After writing about Ethena yesterday, I recalled some of my thoughts on stablecoins in 2021. Now looking back at my previous thoughts, I think they are still valid.

In the crypto ecosystem, the stablecoins we usually talk about are actually just US dollar certificates similar to Hong Kong dollars. This certificate must ensure that users can exchange it for US dollars at the same value.

Therefore, the key to this type of stablecoin lies in whether the collateral assets used by the project party to redeem the certificates (stablecoins) are recognized by users when users take out certificates (stablecoins) to pay for US dollars.

So what kind of mortgage assets are likely to be recognized by users?

It must be a valuable asset recognized by users.

In the crypto world, the most mainstream valuable assets are basically two categories:

One type is the simplest, which is the U.S. dollar pledged in the bank. The most typical stablecoin built with this idea is USDC.

The other type is valuable assets held in trust by security institutions. Specifically in the crypto ecosystem, these are valuable assets held in smart contracts (ETH, etc.). The most typical stablecoin built with this approach is DAI.

For stablecoins such as USDC, since its value is directly supported by the US dollar, as long as the assets can be fully collateralized, it can naturally maintain a 1:1 peg with the US dollar. This eliminates the trouble and challenge of maintaining the exchange rate, but inevitably introduces centralized interference.

For stablecoins such as DAI, since its value is not supported by the US dollar but by crypto assets with greater volatility relative to the US dollar, its biggest challenge is how to maintain a stable exchange rate against the US dollar.

There are two ways to maintain a stable exchange rate against the US dollar:

One is a nearly risk-free approach, which is to ensure that the value of the asset is higher than the value of the certificate (stablecoin) no matter what. Although this approach completely avoids risks, it comes at the expense of capital efficiency. This is why DAI uses over-collateralized assets to maintain the stability of the exchange rate, but it accumulates a large amount of funds and seriously restricts liquidity.

Precisely because of this defect, people have been trying to find the "ideal country" in their hearts - getting rid of liquidity constraints and not sticking to the value of the mortgage assets themselves, that is, the value of the mortgage assets does not need to be higher than the value of the borrowed US dollars.

If the value of the collateralized asset itself is less than or equal to the loaned USD value, then the difference between its value and the USD value is the risk, and this risk is what these stablecoins need to deal with.

When the value of the collateral asset is extremely small, it is 0. In this case, the stablecoin becomes a pure algorithmic stablecoin. When the collateral asset has a certain value, but is not higher than the value of the US dollar, this type of stablecoin is a semi-algorithmic stablecoin.

In the last bull market, several projects have tried pure algorithmic stablecoins, such as AMPL and BAC. They were eventually proven to be difficult to maintain stable exchange rates.

Semi-algorithmic stablecoins were also tried by several projects in the last bull market, including LUNA, ESD, etc. They were eventually proven to be difficult to maintain stable exchange rates, although they may last a little longer than pure algorithmic stablecoins.

Now fewer and fewer people believe in the first type of purely algorithmic stablecoins, but some people are still trying the second type of semi-algorithmic stablecoins.

But in my opinion, both types of algorithmic stablecoins will ultimately fail.

Because as long as the stablecoin is not over-collateralized, its highest price is the US dollar price. However, the price of any asset will fluctuate. Once its price fluctuates, the price of the collateral will be lower than the US dollar. No matter what method is used to make up for this price difference, it is necessary to introduce new valuable "collateral".

Of course, this additional "collateral" can take various forms: either centralized assets or encrypted assets with strong consensus.

It is difficult for algorithms or transactions to make up for this price difference in a sustainable and long-term manner.

So I think that although the ideal of finding a stable solution is great, the laws of finance cannot be violated, just like the ideal of building a perpetual motion machine is great, but the laws of physics cannot be violated.

In the crypto world, breaking away from over-collateralization and full dollar collateral like USDC and seeking other algorithmic stablecoins is probably just a utopia.