In the midst of a general downturn in the cryptocurrency market, the most severely affected appears to be CRV.
This morning, Arkham disclosed that Curve founder Michael Egorov currently has $140 million worth of CRV collateralized in stablecoins (mainly crvUSD) borrowed against $95.7 million in 5 accounts across 5 protocols. Among them, Michael has borrowed $50 million in crvUSD on Llamalend, with Egorov’s 3 accounts holding over 90% of the borrowed crvUSD on that protocol.
Arkham pointed out that if the price of CRV falls by approximately 10%, these positions may start to be liquidated. Subsequently, as the CRV price continued to decline, dropping below $0.26, hitting a historic low, Michael’s CRV borrowing positions on multiple addresses gradually fell below the liquidation threshold.
In the past, Michael would typically add more collateral to save his borrowing positions, however, this time, he seemed to have “given up.”
According to Ember’s monitoring, part of Michael Egorov’s primary address on Inverse is starting to be liquidated. Additionally, according to Lookonchain monitoring, Michale Egorov currently holds 111.87 million CRV ($33.87 million) in collateral and $20.60 million in debt on 4 platforms.
Two months ago, a liquidation crisis began to emerge with CRV when Michael’s borrowing positions fell below the liquidation threshold. However, at that time, Michael was not liquidated, and there was no sign of any remedial action from Michael.
On April 14th, as the market declined, the CRV price dropped to $0.42, and Curve founder Michael Egorov’s borrowing positions once again entered the red zone. According to Ember’s monitoring, Michael collateralized a total of 371 million CRV through 5 addresses on 6 lending platforms, borrowing $92.54 million in stablecoins. Among the 12 debts, the silo lending position was the riskiest.
Since November 2022, the “on-chain bear” ponzishorter has been attempting to short CRV, until the end of July 2023, when Curve was attacked due to a Vyper compiler bug, Michael took frequent actions to save his positions, stirring up the DeFi pot. This series of actions has been likened to a “DeFi defense battle.”
The first “defense battle” may have been Michael’s bear trap, preventing the CRV price from falling and profiting in the battle against the “bears.” The second “defense battle” leveraged the off-exchange OTC power, reducing its holdings, but gaining the support of influential players like Wu Jihan, Du Jun, Justin Sun, and institutions like DWF. It can be said that both of CRV’s defense battles were quite successful.
For more information, read “Curve Issues, a Symptom of DeFi’s ‘Profit Disease.'”
On April 14th at noon, with CRV prices falling to $0.42, according to debank data, 5 of Michael’s 12 positions had a health factor of 1.12 or lower. Ember detected a red line in Michael’s debt position and speculated on liquidation, pointing out that if the CRV price continued to drop by 10% without adding collateral or repayment, the liquidation process would be initiated.
However, while people were pondering how CRV would respond to a third “DeFi defense battle,” an interesting event occurred. It was observed that in the early hours of the morning at 4 o’clock, the CRV price briefly dropped to $0.3592, already below the 10% drop from $0.42. However, Michael’s debt positions were not liquidated as predicted by Ember, and he did not seem to take any remedial action.
Michael’s debt positions are spread across 6 different lending protocols, with the most controversial being silo. After the attack on Curve, because most lending protocols were unwilling to bear too much risk associated with CRV, policies were tightened, and over half of the loans raised by Michael came from silo. During the repayment of AAVE debts, silo provided almost all the necessary loans. Silo became Michael’s biggest ally on the road to debt repayment, with many community members jokingly calling it Michael’s “personal bank.”
At that time, within Michael’s total debt position, approximately 113 million CRV was deposited in the silo protocol, borrowing approximately $27.90 million in stablecoins, accounting for 30% of Michael’s total debt position. However, Curve LlamaLend, UwU Lend, and FraxLend protocols also provided most of the loans to Michael, although not as high a percentage as silo, they still accounted for over 15%, with Curve LlamaLend at 20.7%, UwU Lend at 17.9%, and FraxLend at 17.3%.
On the other hand, silo once forked out a new protocol, Silo Llama, designed specifically for crvUSD isolation lending. Although this protocol was met with skepticism, DeFi should be designed without regard to user emotions. In comparison to borrowing, the impact of CRV’s lockup rate on the sale of CRV is more significant. Setting up a separate pool for CRV in the DeFi lending treasury is one of the working methods, and the silo team members have explicitly denied the accusation of “creating Silo Llama for one person.”
Setting aside the relationship between silo and curve, the essence of the debate is that silo did not liquidate CRV. Informed sources indicated that due to the use of Chainlink oracles for CRV positions on silo, there is a delay in price updates compared to debank, so there is doubt as to whether the oracle tracked the liquidation price.
According to Chainlink data, the recorded CRV price on April 14th at 5:30 am fell below $0.4, ranging from $0.36 to $0.38. Subsequently, the author reviewed data from dexscreener, coingecko, tradingview, coinmarket, and other sources, showing that CRV fell to around $0.36 in the 30-minute timeframe.
As the lowest price for CRV occurred in the early hours of the morning, the author could not verify whether the health factor was reset at that time. However, regardless of what happened between CRV and various lending protocols that night, the only certainty is that all of Michael’s debt positions remain intact.
This crisis has brought attention to silo’s manual liquidation mechanism. Since silo’s liquidation is entirely open, liquidators can choose between manual or automated liquidation. When asked if opting for manual liquidation would prevent automated liquidation, an informed source explained that manual liquidation is just a personal liquidation entry provided by the platform. When faced with a debt to be liquidated, individuals still need to compete with the machine and often cannot outperform it.
Therefore, the key to triggering liquidation lies in whether the collateral price has truly dropped to the liquidation value.
The “Price Game” of the Liquidation Mechanism
According to silo’s documentation, the lending protocol has a liquidation application that the core team uses to monitor risk positions and liquidate positions that are undercollateralized if the liquidation robots (including Silo) do not liquidate them first for any reason.
On April 19th, CRV once again fell to $0.4, and according to Ember’s provided Michael addresses, the addresses starting with 0x 9, 0x 4, and 0x 7 had health factors below 0.1, indicating a dangerous situation in silo’s debt positions.
According to silo finance’s collateral factor table, in the silo protocol, CRV has a loan-to-value ratio (LTV) of 65% and a liquidation threshold (LT) of 85%. This means that Michael’s silo liquidation price is in the $0.41 to $0.44 price range, theoretically resulting in a health factor of 0.
The calculation formula is: Liquidation price = total borrowed amount / (collateral amount * LTV * LT) Health factor = 1 – total borrowed amount / (total collateral amount * LTV)
In response to this, BlockBeats reached out to the project team, and they clarified that their price tracking does not simply rely on oracle prices but uses a weighted average algorithm. This means that the liquidation price of a token can be influenced by the prices of other assets borrowed by the borrower, so a drop in the CRV price alone is not sufficient to trigger liquidation. However, the project team did not respond to questions about liquidity supply issues.
Furthermore, an informed source told BlockBeats that in volatile market conditions, liquidators need to consider slippage issues, which also involve the slippage of crvUSD and CRV. In several previous instances of significant price fluctuations, it is normal for lending protocol automated liquidations to remain inactive.
Was Michael “Cut” this Time?
The liquidation of tens of millions of dollars in debt positions has a significant impact on the overall liquidity of the crypto market. The crisis in April could be avoided due to the defense mechanisms of lending platforms, but this time, with CRV falling below $0.26, the crisis has finally arrived.
Liquidators Profit
Whether to buy the price at its new low is a topic of concern for investors, but at least in the case of CRV, liquidators have already started to profit.
According to monitoring by ai_9684 xtpa, the address starting with 0xF07…0f19E is one of the primary liquidators of Michael’s positions. In the past hour, this address has liquidated 29.62 million CRV at an average price of $0.2549, spending a total of 755,000 FRAX. Currently, these tokens have all been recharged into Binance at an average price of $0.2792.
For liquidators, a more economical approach may be to first open a CRV short position (or borrow and sell) on Binance before liquidating. This way, the tokens obtained from liquidation would only be used to close the short position (or repay the debt), without having to bear the profit or loss caused by price fluctuations during the period.
However, even if 0xF07…0f19E does not take this approach, they can still make a profit of $720,000 simply by selling at the average recharge price.
Investors Incur Losses
On the other hand, investors are facing a disaster.
On the one hand, the decline in prices has triggered liquidations on other lending platforms, with borrowers on Fraxlend suffering millions of dollars in liquidations. According to Lookonchain monitoring, a user was liquidated for 10.58 million CRV ($3.30 million) on Fraxlend.
Compared to this, Fraxlend’s liquidation mechanism is easier to trigger, as its risk isolation and dynamic interest rate mechanisms allow Michael to proactively repay the debt without requiring additional measures. In previous liquidation crises, Michael borrowed a significant amount of assets from Aave and sold tokens through OTC to repay his debts at Fraxlend.
On the other hand, early CRV investors are facing significant losses.
Since the CRV crisis last year, the community has been full of comments about “Curve’s good cards being shattered by Michael.” The most noteworthy aspect of this CRV crisis is the major investors who previously helped Michael.
After Curve was hacked at the end of July last year, various OGs, institutions, and VCs all rallied to support. Wu Jihan, co-founder of Bitmain and Matrixport, posted on social media, “In the upcoming wave of RWAs, CRV is one of the most important infrastructure. I have bought the dip and locked it in, not financial advice.”
Huang Licheng confirmed on social media that he purchased 3.75 million CRV from Curve’s founder through OTC and pledged it in the Curve protocol. The next day, an address related to Justin Sun received 2 million USDT from Egorov’s address and obtained 5 million CRV.
Following this, projects like Yearn Finance, Stake DAO, and institutions like DWF participated in the efforts to save CRV.
Now that CRV has fallen to a new low, Michael himself has not made any statements about rescuing the situation. This turmoil is as the community says, “The one who cuts others has finally been cut by Michael.”
