How did the crash of LUNA and UST happen? What are the key factors in the resulting disaster? Let Jump Crypto, a big cryptocurrency whale with heavy holdings and heavy losses this time, review the scene of the three-day death and the eyebrows during the review process. This article comes from the research report “The Depegging of UST” by Jump Crypto, a top cryptocurrency investment institution, compiled and organized by Wu Blockchain, a columnist in the dynamic zone. (Recap:Lazy Pack | The whole story of UST Crash” $84 million, leveraging a 40 billion Defi financial empire) (background supplement:Instead of wondering when the bear market will end, let’s take a look at A16z DCG Three Arrows… How the “Top 10 Cryptocurrency Institutions” work)
(Note: The time quoted in this article is GMT time, which is 8 hours slower than Taiwan time (GMT+8))
During the May 7-9 period, UST initiated a decoupling that is gone forever, and after reviewing the publicly available transaction data, Jump Crypto’s research team has three key takeaways:
The beginning of the decoupling was triggered by a series of extraordinary transactions on the UST/3CRV Curve pool on May 7. (minor decoupling, up to 1.3% deviation) From Saturday, May 7 to Monday, May 9, outflows from Anchor, especially from a small group of large traders, have given UST’s peg a huge pressure. (1% decoupling pressure) On Monday, May 9th, a strong sell-off in the cryptocurrency market added additional pressure, culminating in a non-exchangeable decoupling of UST. (Started to decouple sharply, decoupling 25% on the day, and the price of UST fluctuated between $0.25 and $0.95 in the following days)
Further reading:The Luna group destroys the major whales” South Korea’s Hashed loses 3.6 billion magnesium! Galaxy Three Arrows Pentera DeFinance Jump… suffers billions of losses
Further reading:Anchor’s founding team was exposed to “you know it will collapse before it goes online”, Luna quit TFL before UST’s 20% annualization
The chart below is a chart of UST’s peg deviation before and after the period:
May 7: Changes to the Curve Pool
On Saturday, May 7th, liquidity conditions gradually deteriorated, especially on Curve. That day, within 75 minutes, a series of key transactions took the lead to unanchor UST. At 21:44 GMT, Terraform Labs (TFL) removed 150 million UST, which, while restoring relative equilibrium to the Curve pool, reduced liquidity depth. At 21:57, a relatively inactive wallet (Wallet A) exchanged 85 million UST for USDC (the largest transaction ever made by the Curve pool), again breaking the balance of the Curve pool. 22:32~22:38: Another wallet (Wallet B) throws a total of 75 million UST through three transactions. At this time, the Curve pool is already in an extremely unbalanced state. 22:52: TFL took another 100 million UST, but it was still in a more serious imbalance, but more importantly, the liquidity depth of the capital pool was worse at this time, and it was more affected by small transactions. After 23:00, the Curve pool transaction began to riot. For details, please refer to the appendix of the original text. Additionally, Wallet A transferred $108 million in UST to Binance earlier in the day, and these transfers happened to be related to Binance’s increased trading volume and the deterioration of Curve’s liquidity.
The following picture is the peg deviation chart of UST on Binance and Curve on this day:
Trend chart of the size of the Curve fund pool:
The imbalance trend of Curve’s fund pool and the trend of UST’s trading volume on Binance:
Therefore, taking into account the above facts, one assumption we infer is that Wallet A sold UST on Binance, and this relatively one-sided selling pressure manifested itself in the deterioration of the liquidity of UST in the later period of Binance and Curve. We cannot touch on Binance’s specific transactions, so we can neither confirm nor deny this assumption.
We have no way of knowing the owner behind Wallet A. Although its transaction volume is huge, its transaction behavior is neither active nor complicated. We provide some additional clues to complete its portrait:
From March 15th to April 11th, Wallet A received a total of 200 million UST and subsequently deposited Anchor. Other than that, there are very few active transactions on this account. This address is just a basic account, not a multi-signature account. Once I wanted to withdraw 10 million UST from Anchor, but I made a typo to withdraw 100 million UST. A higher fee was paid for using the Wormhole cross-chain bridge on May 7th. Consciously use Flashbots to go to the Curve pool to complete transactions on May 7th. Interestingly, Wallet A seems to have previously prepared to go to the Curve pool for transactions, including from Terra across UST to Ethereum via Wormhole, and withdrawing 0.5 ETH from Binance to an Ethereum address for fees. On May 8, Wallet A divided all USDC obtained from the exchange into multiple deposits to Coinbase.
On the night of May 7th and the night of May 9th, Anchor experienced a large outflow of funds, which brought great pressure on the peg. The funds of large households have been flowing out one after another, but some small households have increased their funds in Anchor during this period. A large number of Anchor borrowers were forced into liquidation during this event, although we do not think these put much pressure on UST. At that time, 75% of the UST gathered by Achor, and within a week after the incident, a total of 10 billion UST flowed out, of which the outflow in the first three days had already reached 5 billion.
The following picture is the outflow chart of UST in Anchor:
The chart below is the outflow chart of UST in Anchor vs the peg deviation:
As can be seen from the above figure, Anchor’s UST outflow is divided into three stages:
During the period of May 7-8, although there were 2.5 billion outflows, it still remained anchored. During May 8-9, there was no significant outflow, and the anchoring was relatively stable. After May 9, the outflow began again, and the decoupling further increased.
In addition, we classified according to the user’s UST holdings, and got the following results:
Big players (anchor store worth more than 1M before 6th): quickly fled the protocol, reducing positions by 15% almost immediately and by more than 40% in the first three days of these events. Nakato (anchor store value of 1~1 million before the 6th): The speed of escaping the protocol was less happy, reducing positions by 5% almost immediately and by more than 30% in the first three days of these events. position. Small households (Anchor stored value of less than 10,000 before No. 6): increased risk exposure in Anchor, but since its total position size is an order of magnitude smaller than that of medium and large households, the increase in exposure is not enough to offset capital outflows.
The following picture is the depositor’s position:
We finally checked the forced liquidation of borrowers at Anchor who deposited bLUNA, bETH and other collateral. Some commentators have pointed out that the speed of Anchor’s liquidation may have further contributed to these events, given the falling prices in the cryptocurrency market, especially LUNA.
However, we found that forced liquidation anchor borrowers did not appear to play a key role in the event. There are three reasons:
Conceptually, the impact of liquidation on pegs is not obvious. Liquidation puts downward pressure on collateral prices (like LUNA) but upward pressure on UST, which shrinks the net supply of UST as outstanding loans are repaid The borrowings are small. At its peak, Anchor had about $14 billion in deposits and only $3 billion in loans. In terms of timing, the liquidation basically occurred as Anchor outflow accelerated. So at best it just adds incremental pressure.
We illustrate the latter two points by plotting liquidations versus Anchor outflows, where forced liquidations do not appear to play a prominent or dominant role in these events, either quantitatively or visually.
On the evening of May 9, the UST decoupled by a few percentage points, and the risk aversion in the crypto market may have contributed to this event.
This sentiment extends beyond the cryptocurrency market and is even felt in the traditional stock market.
The following graph is Bitcoin price vs peg deviation:
The chart below is S&P 500 price action vs peg deviation:
As of May 9, the UST has been severely decoupled. In the next few days, the price of LUNA, the main channel for absorbing UST, also plummeted.
In particular, before May 7th, the total supply of LUNA was several hundred million, but from May 7th to May 9th, the minting and burning mechanism on the Terra chain minted several million more. But after that, the pace picked up. On May 10, the agency minted 40 million; on May 11, another 1 billion was minted; on May 12, another 200 billion was minted; and in the first hours of May 13, another 6 was printed. trillions.
US dollars tampered with UST!Revenue-generating app Stablegains was bombed to invest all customer funds in Anchor, killing users after huge losses
South Korean prosecutors summoned all Terra employees to conduct a comprehensive investigation: Do Kown knew that the LUNA model was flawed and still implemented it?
Some users complained that “the number of airdropped LUNA is less than expected!” Terra: The team is trying to solve it
Make the Telegram news channel of the dynamic zone powerful again! ! Join now to get first-hand blockchain, cryptocurrency news coverage.
LINE and Messenger serve you from time to time