The collapse of Terra’s ecology has been destined to be recorded in the history of blockchain. The decentralized under-collateralized algorithm stablecoin behind it has also ushered in a historical turning point.
This article analyzes the problems of several historical algorithmic stablecoin projects and draws some lessons they can teach us. At the same time, I also want to make a phased summary at this turning point, so as to bring some references for better algorithmic stablecoin solutions that may appear in the future.
Unsecured dark age, the game between air and human nature
Unsecured algorithmic stablecoins have no backing, everything is based on the consensus of participants on their value. However, in the infancy of algorithmic stablecoins, the entire blockchain ecosystem did not have enough usage scenarios, making it difficult to endow such tokens with intrinsic value. This is a speculator’s carnival, and it is also the most intuitive display of the “Greater Fool Theory” (Greater Fool Theory).
Ampleforth: Amplified volatility and rebase token compatibility issues
Ampleforth is often mentioned as one of the ancestors of algorithmic stablecoins. It was also originally designed to stabilize the price of the currency at a $1 value in 2019.
As a rebase token, its token quantity will change towards the target price every 24 hours:
- When the price is within 5% of the target price (current CPI-adjusted target price is $1.1), the number of tokens is unchanged.
- When the price is more than 5% below the target price, the tokens owned by all token holders will continue to decrease until the target price is reached.
- When the price is more than 5% above the target price, the tokens owned by all token holders will continue to increase until the target price is reached.
The entire increase process of AMPL is currently controlled by a proposal proposed and passed in February 2022, with a maximum daily change of 10% increase or 7.78% decrease. Appears when the market price is above $3 and $0, respectively.
The volatility of AMPL mainly comes from the lag of market adjustment and the amplified volatility brought by rebase .
When the market changes significantly in a short period of time (such as a 50% drop in 24 hours), the Ampleforth protocol cannot directly absorb all the fluctuations at the next rebase time point. Instead, there will be an adjustment period Δt. During the cycle, the number of AMPL tokens will continue to decrease, which is used to reduce the supply and increase the price.
But in the process of its decline and rebase, the real holders are actually under a double loss pressure. On one side, a token that was originally worth $1.1 has now become less than $1.1, and on the other side, 1 token that was originally held will become less than 1 token after each rebase.
At this time, if no action is taken, and the rebase token supply is allowed to decrease and the price rises back to the range of $1.1, then the price of the tokens owned by the holders will go back, but the number of tokens will not go back. of. So during this period of volatility, holders actually suffered a net loss in the number of tokens.
When the loss in the number of tokens caused by the decline occurs, the most efficient remedy other than selling tokens is to continue to buy AMPL tokens when the price is less than the target price, in order to reduce the average price and hope that after the subsequent price increase profit. If you insist on implementing this strategy, as long as the token price is lower than the target price, you must continue to buy AMPL tokens until the average price is lower than the current price to reduce losses by profiting. But buying is also risky, because if the token cannot rise back to the target price of $1.1 for a long time, the token held will continue to shrink, resulting in continuous losses. Therefore, in this case, holders generally tend to sell early to stop losses. The selling pressure will also keep the price lower until enough people form a consensus that “buy is profitable” and the token price bottoms out.
On the other hand, when AMPL is higher than the target price, the additional issuance of AMPL tokens will make token holders continue to profit. At this time, the daily income brought by holding AMPL will be very considerable. In the blockchain market, the fear of missing out (fomo) of various funds will make a coin rise higher when it rises or has higher returns. At this time, if the AMPL price remains 50% above the target price, according to the curve, there can be a daily gain of about 5%, which is enough to drive all speculators crazy. In this income carnival, even though many people know that its continuous issuance every day will actually bring about selling pressure, but as long as speculators continue to publicize their high yields, more speculators will take over the market at a high level, and form together with the entire community. The consensus of reinvestment (that is, doing nothing), its currency price will continue to rise. Of course, the pressure of additional issuance will always reach its critical point. Once there is trouble, the consensus will dissipate and bring great selling pressure to the entire token market. And because everyone earns additional tokens above the target price, selling can result in an eventual positive or flat return, even if the current price is below the target price. So basically all of the big gains in AMPL’s history have ended up with trading prices well below the target price.
Incompatibility between Rebase and Defi
In many Defi lending or pledge agreements, changes in the amount of collateral assets make it difficult to calculate the collateral ratio of loans, and the high volatility of AMPL tokens leads to frequent rebases. At the same time, because the rebase mechanism itself is a direct change of the number of token addresses, various defi protocols need to do special logic to provide their AMPL users with real-time updates of the token number. In addition, as a decentralized liquidity provider, it inherently bears the risk of impermanent losses caused by providing liquidity. The higher price volatility and changes in the number of tokens brought about by AMPL’s rebase make providers almost unwilling to to take this risk.
AMPL is the earliest attempt of algorithmic stablecoin in web3. Although its simple and rough adjustment of supply relationship directly points to the essence of algorithmic stablecoin, in terms of “stability”, it is destined not to be stable for a long time because of its rebase mechanism. Target price. AMPL no longer advertises itself as an algorithmic stablecoin because of the poor actual stabilization effect. But at the same time, its stable mechanism also provides an improved basis for various later stablecoins with different algorithms.
ESD: Losing consensus means losing everything
ESD is an algorithmic stablecoin improved from the original Basis white paper. Its mechanism is similar to Basis, so Basis will be skipped in this article. The setup of the bond concept introduced by ESD in order to stabilize the value of the currency seems to have natural problems, but at the same time it allows speculators to continue. However, if all players come for speculation, it will be difficult for the project to have a firm value consensus. Coupled with the problem of its redemption mechanism, the entire ESD unsecured ecosystem collapsed 4 months after its launch.
ESD has an oracle with Uniswap v2 as the source of quotes.
When the price falls below $1, ESD holders can destroy their ESD holdings and receive bonds of equal value. Bonds have a validity period of 90 days (the time of the original white paper).
During the life of the bond, if the price is higher than $1, then the bondholder will follow the redemption submission sequence to redeem the bond one by one. That is, the system will generate new ESDs and meet bond redemption needs. If the bond is fully cashed and the price is still above $1, the new ESD will be split equally between all ESDs pledged in the DAO.
ESD shifted the selling pressure to the water, a practice that doomed its token price to be below $1 most of the time.
When the market price is lower than $1, speculators can buy low-priced ESD and get a priority to cash out when the market price is higher than $1. These bonds enter a lapse countdown when they are bought. Expiration will reset the face value to zero and achieve deflation. All bonds will be “unlocked” through inflation when the market price is above $1. When unlocked, the vast majority of bondholders will try to cash out the bond due to the pressure of the countdown to failure. At this time, the ESD that was originally “locked” by buying bonds will be unlocked by its inflation mechanism, increasing liquidity and creating new potential selling pressure on the entire market.
When the price is above $1:
- On the supply side, ESD begins to inflate, and the higher the price at this time, the faster the supply increases.
- On the demand side, because of the potential effect of ESD inflation, when the price remains above $1, it means that dividends will be distributed to ESD pledgers, which will bring a considerable demand for pledge in the early stage. But as prices get higher and higher, speculators will gradually stop staking new tokens. At this time, the relative relationship between supply has not been destroyed by the changes on both sides, and an equilibrium point of $1 is still maintained.
When the price is less than $1:
- If it is slightly less than 1 dollar, the market will be more confident that it will return to 1 dollar and reach more than 1 dollar. At this time, both holders and new entrants speculators will tend to choose to buy bonds to obtain preferential profits above 1 dollar. s right.
- However, when the price gradually deviates from $1, the market’s confidence in its return will gradually decrease, and the risk of buying bonds will increase. At the same time, the potential benefits of bonds will also increase, so there will be more high-risk appetite funds. Admission.
The game between venture capital and holders is not without a lower limit. When prices are low enough, markets that do not see a possible price recovery experience panic selling. At this point, the consensus on a $1 standard ESD price began to fade. Venture capital will also become reluctant to pay for a bond that has no consensus and cannot be cashed. So supply and demand at this point will go in reverse until a new equilibrium point is reached, well below $1.
At this point the market has completely lost the consensus on the ESD standard price of $1, and the mechanism used to pull the price back to $1 below $1 has completely failed because no one will buy a bond that cannot be redeemed. While there will be no inflation at this point, there will also be no deflation at the same time. The market will go silent and then die.
The mechanism of ESD does not seem to be a problem at first glance. The disguised destruction mechanism attached to the bond can effectively increase the price when it is below $1. But once its consensus is broken, unlike AMPL, which can deflate indefinitely, ESD will have no remedy. Because at this time, no one is willing to buy bonds that cannot be redeemed but have a maturity period. So for ESD, losing $1 of consensus means losing everything.
The mortgage era, playing with humanity and greed
After experiencing numerous failures in the unsecured era, everyone finally realized the importance of a full mortgage. But at the same time, the stability mechanism of algorithmic stablecoins also has its merits, so starting from Frax, algorithmic stablecoins have gradually entered the era of collateralization. Frax, as a project with close to 0% mortgage in advance and 100% mortgage in refund, can be said to be one of the most stable algorithmic stablecoins at present. At the same time, the stablecoin system of Terra ecology using mainnet tokens as collateral is also developing rapidly.
UST (Terra Money): Mainnet (L1) and Mainnet Token-Powered High Risk High Yield Stablecoin
The emergence of UST answers a question of L1 (main network): how to better capture the expected value of the main network token in the future? Terra chose to build a stablecoin system.
UST uses the Terra mainnet token Luna as collateral, and there is no cash reserve in the initial stage of the ecosystem.
When the price is above $1, arbitrageurs can burn the Luna minted UST bought with $1 and sell at the current price above $1 and make a profit.
When the price is below $1, arbitrageurs can buy UST at a price below $1 and exchange it for Luna at the market price of $1 and sell Luna for profit. The maximum arbitrage amount for this method is $300 million per day, and the amount is reset every 36 blocks (later, it was raised to $1.2 billion per day through a proposal in the UST crash).
The extreme interest-earning debt structure that UST failed to adjust in time gave it a fatal blow
In 2020, Do Kwon proposed a solution to the problem of the low asset pledge rate on the Terra chain. In mid-2020, one of the most efficient protocols in blockchain history, Anchor, was launched. After that, attracted by the interest rate of 20%, Anchor’s lock-up volume (TVL) continued to rise. By January 2022, it will reach nearly $5 billion . And it climbed to a terrifying more than 14 billion TVL in just four months. On the eve of Luna’s collapse (May 8), the total market value of UST was only more than 18 billion, and the amount of UST locked in Anchor was more than 14 billion.
At the same time, in the face of high returns, the seemingly harmless and stable UST makes many people rush for high leverage. Therefore, although the high interest-earning lock-up volume is very risky, the centralized liquidation of leverage is also one of the key elements of the collapse.
At this point, it’s no wonder that a turbulent $85 million sell-off can cause a series of follow-up events.
UST is a game against Luna, the mainnet token of Terra ecology. The value of Luna comes from its gas fee and POS mining mechanism as a mainnet token for the entire ecosystem. UST will transfer the value generated by Luna and exist in the Terra ecosystem in the form of stable coins, and provide Terra with a better and more stable transfer platform for usage scenarios (compared to other mainnet tokens such as ETH, which are more volatile) . In this process, the value purchased by the holders of Luna and UST is the same, that is, the ecological value of the Terra main network.
So at this time, if someone says that the most important value of Terra ecology is Anchor’s UST interest-generating pledge, does this sound very much like a Ponzi scheme where pure money generates money?
In addition, the ecological problems of Luna and UST are not limited to Anchor. During the process of generating UST, the total market value of Luna + UST should be constant. Because UST will exist in the entire ecosystem on behalf of the market value burned by its burning Luna. The act of minting UST should not be seen as a requirement for Luna tokens, but as a formal conversion. People who want to buy Luna can also get Luna by buying UST and then changing to Luna. But in fact, the market regards the casting of UST as a demand and consumption of Luna, which makes the market extremely overestimate the value of Luna. What overvaluation brings is that more UST is minted by less Luna, creating a vicious circle that makes the false prosperity even worse.
In fact, Terra, as a main network ecology, has a very delicate UST construction. But for the Terra ecosystem, on the one hand, Anchor’s high income quickly built a tens of billions of financial empires for it, and on the other hand, Anchor’s high income immersed the public in a high-value false prosperity and eventually led to collapse. The market also lacked sufficient understanding of the valuation of this model of swapping the value of mainnet tokens and stablecoins, which ultimately exacerbated the entire collapse.
In the whole process, in the final analysis, Anchor played with the public’s greed, and UST played with the public’s perception of value. Both sides packaged the wealth dream of a financial empire, which made countless people addicted to it and eventually collapsed together.
How to view algorithmic stablecoins in the post-Terra era
Do we really need algorithmic stablecoins?
Terra’s thunderstorm has made many people very disgusted with non-over or fully collateralized stablecoins, thinking that it is a Ponzi scheme that produces air. However, some people think that Terra’s thunderstorm has little to do with algorithmic stablecoins, but is caused by the imperfection of the entire Terra Money ecosystem.
Vitalik on May 25, 2022 is obviously more in favor of the latter, and praises the RAI stabilization mechanism in the article. On the same day, he blogged about two criteria he used to measure stablecoins.
- Can stablecoins safely clear all users?
- What happens when the stablecoin mechanism is anchored to an index that rises 20% annually?
For specific analysis , please refer to the original text .
The essence of algorithmic stablecoins is a decentralized movement of coinage rights, which can only be done by the state on the Internet through the consensus of users and a series of algorithms. Of course, the current road of algorithmic stablecoins is not satisfactory, especially the failure of Terra has caused many people to lose confidence, resulting in “Since there are already centralized and compliant stablecoins such as USDC and over-collateralized stablecoin DAI”, why? Still need algorithmic stablecoins?
For these two types of stablecoins, the answer given by algorithmic stablecoins is actually very simple, because centralization will face the risk of regulatory ban, and over-collateralization will face the problem of capital efficiency. Since a completely decentralized blockchain network has emerged, why is everyone still required to obey centralized control? Since there is a better capital utilization rate to earn higher returns with lower margin risk, why should you mortgage more margin assets? We cannot fall into a storm of criticism just because of the failure of the algorithmic stablecoin project at the current stage. Algorithmic stablecoins, as the practice of future stablecoins, still have their value and necessity.
Does each public chain need its own algorithmic stablecoin?
After the success of Terra’s algorithmic stablecoins, public chains such as Waves, NEAR, and TRON have launched their own algorithmic stablecoins. Now that Luna has stepped down from the altar, one question is whether it is necessary for other public chains to launch their own calculation?
Algorithmic stablecoins that coexist with the native tokens of the public chain can provide a better and more stable economic model for the public chain itself. But at the same time, it needs to be cautious about the macroeconomic aspect to avoid repeating the mistakes of Terra. So far, there is no stable solution that has passed the test of time and can improve the ecosystem of a public chain token. Therefore, in the short term, smaller public chains will take a cautious approach to any algorithmic stablecoins that intersect with their native tokens.
It is undeniable that stablecoins add an important asset to the public chain and enrich its own ecology. At present, the cross-chain bridge assets of various other stablecoins are enough to support the stablecoin demand of an early public chain. However, in the later stage, when the nascent public chain develops into a complete ecosystem and forms a system similar to a “country”, then the native algorithm stablecoin will definitely become an indispensable part of forming a complete financial system.
write at the end
The early unsecured algorithmic stablecoins proved that the unsecured road would not work. Over-collateralized or fully-collateralized traditional stablecoins of the same era have gradually become the mainstream of the market. The rise of Terra and UST is not accidental. It is a feasible way to use mainnet tokens as collateral to provide a stable settlement system for the ecosystem, but the financial regulation and balance are far more than that simple. Terra’s story also tells us at all times that we must look at the value of the main network rationally. Don’t abandon the value that computing stability itself can bring to an ecosystem for the so-called TVL, the so-called false prosperity, and the greed for money.
The information in this research report comes from publicly disclosed information, and the opinions in this article are for research purposes only and do not represent any investment opinions. The opinions and forecasts issued in the report are only analysis and judgment on the date of issuance, and do not have permanent validity.
Without the authorization of DODO Research Institute, no one may use (including but not limited to copying, disseminating, displaying, mirroring, uploading, downloading, reprinting, extracting, etc.) or permit others to use the above intellectual property rights. If the work has been authorized to use, it should be used within the scope of authorization, and the source of the author should be indicated. Otherwise, its legal responsibility will be investigated according to law.
“DODO Research Institute” led by the dean “Dr. DODO” led a group of DODOs to dive into the Web 3.0 world, doing reliable and in-depth research. With excellent awareness, Dr.DODO feels the magic of data, aims to decode the encrypted world, outputs clear opinions, and discovers the future value of the encrypted world. “DODO” is a decentralized trading platform driven by an active market maker (PMM) algorithm, aiming to provide efficient on-chain liquidity for Web3 assets, allowing everyone to easily issue and trade.
Posted by:CoinYuppie，Reprinted with attribution to:https://coinyuppie.com/failed-algorithmic-stablecoin-projects-whats-wrong-with-ampl-esd-and-terra/
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.