of sUSD is higher than the mint price, arbitrageurs may mint new sUSD increasing market supply, then sell it at the market price thus lowering its price. When the market price is lower than the mint price, arbitrageurs can buy back sUSD from the market, then burn it, to reduce debt.

The application of sUSD is based on the “debt pool” formed by SNX collateral. The debt pool is a unique mechanism of Synthetix. All users who stake SNX to mint sUSD share a debt pool. When a user mints sUSD, the proportion of the minted sUSD to the total sUSD in the debt pool is the user’s share, and all minted sUSD is the total debt of the entire system. If a user’s investment strategy achieves asset appreciation (such as purchasing sETH with sUSD and the price of sETH rises), it will increase the debt of other users.

The debt pool can provide liquidity with zero slippage and can act as a counterparty for various protocols to provide liquidity services, with good composability.

On the basis of this debt pool, SNX has built its ecosystem. Synthetix does not directly provide any frontend, but serves as the backend liquidity provider for some DeFi protocols. The current ecosystem includes Curve, the contract exchange Kwenta, the options exchange Lyra, etc. sUSD has a relatively stable usage scenario. Recently, the trading data and revenue data of Kwenta have increased significantly.

During the USDC de-peg panic, although the underlying assets of sUSD did not include $USDC, the price was still affected to some extent, dropping to around $0.96 at its lowest point. However, it was quickly balanced by arbitrageurs. The related panic selling was mainly dominated by emotional factors as the Synthetix system does not have direct exposure to the risk, and the drop in the prices of non-sUSD synthetic assets during the same period actually reduces the debt (liquidation risk) of sUSD minters. Therefore, compared to stablecoin projects with a large amount of USDC on their balance sheets, the certainty of returning to peg for sUSD is higher.

Chart: Comparison of sUSD and USDC prices

In addition, there are theoretically still arbitrage opportunities within the protocol at this time, such as purchasing sUSD at a price of $0.95 in the secondary market, exchanging sUSD for other synthetic assets such as sETH in the Synthetix system at a ratio of 1:1, and selling sETH in the secondary market at a price above $0.95 to obtain arbitrage profits, provided the friction costs are low enough.

Synthetix is currently undergoing modifications for version 3.0. In V3, there will be new types of staking assets, in addition to SNX, other cryptocurrencies such as ETH can also be staked to generate sUSD. Previously, the size of sUSD was limited by the market capitalization of SNX, but after the implementation of V3, it will no longer be limited by this, and the scalability of sUSD will be enhanced. It is anticipated that Optimism will develop a more diverse ecosystem and gain a greater market share as funding flows into it.

MakerDAO

MakerDAO is a smart contract application built on Ethereum in 2014, which issues a decentralized stablecoin called DAI that pegs to the US dollar in a 1:1 ratio through a DAO model. The protocol uses various types of crypto assets as collateral and issues stablecoin DAI based on a certain collateralization ratio, essentially creating a trustless over-collateralized loan. When the value of the collateral falls below the minimum collateralization ratio (150%), the user’s collateral may be liquidated (forcibly sold to repay the DAI), ensuring that Maker does not experience a debt shortfall.

DAI’s goal is to minimize cryptocurrency volatility, but market behavior often causes DAI to deviate from the original price of $1. Therefore, Maker’s primary goal is to maintain the stability of the DAI price..

One of Maker’s tools for regulating the DAI price is adjusting its stability fee. Because the stability fee represents the interest rate users have to pay for borrowing DAI, it can influence their borrowing behavior by increasing or decreasing the borrowing rate. However, the adjustment of the stability fee is decided by MKR holders through voting, and the governance cycle is relatively long, resulting in a longer period of price control. Moreover, the actual market scenario for DAI is that demand for DAI decreases as ETH increases while demand for DAI rises as ETH declines, but the market supply rules are the opposite.

To address these two issues, Maker has designed the Peg Stabilization Module (PSM). PSM’s first implementation was the USDC PSM, which allows customers to deposit USDC and withdraw DAI with only 0.1% transaction fee at a 1:1 exchange rate. The module is a currency exchange protocol based on the fixed price of DAI, similar to a rigid redemption with a certain amount of funding, providing bilateral buffer protection for the DAI price.

PSM largely addresses DAI’s price stability problem and allows Maker to maintain control over borrowing costs without constantly changing their borrowing rates. PSM’s funding scale has expanded rapidly, making USDC the largest source of collateral for DAI. While a smooth exchange mechanism, this was exactly what caused USDC to be quickly dumped into the PSM during the most recent panic event. Currently, the debt ceiling for DAI issuance through USDC PSM has been set, and other more volatile assets (such as MATIC) have been largely redeemed, which has resulted in USDC PSM’s share of DAI issuance rising from 40% to 62%.

Image: Percentage of MakerDAO locked assets and resulting DAI debt.

There are two main types of market crises that DAI has gone through:

When most collateral falls in price due to stablecoin panic:

On March 12, 2020, the price of Ethereum plummeted 43% in one day, which resulted in significant deficits for many users who created DAI using Ethereum and other cryptocurrencies as collateral. These deficient Vaults were forcibly liquidated, and their collateral was auctioned off to repay debts and fines. Nevertheless, several auctions resulted in no bids due to market panic, network congestion, and system problems, resulting in a user winning a significant amount of collateral for 0 DAI. Due to this, there were 5.4 million DAI lost to MakerDAO, and the supply of DAI decreased significantly.

Since DAI demand far exceeded supply, the peg was broken, and DAI experienced a premium of up to 10%. There was a high risk-to-reward ratio for shorting DAI and waiting for its price to rise or fall even further at this time.

When only USDC continues to trade at a discount:

This was the first crisis in history where USDC had come off its peg. As half of the DAI was generated with USDC as collateral, this posed a significant risk to the DAI system. The price of DAI will fluctuate or become unredeemable if USDC loses its peg. As a precaution, the MakerDAO community approved a number of urgent measures to reduce the debt ceiling of several liquidity pools to zero DAI, meaning that they cannot continue to issue new tokens. In addition, the daily issuance limit of the portion exposed to USDC risk in the so-called “stability module” (PSM) was reduced from 950 million DAI to only 250 million DAI, and the fee was increased from 0 to 1%.

Liquidation is not feasible because USDC is not overly collateralized when creating DAI. Users who generate DAI using USDC as collateral will run out if the price of USDC drops below $1, thus they will have to pay more DAI to redeem their USDC. This puts them at risk of losses or being unable to exit, and the entire system may be paralyzed. Therefore, there may be another auction of MKR to make up for the shortfall, which is why the price of MKR plummeted by more than 30% during this crisis period. However, after the USDC crisis was resolved, the price of MKR quickly recovered to its pre-crisis level.

Image: Price changes of USDC and MKR during the de-pegging crisis.

In addition to the trading opportunity to recover the value of MKR, there is also a trading opportunity for DAI. DAI is a strengthened version of USDC as the overall collateralization ratio behind DAI is primarily above 150%. Due to this, the price of DAI should rise even faster when the price is below USDC and once the risk surrounding USDC has been eliminated. According to the chart below, the DAI price was always slightly higher than the USDC price as the crisis on the 11th was slowly resolved by the market.

It is also important to note that the stablecoin minting module PSM saw a huge inflow of 950 million USD during this crisis, while the safe GUSD deposits also flowed out significantly. Other collateral pools also had more or less outflows, showing the effect of “bad drives out good”. How to deal with similar structural risks is worthy of more attention and discussion.

Liquity

Liquity was launched in April 2021 as a decentralized stablecoin lending platform built on Ethereum. Users can only generate the stablecoin LUSD, which is pegged to the US dollar, by collateralizing ETH. Liquity charges a one-time minting and redemption fee to support long-term LUSD holdings rather than interest on loans. Liquity is managed by smart contracts that cannot be modified after deployment and do not handle front-end operations, so user interaction requires third-party frontends, making it highly decentralized and censorship-resistant.

Currently, the circulating supply of LUSD is around 243 million, with a TVL of $572 million and 388k ETH collateralized, resulting in a total collateral ratio of 235.1%.

The minimum collateral ratio on Liquity is 110%, and the system enters recovery mode when it falls below 150%, with liquidation triggered below 110%. In recovery mode, vaults with a collateral ratio below 150% may also be liquidated, and the system prohibits further reduction of the overall collateral ratio. The goal of the recovery mode is to rapidly increase the overall collateral ratio to above 150% to reduce system risk.

Liquity employs a graded liquidation mechanism to maintain system stability. The stability pool incentivizes users to store LUSD through liquidity mining, and during liquidation, it destroys the debt and receives ETH. When the stability pool is exhausted, the system will reallocate the debt, distributing the remaining debt to ETH proportionally to other vault owners.

Stability pool providers and frontend operators can earn governance token LQTY rewards. LQTY represents the right to claim protocol revenue (minting and redemption fees) and governance rights (voting power).

When the price of LUSD falls below $1, users can buy LUSD at a lower price from the market and then redeem ETH for a profit. When the price of LUSD exceeds $1 (e.g., $1.1), users can collateralize ETH to mint LUSD, and then sell LUSD at a higher price on the market for a profit. This way, the price of LUSD fluctuates between (1-redemption fee and 1.1) and tends towards $1, creating a stable peg mechanism for LUSD. Also, since users can mint and redeem LUSD at any time for $1, they form a Schelling point and believe that 1 LUSD = 1 USD.

Image: LUSD peg mechanism

On May 19, 2021, the price of ETH rapidly dropped from $3400 to $1800, and over 300 addresses were liquidated. Liquity initiated two recovery modes, but the data was not captured by Dune because the recovery was too fast (the collateral ratio quickly returned after falling below 150%). During this period, a total of 93.5 million LUSD debts were liquidated, and 48,668 ETH were allocated to the depositors of the stability pool. All the liquidation was completed by the stability pool, and the participants in the stability pool were able to purchase ETH at a discount. This stress test proved the robustness of Liquity’s model.

Image: Liquity System Collateral Ratio Changes, TCR = Total Collateral Ratio.

During the USDC panic, LUSD also experienced fluctuations, with a minimum of 0.96 and a maximum of 1.03. Arbitrageurs quickly brought the price back to normal. Specifically, users can purchase LUSD on the secondary market for $0.96, repay their LUSD loan in the Liquity system to redeem their collateral ETH. Since the price of LUSD in the Liquity system remains at $1, the value of the ETH asset that users can redeem is greater than the value of LUSD they purchased on the secondary market, resulting in arbitrage opportunities.

During the market crisis, the reason for the rise of LUSD was that some users needed to repay LUSD to avoid liquidation, and some users had the motivation to deposit LUSD into the stability pool in hopes of obtaining discounted ETH in liquidation. Both factors contributed to the demand for LUSD. In addition, USDC holders who rushed out of the LUSD-USDC liquidity pool also helped passively boost the LUSD price.

Price fluctuations on the day of the USDC panic resulted in arbitrage opportunities, which resulted in a significant increase in mint and LUSD burn. There were 11.22 million burned LUSD and 21.98 million newly minted LUSD. The net increase in LUSD supply was about 10 million, accounting for about 4% of the total LUSD supply on that day. As a result, the protocol’s revenue on March 11th increased significantly, with 377,000 LUSD earned in the minting process and 97.4 ETH earned in the redemption process. This revenue is fully attributed to LQTY stakers, resulting in a short-term increase in LQTY yield. At the same time, the USDC panic increased the market’s attention to the decentralized stablecoin LUSD, and the supply and trove numbers of LUSD showed an upward trend. The overall supply of LUSD increased by roughly 12% from March 11 to March 16.

Chart：LUSD mint and burn, protocol insurance income, and redemption income.

Reflexer

Reflexer is an overcollateralized decentralized stablecoin platform where users can generate RAI by staking ETH, a stablecoin not pegged to any fiat currency or asset. The redemption price of RAI is automatically adjusted by a market supply and demand-based algorithm using a PID controller to achieve low volatility. Users can mint RAI by over-collateralizing ETH, pay 2% annual interest, and redeem ETH by repaying RAI. The liquidation threshold is 145%, but currently, over-collateralization rates are between 300%-400%. Reflexer has a triple liquidation mechanism to ensure system security and charges a 2% stability fee as a surplus buffer. FLX is Reflexer’s governance token and also the final lender in the system.

When the market supply and demand are imbalanced, Reflexer proactively adjusts the RAI redemption price by incentivizing users to arbitrage and guide the market price back to the redemption price. The RAI system uses a PID control mechanism based on a series of parameters to adjust the control process mentioned above.

When the value of a user’s collateral is lower than a certain threshold relative to the borrowed amount, liquidation is triggered. The liquidator acquires the liquidated user’s ETH collateral through a fixed discount auction and repays the RAI debt on behalf of the liquidated user. When the surplus buffer is not enough to handle bad debts, the protocol enters the “debt auction” process, and the system issues more FLX to exchange RAI to complete debt processing on the market. Reflexer’s 2% stability fee is allocated to the following purposes: Stability Fee Treasury smart contract, FLX stakers, for buyback and burn.

Use cases for RAI include currency markets, stacked funding rates, yield aggregators-leverage positive/negative redemption rates, and sophisticated arbitrage tools.

Overall Reflexer has the advantages of being completely decentralized, ultimately moving towards no governance, and having a collateral mechanism that does not peg to any Fiat currencies, with positive comments from Vitalik and support from the Ethereum community. However, it lacks passive demand and use cases, and the over-collateralization rate is currently 300%-400% (357%), with low capital efficiency and token value capture, and insufficient tokens for incentives (the FLX reserved for promoting use cases may not be sufficient).

In terms of capital efficiency, Liquity is better than Reflexer (Liquity 260% collateralization rate vs. Reflexer 357%).In terms of borrowing volume, Liquity’s records is several times of Reflexer, and in terms of P/S valuation, FLX is undervalued compared to LQTY.

In this crisis, Vitalik minted RAI purchasing power for USDC and USDT, which seems to be a recognition of an entirely decentralized (non-pegged to fiat currencies) stablecoin backed by ETH. However, as RAI does not have a fixed peg price and is inherently unstable, it is still challenging for Reflexer to attract mainstream users.

Additionally, in January of this year, Vitalik suggested improvements to Reflexer’s collateral mechanism. He believes that ETH holders need more incentives to over-collateralize ETH and borrow RAI on the Reflexer platform because by staking ETH users can earn a risk-free 5% reward, and the arbitrage floating return on the redemption rate obtained on Reflexer is not very attractive when it does not exceed 5%. However, the community has rejected the proposal to use ETH as collateral due to additional contract risk considerations.

Celo

Celo is a mobile-focused open-source payment network. Its mainnet was launched in April 2020. The network combines a PoS mechanism with EVM compatibility, providing users with various DeFi services. These include using stablecoins for remittance and cross-border payments, supporting multiple token payments for gas fees, and mapping phone numbers to wallet addresses to simplify transfer operations. Celo has also introduced its stablecoins CUSD, CEUR, and CREAL on-chain.

The stablecoin mechanism of Celo works as follows: users can send 1 dollar worth of Celo to the official Mento pool and receive 1 dollar worth of stablecoins such as cUSD. Conversely, they can also send 1 dollar worth of cUSD to Mento and receive 1 dollar worth of Celo. Under this mechanism, when the market price of cUSD is lower than 1 dollar, someone will buy cUSD at a low price to exchange it for 1 dollar worth of Celo. Similarly, when the price of cUSD is higher than 1 dollar, someone will mint cUSD using Celo and sell it, and the existence of arbitrageurs will ensure that cUSD does not deviate too far from its peg price. Currently, the reserve pool behind stablecoins is CELO (81.9 million USD), ETH (48.89 million USD), and BTC (7.91 million USD).

The difference between the Celo mechanism and the LUNA/UST mechanism is that the funding for CUSD, cEUR, and cREAL is backed not only by CELO but also by ETH and BTC. Therefore, as long as the asset size of ETH and BTC in the reserve pool is greater than the circulating market value of the stablecoins, even if the price of CELO tokens fluctuates greatly, the stablecoins will still be over-collateralized, and there will be no risk of de-peg. The market value of ETH and BTC in the reserve pool may not be sufficient to cover the circulating market value of stablecoins if CELO’s market value declines quickly, which might result in stablecoins becoming unpegged.

Consequently, as long as CELO’s price does not decline by more than 50%, the risk of stablecoin de-peg is fairly low. It will increase the possibility of depegging, though, if the size of stablecoins keeps growing in the future and the size of BTC and ETH in the reserve pool does not rise as a result.

Looking at the historical volatility of CUSD, except for the market panic caused by the FTX incident last year, the overall price has remained above $1 for most of the time.

Generally speaking, there aren’t many external factors that can affect Celo’s stablecoin mechanism, and the primary factor is the asset size of ETH and BTC in the reserve pool. In addition to the fact that CUSD is mostly circulated on the Celo chain and is not directly connected to USDC, the recent USDC de-peg has no direct impact on it.

Frax Finance

A core component of the FRAX stablecoin system, the Algorithmic Market Operations (AMO) controller, was introduced in December 2020 and is currently in version 2.

Although the FRAX token was originally intended to have 100% USDC as collateral, it was later transformed into a mixture of USDC and FXS, gradually reducing the proportion of USDC collateral. As of February 2023, the community voted to permanently increase the collateral rate to 100%, suspending FXS buybacks until sufficient revenue is generated by the agreement. The protocol’s collateral rate is currently 92%. A lending market called Fraxlend was also introduced by Frax Finance in September 2022, allowing users to borrow FRAX against their assets, similar to the MakerDAO protocol. The difference is that FRAX holders can deposit FRAX into the lending pool and earn interest on their loans.

Without lowering the collateral rate or altering the FRAX price, the AMO is designed to create FRAX monetary policy and invest in reserve assets, improving capital efficiency and capturing more value for FXS holders. Following the activation of the mechanism, the scale expansion of the stablecoin is largely under the control of the AMO. Currently, Liquidity AMO (multi-chain DEX providing liquidity), Lending AMO (lending pool), Investor AMO (investment), and Curve AMO are the core AMO pools (Curve ecosystem). About 800 million FRAX tokens are under the control of the protocol.

The recent USDC event had a significant impact on the FRAX price, which dropped to a low of $0.87. The main reasons were that 1) Since USDC+FXS has 92% of the actual value of the FRAX algorithmic stablecoin, and after the community voted to raise the collateral rate to 100%, only 100% USDC is utilized as collateral. 2) The FRAX amount in the Curve AMO LP pool is 270 million, and the core pools are FRAX/USDC and FRAXBP (FRAX/USDC/USDT/Dai).

In spite of the fact that Frax Finance did not make any public announcements of any remedial measures during this event, the team informed the public that the long-term plan is to address such risk concerns by establishing Frax Market Assurance (FMA).

Due to the fact that the collateral is a stablecoin and the collateral rate will eventually rise to 100%, it is not possible to increase the efficiency of user capital. There are few other FRAX use cases on the market right now, and Curve and Convex are FRAX’s main ecosystem collaborations. Overall, there is a low demand for FRAX in the market.

Since FRAX tokens are not commonly used on other dApps, most arbitrage activity takes place on Fraxlend, where users who have borrowed FRAX tokens can purchase them at a discount in the market to pay off their loans.

Image: Changes in the prices of FRAX and USDC before and after the de-pegging crisis.

Despite this, its governance token also experienced a sell-off with a decline of more than 20% during the crisis due to the uncertainty brought on as a result. Like MKR token, the reason for the selloff might be due to potential risks of system imbalance and token inflation. However, after the USDC crisis was resolved, FXS prices rebounded rapidly by more than 40%, similar to MKR.

Chart: Changes in the prices of governance tokens FXS and USDC before and after the de-pegging crisis.

In addition, ETH 2.0 liquidity staking is a feature of the Frax system. In the midst of the market turbulence, node revenue increased more than five times around March 10 due to the rise in front-running transactions.

Chart: frxETH node income. Blue is the basic reward, and green is the fee “tip” income.

For example, on March 10th, a USDC user who exchanged 2 million USDC for USDT on-chain, and a MEV bot unexpectedly made a profit of $2.045 million after the user paid $45 in gas fees and a $39,000 MEV bribe because the user did not set a slippage limit. Unfortunately, the user traded 2.08 million USDC but received only 0.05 USDT.

GHO

The Aave lending protocol supports any kind of lending and borrowing among various cryptocurrencies. The team announced its intention to introduce stablecoin GHO in July 2022. Based on over-collateralized lending, the overall mechanism for creating GHO is similar to DAI, which is maintained at a 1:1 peg to the US dollar. Essentially, the underlying logic is that users deposit supported collateral types into the vault and receive GHO stablecoins based on a certain collateralization ratio. In the event that these stablecoins are repaid or liquidated, they will be burned.

In order to generate more revenue, Aave first started its stablecoin business because stablecoin minting fees are typically higher than interest income from general lending and borrowing, which can generate extra profits for Aave. The money will also go straight to the vault.

Aave’s current lending and borrowing business is relatively mature, and compared to the difficulty of promoting stablecoins for other newer protocols, Aave has some market credit endorsements. Secondly, Aave’s current protocol version has been iterated to V3, and the V3 upgrade plan has not yet been fully deployed. V3 will also provide significant support for stablecoins.

One of the unfinished plans is the efficient lending market (e-Mode), which allows borrowing against the same type of collateral with a higher LTV. Firstly, borrowing between ETH assets (WETH, wstETH) can achieve a 90% LTV, meaning that originally collateralized $2,000 WETH can now borrow up to $1,800 wstETH, improving the efficiency of capital utilization. The next step is to implement e-Mode lending between stablecoin assets (L2 has been deployed, but Ethereum mainnet has not), with an LTV of up to 93%. This implementation will greatly improve the efficiency of using stablecoin assets. In the future, after the launch of GHO, it will have a positive effect on the stable price of GHO and its adoption. However, it faces the same problem as MakerDAO PSM, which is the regulatory issue for centralized stablecoins within the module.

The second unfinished plan is cross-chain deployment. Although Aave has already been deployed on multiple chains and maintains its advantage as a top DeFi lending protocol, it has not yet achieved true cross-chain liquidity. After the implementation of Aave V3’s cross-chain deployment, there will be some advantages for the expansion of GHO. Aave V3’s “gateway” feature allows assets to flow seamlessly between V3 markets on different networks, meaning that “aTokens” can be minted on the target chain through a whitelisted cross-chain bridge protocol, and then burnt on the source chain, to achieve the goal of liquidity transfer from the source chain to the target chain.

Image: Top Protocols on various chains

Now, the V3 cross-chain gateway whitelist proposal has been voted through. Aave will use the cross-chain interoperability protocol Wormhole for general message passing between the source and target chains, and use Hashflow’s cross-chain DEX to obtain quotes from market makers with zero slippage and MEV protection to execute transactions.

CRVUSD

In October 2022, Curve published its white paper for the stablecoin protocol, Curve.Fi USD Stablecoin, also known as crvUSD. In order to provide continuous liquidation and purchase collateral, Curve designed the LLAMMA model. Users can over-collateralize to mint crvUSD in LLAMMA and avoid the risk of their assets being liquidated all at once. The relationship between crvUSD’s supply and demand in the market is stabilized by Pegkeepers’ automatic issuance and destruction mechanisms.

The official launch of crvUSD has not yet taken place. However, if a major stablecoin such as USDC were to de-peg in the future, based on the design mechanisms Curve has disclosed, the following situations could occur (assuming USDC de-pegs once more):

1.In theory, crvUSD is over-collateralized with mainstream assets such as ETH and BTC, so if a major stablecoin de-pegs and the collateralized asset prices remain unaffected, crvUSD should not experience direct price fluctuations.

2.Although crvUSD is minted through over-collateralization, its price stability still depends on the liquidity depth of the market pool. As the official stablecoin of Curve, crvUSD may enter the 3pool in the future. Curve’s team holds a large amount of veCRV and can provide significant liquidity incentives for crvUSD early on. Hence, a comparatively smooth start should be made. If crvUSD, USDC, and USDT are all in the 3pool as USDC de-pegs, users may panic and convert large amounts of USDC to USDT and crvCRV, causing crvUSD’s price to temporarily exceed $1. To maintain the stability of crvUSD, Pegkeeper will issue a large number of crvUSD to the market pool, which will restore the price. Therefore, compared to other stablecoins, the mechanism of crvUSD may not be suitable for slippage arbitrage because Pegkeeper’s automatic issuance function will quickly reduce the arbitrage opportunities.

3.If a major stablecoin de-pegs, causing significant market panic and leading to massive liquidation in the cryptocurrency market, including LLAMMA, which converts user collateral into crvUSD, crvUSD may also experience a price de-peg due to panic. However, if users believe that Pegkeeper has funds in the market to buy crvUSD and burn them, thereby restoring the price when crvUSD falls below $1, there may be potential arbitrage opportunities.

4.Since LLAMMA is still an AMM-based mechanism, LPs consisting of stablecoins and collateral may exist in the liquidation pool. When prices fluctuate, on the one hand, the frictional cost of collateralized assets due to continuous liquidation and purchase can cause wear and tear, which is difficult to avoid for collateral providers. On the other hand, LPs may receive a certain amount of transaction fee subsidies from the continuous liquidation process, especially when there is a significant liquidation on the chain, the trading volume of LLAMMA may increase in the short term, and 50% of these fees may be given to LPs.

All of the above four points are based on the current Curve mechanism disclosed and require further testing in the market after the official launch of crvUSD.

Impact and opportunities of borrowing and trading agreements

Image: Price changes of the four largest stablecoins and TUSD after USDC de-peg

The largest stablecoin on the market, USDT, remained the top choice for capital outflows despite the USDC de-peg crisis and had the strongest price performance, with a growth rate of one point even exceeding 1%. As approximately 50% of DAI’s collateral is in USDC, the continued decline of the USDC market price caused DAI to experience severe de-pegging as well, although to a lesser extent than USDC.

Despite recent additions to Binance’s stablecoin trading pairs, the price performance of BUSD and TUSD has remained relatively stable without experiencing price fluctuations exceeding 1%.

Measures taken by different protocols

In this USDC de-pegging crisis, besides a large number of people “running for the exits” to reduce their losses, there is also a group of arbitrageurs who have a deep understanding of mainstream DeFi project mechanisms. The core idea is to use protocol mechanisms for USDC by linking the value of USDC to $1, seeking to exchange USDC when its price is lower than $1 for other stablecoins or cryptocurrencies with a value greater than $1, successfully booking a profit. This event will also bring about reflection on current and future projects, whether the price of all trading targets can be replaced by a fixed value instead of the actual market value. The following are several cases for reference.

PSM (Peg Stability Module) of MakerDAO

The PSM is a mechanism designed to help maintain the stability of the DAI price. It allows users to exchange other tokens for DAI at a fixed rate, without having to go through an auction. USDP is a stablecoin issued by PAXOS and was not affected by the recent crisis. When USDC and DAI experience a discount, the PSM price mechanism is applied to take advantage of the arbitrage opportunity: 1 USDC = 1 USDP.

Arbitrage path: Use assets to borrow USDC on AAVE, then exchange USDC 1:1 for USDP on the PSM. When the price of USDC drops further, exchange USDP back to USDC, repay the USDC loan, and keep the remaining USDC as profit. As shown in the figure below, during the USDC de-peg crisis, USDC experienced a significant inflow while GUSD and USDP quickly experienced outflows, causing a shortage of USDP supply.

Image: Changes in the three different stablecoin collateral in the Maker system during the USDC de-peg period.

Curve

The Curve price formula is designed for stablecoin trading, with lower slippage for exchanging large amounts of stablecoins. When using the algorithm for non-stablecoins, this model provides greater trading depth than UNI at the current order book mechanism, resulting in price differences with other markets, with Curve’s USDC/USDT price > UNI’s USDC/USDT price. Similar to traditional arbitrage, one can buy USDC on UNI and sell it on Curve to earn the price difference, but there is a risk of rapid USDC price drops in between.

AAVE

In the case of AAVE’s long ‘leverage looping’ of USDC, the buyer makes the assumption that the USDC price will eventually re-peg and is willing to take on the risk of further USDC price drops. The specific process involves borrowing USDC using asset lending (borrowing USDT and then trading for USDC through a DEX), collateralizing USDC to borrow USDT, using USDT to purchase USDC, and then using USDC to borrow USDT again in a loop.

Centralized exchange arbitrage

Binance supports USDC deposits, and USDC can be exchanged 1:1 for BUSD, completing the exchange of low-value USDC for high-value BUSD. Coinbase users can withdraw USDC 1:1 as USD into their bank accounts to complete the arbitrage process. However, both exchanges react quickly and the arbitrage process lasts for a short period of time.

Depending on USDC’s collateral issues and USDC’s de-peg price, several types of DeFi applications may be subject to the following risks:

Risks faced by lending protocol

Aave and Compound are the two largest on-chain lending protocols, allowing users to borrow and lend various cryptocurrencies and earn interest income or pay interest fees. USDC is one of the commonly used assets on these platforms. If a user uses USDC as collateral for borrowing, they may face the risk of being unable to repay or being liquidated. Aave uses Chainlink as the oracle for USDC value calculation, and Uniswap as the backup oracle. For users who borrow USDC by depositing ETH, they can take advantage of the opportunity to buy back USDC with less capital to redeem their original collateral. Compound, on the other hand, uses a fixed value of $1, which may lead to risks.

Specifically, when the USDC price de-pegs, lending protocols may have some risks associated with lending protocols:

These risks can be mitigated through the following measures:

For trading protocols: If USDC is a trading pair or for a liquidity provider, they may face the risk of loss. For example, on the Uniswap platform, users can trade between any two cryptocurrencies or provide liquidity to earn fees. If users use USDC as a trading pair or one of the currencies in the liquidity pool, they may lose money due to the deterioration of the exchange rate or inability to exit when the USDC price falls. In addition, there are some default USDC applications in derivative trading applications that bring arbitrage opportunities.

Taking the derivative protocol GMX as an example

GMX is a decentralized perpetual exchange built on Arbitrum. In the trading mechanism of GMX, users take long positions and receive the underlying assets, rather than the margin itself, which is priced in USD. Arbitrageurs use discounted USDC to open positions and are paid according to the standard USD price.

After this incident, the protocols of GMX and similar mechanisms may optimize the stablecoin price curve, calculating the actual value of the stablecoin used as margin by users when placing orders.

In GMX’s Swap, the price of USDC/ETH is based on the price of ETH/USD, creating arbitrage opportunities in the Swap market as well.

Moreover, such leveraged trading platforms received 2–3 times as much fee income as usual as a result of the market turbulence.

Image: Revenue of the four major on-chain trading platform

Prediction applications: If users use USDC as a betting or reward currency on prediction platforms such as Augur, they may face the risk of reduced payouts or profits if the USDC price drops. For example, users can create and participate in various market prediction events on the Augur platform and win rewards based on the results. If users use USDC as a betting or reward currency, they may lose money if the payout amount decreases or their profits shrink due to a drop in the USDC price.

Summary

During the USDC de-peg crisis, the discount of USDC brought about significant uncertainty to the DeFi ecosystem, but it was eventually resolved by government and regulatory intervention. In this process, there were both one-way bets on the price re-pegging (ie. a wallet receiving 215 million USDT from Binance to purchase USDC and DAI stablecoins, earning about 16.5 million US dollars in profits); low-risk arbitrage exploiting mechanism loopholes (such as the Maker PSM arbitrage mentioned in the article); panic selling of various assets at discount prices (such as BUSD and USDP mentioned in the article); in addition, the passive income increase brought to leveraged trading and LSD applications due to the significant market fluctuations cannot be ignored (such as Frax and crvUSD can capture relevant income). These applications themselves have small USDC risk exposures and instead become passive beneficiaries.

Read more: https://ld-capital.medium.com/trend-research-by-ld-capital-reviewing-the-dangers-and-opportunities-in-the-usdc-de-peg-crisis-3ba1827f43cc