Malicious shorting and big investors smashing the market?
In the early morning of May 19 Beijing time, the price of XVS, the token of the lending protocol Venus on BSC, began to show a dramatic increase, from a low of $70 to a high of $144 in less than a few hours, but in the subsequent price began to show a huge retreat, with the lowest price as low as $31, a 78% drop from the high point.
According to the address of the large account provided by the community user: 0xEF044206Db68E40520BfA82D45419d498b4bc7Bf, which is also the address of the largest position in vXVS (XVS pledged deposit certificates in Venus), by checking the records of this address on BscScan, it can be found that the address started at 0:00 am on the 19th on the chain as well as the CoinExchange to buy a large number of XVS tokens, and because of the lack of liquidity of XVS on the market flow, the large number of purchases in a short period of time led to a further increase in price.
Crypto KOL Wang Dayou said on Twitter that the large investor pledged close to 2 million XVS at the high price point (estimated to be around $140) and then lent 4,100 bitcoins and 9,600 ethereum for an estimated $200 million in debt, which later led to a serial liquidation.
The community’s biggest push regarding this event is that XVS is collateralized at 80%, and such a setting comes from the recent VIP-22 proposal passed on May 8 to increase the lendability factor of some collateral, such as mainstream coins BTC, ETH, BNB, XVS and its token SXP from Swipe, the development team behind it, and stablecoins USDT, BUSD, USDC, the proposal Inside it explains that this behavior is because these assets have enough liquidity to improve Venus’ capital utilization and enhance its competitive advantage.
Although the increase and collateralization rate can improve the utilization of funds, it also increases the risk of the Venus system, as an 80% collateralization rate means that a 14% drop in collateralized assets could trigger the liquidation of the system.
In the DeFi lending system, there is generally a concept of health factor, which is generally related to the borrowing amount and collateral of the account, and can be expressed as follows: Health Factor = ∑(Collateral * Liquidation Threshold)/(Borrowing Amount + Interest on Borrowing)
When the health factor <1, it will trigger the DeFi lending system to liquidate the collateral of the user’s account, liquidation of individual users is to avoid the occurrence of systemic financial risk, so the borrower will be punished by the system, while the liquidator will receive a certain incentive, in the recent VIP-19 proposal just passed on May 3 is in the original 10% incentive adjusted to 15%, further incentivizing the Liquidators to participate in this liquidation, and because the collateral XVS price from $ 70 to $ 140 this process belongs to the individual operation, the market does not have much consensus, so in the liquidation process the price of the collateral intensified decline to lead to the final 100 million dollars of bad debt.
So was it a problem with the prophecy machine, Compound had previously used only the price of a single DAI on Coinbase resulting in a $90 million liquidation, so Venus also changed the original feed price in Prop 22 from having the original Band to Chainlink, and in this case both the chain and the exchange did reach that price.
More than once
The entire liquidation process was also carried out for several hours, by observing the previous transaction records of this address can also be found, he as the first XVS position address, has been cycling supply XVS, and then lend XVS, cyclic operation, until the lending limit is used up, there is no risk of bursting the position. According to the token economy model of the Venus platform, 79% of the total XVS tokens will be reserved for community mining, and 35% will be allocated to the lending pool, i.e., to users who pledge assets and receive interest, and 35% to the borrowing pool, i.e., to users who lend assets and pay interest, and because XVS has a lending limit, most of the XVS is borrowed by this large account. while enjoying an annualized return of 121%.
And inside the latest proposal, the official lowered the output of XVS, and instead used a portion of the VRT as an incentive, which was officially adopted starting May 16, which is also mentioned inside the roadmap of 2021. It may be because this large investor is not optimistic about the market, want to rush to dump the assets in their hands, but because of the XVS depth difference significant selling easy to lead to shrinkage of assets, but also because of the depth difference, so choose to pull up the price through a short period of time in such a way to complete a successful sniping.
Such means can’t help but think of this January, a cross-chain asset protocol Canno tokens on the Coinan smart chain on Swipe Wallet (Venus behind the development team) on public sale, and then Venus official tweeted that it will go online and support CAN tokens. After that, the price of CAN was pulled to $0.35 on DEX, which only cost a few dozen ETH because there was not much liquidity.
Then there was a flow of 448 million CANs into the Venus platform, collateralized by $0.35 (worth $157 million), lending 3,000 bitcoins and 7,000 ETH and other assets, and then Venus incurred about $100 million in bad debt.
What can we learn from this incident?
Again, it seems that the project didn’t learn anything from this incident, again because it pledged illiquid assets, and again because it borrowed at a high point by pulling up collateral.
First of all, the choice of collateral, in the traditional financial market, the bank president and the board of directors have the power to choose the type of assets to be collateralized because they are big enough to do so, and that is why the CBRC exists. After the famous Anbang Group acquired Chengdu Agricultural and Commercial Bank, it sat on the board of directors and began to let the bank sell its own insurance and later redeposit the premiums in its own bank. In such layers of nesting completed two trillion financial empire, of course, the end we all know very well, of course, the end we all know very well, maturity mismatch, resulting in liquidity crisis, the king this beginning, will be this end.
If a project party can choose a collateral at will, then the decentralized world becomes more of a haven for evil doers, where there is no regulation and even theft can be supervised. In a way, the project owner and the big players have become another center. As a community-governed project, it is more important to give the choice of collateral to the community when the project is mature, and to set up a mechanism to prevent the evil-doers.
For different collaterals, design reasonable deposit cap as well as borrowing cap, such as poor depth collaterals need to have some reduction in collateral coefficient when borrowing more liquid tokens.
Secondly, high collateral rate can certainly bring higher capital efficiency, but the price volatility of the collateral and the market trading volume should be fully considered. XVS saw even a 30% price drop in a short time after completing the VRT snapshot. Compare to the well-known lending agreements on Ether, such as Aave and Compound for their own tokens, the collateral rate has been maintained at 60% up and down, not to mention that the circulation of the tokens of the two is decentralized enough, and the market depth is much greater than XVS.
As well as being able to take into account all possible scenarios in liquidation, such as the lending protocol Liquity, although the collateral rate is 110% (typically 150%), implemented a liquidation process based on a pool of stable coins, a mechanism that provides a natural way for a loan to be given directly to the acquirer if the collateral value drops below the minimum collateral rate, allowing the system to use their presence funds in the stabilization pool to collect the debt instead of the borrower. In this way, the collateral surplus or excess collateral is a profit for the acquirer or a loss for the borrower.
At the same time, whether the feeding process of the prophecy machine can be supported by more dimensional indicators, such as trading depth, trading volume, rather than just providing prices. The core factors that determine the stability and robustness of the prophecy machine node quotes are: the depth of the Cex quotes and the structure of participants, as well as the efficiency of Dex’s capital utilization. The deeper the Cex quotes and the richer the structure of participants, the more representative the acquired prices will be The deeper the Cex, the richer the structure of participants, the more representative the prices obtained, and the more difficult it is to do evil and manipulate. The deeper the Cex, the richer the participant structure, the more representative the prices are, and the more difficult it is to commit fraud and manipulation.
As well as for poorer depth of collateral, lending platforms also have to consider the time factor, such as MakerDAO In order to prevent attacks on the feeding process, MakerDAO Oracle Security Module (OSM) will delay the release of the new reference price by one hour, of course, in the very fast falling market, faster prophecy machine offer, can be greater to reduce the risk of bad debts.
In short, DeFi is still a very emerging industry, after each risky attack, are paving the way for a larger amount of money in the future, from each incident to learn lessons, and then apply them to the next version, to improve the security of the entire DeFi world, to prevent systemic risk, which is the main wealth left to us by this event.
Posted by:CoinYuppie，Reprinted with attribution to:https://coinyuppie.com/revisiting-venus-coldest-day-200-million-liquidation-revelation/
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.