During the process of designing Elena Protocol, we have been monitoring several well-known algorithmic stablecoin projects. When choosing a price anchoring mechanism, we prefer the arbitrage strategy which is used by both FRAX and FEI. We have conducted further research related to the price anchoring mechanism of FRAX and FEI protocols.
The FEI and FRAX will be anchored at $1 value by design. The price anchoring at $1 should be separated into two circumstances to be discussed separately, one is when the price is greater than $1, and the other is when the price is less than 1. When it is greater than 1, the two protocols use the same approach by using arbitrage. Therefore, we focus on analyzing how to anchor the price back to $1 value again through an effective mechanism when the price is less than $1.
FRAX’s Price Recovery Strategy
FRAX’s strategy corresponds to when it is greater than $1, and it is also implemented using an arbitrage mechanism. When the price is less than $1, the arbitrageur will find that there is an arbitrage opportunity so that he can buy at a price less than $1 in the exchange, and then sell at a fixed price of $1 in the vault, thus completing an arbitrage. This method is very friendly to arbitrageurs, because theoretically, the smaller the time interval between two online transactions of an arbitrageur, the less risk of its arbitrage strategy. In theory, it is infinitely close to risk-free arbitrage. This is very tempting for arbitrageurs, making them more willing to do arbitrage. In this way, the driving force behind the price return is produced.
FEI’s Price Recovery Strategy
When FEI is less than $1, there are two important mechanisms to protect it. One is direct motivation, and the other is reweight. The system will give priority to direct incentives to smooth out price fluctuations. When the mechanism fails, reweight will be activated. This is an efficient buyback scheme that returns the price to 1 under the condition of extremely low consumption.
Direct incentives are divided into two parts, mint reward and burn penalty. Buying when the price is less than $1 will be rewarded with profit. Therefore, when the user purchases, there is an unrealized profit, but this cannot be realized. Because the amount of the reward according to the agreement is less than or equal to the burn penalty. That is to say, only when the price returns to $1, the speculator can realize profit. This is risky for speculators. When the market lacks confidence in the mechanism or does not have a strong consensus on the mechanism, prices may move in the opposite direction and form a death spiral. At the same time, this part of the reward has created new selling pressure. From a certain point of view, the mint reward mechanism is similar to the bond mechanism of basis protocol v1. The mechanism has proven to be unsuccessful.
Before we published this article, the burn penalty mechanism had been canceled. We also hope that FEI can work with the FEI community to give a more effective optimized plan. At present, we only analyze the original plan. The previous mechanism is that when the price deviates from 3%, selling FEI will suffer an additional 9% penalty. This is very scary for users in the genesis period. It shows that either abandoning the benefits of tribe rewards in the genenise period may even cause losses, or be imprisoned to hold FEI for a long time. Users need to balance between money loss and extremely low capital utilization rate, which could be very painful.
Besides, FEI’s long-term positioning is a stablecoin, and one of the most important attributes of stablecoin is sufficient liquidity. The burn penalty is contrary to this attribute.
One scenario of PCV’s current application is to activate the buyback function when the price is less than $1 and when the direct incentive fails. Through its mechanism, the price can return to $1 at a very low cost. From a theoretical analysis, this program consumes less capital than arbitrage, and the arbitrage program will distribute this part of the profit to market participants, thereby increasing market activity and transaction volume.
FRAX currently mainly uses USDC stablecoin as value collateral, while FEI uses ETH. From the price volatility, it can be seen that ETH has higher volatility. In this way, when ETH falls, FEI’s treasury may have insufficient collateral or even the possibility of collateral value being zeroed out, which is terrible for its investors.
FRAX vs FEI
Withstand the impact of the violent fluctuations in the price of ETH, there is a greater risk of insufficient collateral.
After multi-directional comparison, we feel that the reliability of FRAX’s solution is better, and FRAX has undergone 4 months of market verification. So ELENA chose the FRAX model as its basis.
The improvements of ELENA
When we stood on the shoulders of giants, the team conducted a more in-depth discussion. We believe that the relationship between the reserves of the vault in the agreement, that is, the ratio of the collateral to the pool in the secondary market, is very important. We tried to define the reserve ratio (white paper) to evaluate and maintain the reliability of the system.
Since the vault is fixed at 1 USD for minting and redeeming, in this case, the smart participant will preferentially choose the vault for “transaction” than the trading pool, because this way he can have close to 0 slippages in exchange between stablecoins and FRAX. Even in the fractional stage, such as the case of a 90% collateral, the vault will be preferred, which can ensure that at least 90% of the funds are exchanged without slippage. This method gives users a very smooth trading experience. On the other side of the coin, this is a test for the reserves of the vault. When there is a large amount of selling, the reserves of the vault will be consumed before the stablecoins in the trading pool. The frightening thing is that when the reserve ratio is very low, arbitrage will not be possible, and the mechanism of returning the price to 1 will fail.
In this regard, we try to alleviate this problem by dynamically setting the state of redeeming. When the collateral ratio drops to the threshold, redemptions will no longer be allowed. At this time, if the funds still need to be withdrawn, ELENA needs to be sold in the trading pool. When the collateral ratio returns to above the threshold again, we will reopen the redemption function. In this way, the stablecoins in the vault and trading pool can be consumed relatively evenly, thereby reducing the risk of anchor failure.
Besides, to reduce the cost of transactions, we set the minting and redeeming fees to 0. This may more effectively enhance the effect of the anchoring mechanism.
We believe that true algorithmic stablecoins should have the feature of anti-volatility in the market, rather than a purely speculative opportunity in a bull market. Just like what Travis Moore (Co-Founder @fraxfinance) commented out in our Telegram group: “The true test of algo stablecoins will be a bear market like 2018–2019”.