Author: CYC Labs

As the big brother of true algorithmic stablecoins, AMPL has maintained a good vitality since its inception, and its mechanism has been analyzed by many. Today, we will not elaborate on this, but rather think from another perspective, that is, "What is a true algorithmic stablecoin?" to see how AMPL measures up. Can it be matched with the title "Progenitor Giant"?

What is a True Algorithmic Stablecoin

Without further ado, let's get straight to the point. At least from my personal perspective, a true algorithmic stablecoin needs to meet the following four characteristics:

High Stability: Undoubtedly, the main reason stablecoins can be used as a bridge for asset trading lies in their stable value, or in other words, the stability of purchasing power.

Low Volatility: This is actually similar to stability. If we observe the "price" (or should we call it "exchange rate") of existing more traditional stablecoins, we will find that there is a positive correlation between their "exchange rate" and their purchasing power. Therefore, the lower the volatility, the more stable the purchasing power.

High Liquidity: The value endorsement of stablecoins is essentially based on consensus. Consensus is related to people. Only when people use and recognize its value as a bridge for asset trading can it be called a "stablecoin" similar to traditional currencies. This is especially true for algorithmic stablecoins; high liquidity and broad utility enhance their stability. If liquidity is limited to a small circle, any transaction of a certain scale will affect its stability, and thus there is no such thing as "stability."

Low Governance: This can be said to be determined by the nature of algorithmic stability itself. The underlying algorithmic stability should rely on mathematics. It is paradigm-based. In principle, as long as the mechanism design is excellent enough, there will be little need for human intervention. In other words, ideally, algorithmic stablecoins should be able to operate independently of governance, relying entirely on algorithms to maintain stability.

So, let's examine AMPL's performance from these four aspects.

Stability & Volatility

Here, we should look at stability and volatility together. Because whether or not a coin is stable is the result of its performance, while the size of volatility is its process. Therefore, looking at them together will provide a more intuitive judgment.

First, let's take a look at the price performance of AMPL and various other stablecoins over the past year (for stablecoins that have not been online for a year, we use their price at launch as the price one year ago) to compare whether AMPL is "stable" in terms of results.

Data Source: intotheblock, compiled by CryptoYC

From the table above, we can see that in terms of stability results, AMPL's performance is actually quite good. Although there is a significant price difference between the time of its launch and the latest time point, if we look at the average price over a longer period, AMPL's price is still similar to that of other stablecoins, hovering around 1. However, we must admit that traditional USD-pegged stablecoins have the strongest stability. The main reason is that the mechanisms of the two are fundamentally different; AMPL is purely algorithmically supported, with no collateral and no institution responsible for adjusting liquidity, purely adjusting its supply based on market demand to maintain purchasing power stability, known as the "elastic supply mechanism." This is different from USDC, USDT, BUSD, etc., which rely on centralized institutions to maintain price stability. It is also different from Dai and Fei, which require collateral to adjust liquidity and stabilize the coin price.

Therefore, if we do not look at the price but rather at the purchasing power of each stablecoin itself, since its mechanism is designed around purchasing power stability (benchmarking against the US CPI) and is currently effective (not zeroed), we can say that compared to other stablecoins, AMPL's purchasing power is relatively stable.

Now, since the results are acceptable, what about the process? In other words, how is the volatility during this period? So, let's take a look at the volatility of AMPL's price. For comparison, we will use Fei, which is also touted as a star algorithmic stablecoin, to examine the volatility performance. Of course, we need to note that the mechanisms of the two are still different.

Data Source: intotheblock, compiled by CryptoYC

From the above chart, we can see that AMPL's price volatility is still quite large compared to FEI. The reason lies in the fact that FEI is a collateralized stablecoin. Therefore, it can use collateral assets to recycle/release liquidity to maintain price stability. AMPL, on the other hand, has no so-called collateral to regulate liquidity, so its volatility is similar to that of traditional blockchain tokens like BTC, where the entire issuance and recycling rely on its rebase mechanism, which, at least for now, is still quite effective in most cases. Personally, I have never considered FEI to be an algorithmic stablecoin; it is essentially just an improved version of MakerDAO.

Liquidity

The importance of liquidity for stablecoins has already been mentioned. Now, the question arises: How do we judge the liquidity of a stablecoin? I personally believe that several reasonable criteria for judgment are:

Where the token can be used or traded: A token must have utility to circulate better, and circulation cannot be separated from trading, so I believe this condition is very important.

The number of new addresses holding the token and the number of active addresses: The token itself serves people. People use tokens for various financial activities. In blockchain, people can simply be represented by wallet addresses, so we look at this condition here.

The issuance volume of the token: The issuance of stablecoins should match market demand. Demand accompanies financial activities, and financial activities are accompanied by token circulation. Therefore, to some extent, good liquidity will lead to a higher issuance volume (some may argue that due to AMPL's mechanism, its token quantity may increase or decrease with financial activities, but we need to note that the goal of rebase is to maintain a stable ratio of users' wallet balances, and financial activities include not only direct activities like transfers but also activities like "collateral" that will "consume" tokens).

So, let's examine AMPL's performance from these three aspects.

First, let's start with the simplest condition, which is to see which exchanges AMPL is listed on. Of course, data is meaningless without comparison, so we will again use FEI for comparison. The results are as follows:

In terms of algorithmic stability, compared to FEI, AMPL's performance is quite good, being listed on many exchanges, including both low-threshold DEXs and some relatively well-known CEXs. This provides a certain guarantee for AMPL's liquidity. However, simply looking at the number of listed exchanges is not enough. We should also look at the trading volumes of the two. Therefore, let's compare the daily average USD trading volume of both.

It seems a bit dismal. Compared to FEI, it is about ten times lower. However, if I were to speculate on the reason, I would personally attribute it to FEI's initial promotion. After all, FEI was a top-tier project at that time, with endorsements from many big names, leading to a lot of people getting involved and then getting stuck. Meanwhile, AMPL has remained in a lukewarm phase. Therefore, in terms of trading volume, AMPL still lags significantly behind other mainstream stablecoins. As for the peaks in July and August, they stem from the later famous "Geyser Plan," which we will discuss in detail later.

After discussing exchanges, let's look at where AMPL can be used. According to the partners currently announced by AMPL, it can be used in very few places; currently, the only one we can find is AAVE (which launched a pool supporting AMPL lending on July 23). The rest are all exchanges.

Having looked at the somewhat disappointing trading volume and where AMPL can be used, we can move on to the next topic: how is its user growth and activity? Here I must state that due to personal reasons, I have not found data on its competitor FEI, so I can only compare it with other stablecoins.

From the above chart, we can't see much information; the most useful information might be that, except for USDC, the daily average growth and daily average active address ratio of various stablecoins are similar. Since that is the case, let's take a closer look at AMPL's user situation.

As of July 26, the total number of AMPL addresses reached 96k, and the daily active and daily new address numbers are shown in the above chart. We can clearly see that during the period from July to September 2020, both daily active and daily new addresses reached a very high level. Why is that? The reason, as I mentioned earlier, is the famous "Geyser Plan." On June 23, 2020, Ampleforth proposed its Geyser plan on Uniswap V2, which sparked a lot of discussions. Essentially, it provided liquidity rewards (AMPL-WETH) on Uniswap V2. The plan aimed to provide 25,000 AMPL as rewards each month. Of course, if it were a regular reward scheme, there wouldn't be such dramatic changes, so the project team designed a very clever mechanism for distributing rewards.

To limit user exits, Ampleforth implemented a Bonus period where the longer users hold AMPL, the greater the rewards they can earn. "Initially, the reward multiplier for users is 1, and as the staking period increases, the reward coefficient increases, reaching a maximum of 3 times after two months." The reward multiplier here can be indirectly seen as a distribution weight. The main purpose of the rewards is to attract users, thereby increasing AMPL's liquidity. Thus, this reward mechanism achieves its goals in the following two ways:

1. The existence of this incentive encourages users to stake a large amount of AMPL in Uniswap. Especially in the early stages, the APY was quite high (up to 134.52%), leading to a sharp increase in AMPL's price in the short term, attracting a large number of LPs to provide liquidity and increasing the number of new users.

2. At the same time, due to its rebase mechanism, the number of AMPL in users' wallets is automatically adjusted. In such a scenario of drastic price changes, a large number of arbitrage opportunities arise, naturally stimulating some user growth and having a feedback effect on the price.

At this point, we have briefly analyzed the main reasons for the sudden increase in these three data points. So why did the data start to decline in August? Actually, once we understand the liquidity reward distribution mechanism mentioned above, we can also understand:

Because the highest multiplier for liquidity mining rewards is set at 3 months, and as the number of LPs increases, the yield will decrease. Coupled with the fact that a large influx of new users and active users are mostly short-term speculative arbitrage participants, once the reward multiplier reaches its peak and the yield decreases, a large number of LPs will withdraw, leading to a rapid decline in various data points. This is a normal phenomenon. However, we need to note that during this process, a large amount of AMPL should have flowed into the market, increasing the utility of AMPL. Because the issuance mechanism of AMPL is inherently linked to market demand, if our assumption holds, a large amount of AMPL should have been issued and flowed into the market during this period starting from June 2020. So let's take a look at the supply change chart of AMPL.

Sure enough, the amount of AMPL distributed began to rise sharply in July. Although it fell back to half after the Geyser, it is still much higher than before. One cannot help but say that the project team is very clever; both the stability mechanism and the stimulation of liquidity to expand the utility of their stablecoin fully utilize the market. Game theory at its finest. The subsequent fluctuations in supply can basically align with price changes, to some extent representing the effectiveness of its stability mechanism.

However, while the Geyser plan seems promising, there are two issues that cannot be ignored:

We know that neither lending nor exchanges, especially DeFi lending and exchanges, open "liquidity pools." The AMM mechanism of DEX and the over-collateralization mechanism of lending seem to be inherently incompatible with AMPL's rebase mechanism. So how does AMPL solve this problem?

Where do the rewards for the Geyser come from? Are they newly issued or existing?

With this question in mind, I consulted relevant materials and asked questions in the community. First, regarding the rebase issue, according to the official Medium post about the Geyser plan, it states that staking AMPL in the liquidity pool to provide liquidity is considered "using" AMPL, and this portion of AMPL does not belong to the "balance" part, meaning it will not participate in rebase. The same applies to lending. However, lending has a slight difference in that if users stake and lend out AMPL, it will still participate in rebase, but when repaying, they must return the amount borrowed initially. A bit of a catch. However, according to the official statement, purchasing power remains consistent.

Next, regarding the second point: where do the rewards for the Geyser come from? This was answered by personnel from their official Discord: "Ampl ecosystem fund iirc" and "Initial funds, still not drained fully." So, let's first look at how AMPL was initially distributed when it launched. A single chart can illustrate this:

We can see that 23.2% of the initial AMPL was allocated to the AMPL Ecosystem Fund and 20% to the treasury. This means that the funds used for Geyser rewards are not newly issued but are drawn from their own Fund. However, it is important to note that if this continues, the fund will eventually run dry. The total rewards for the completed Geyser V1 amounted to 11,248,520.31 AMPL, while the ongoing V2 is set to distribute 754,853.55 AMPL, totaling 12,003,373.86 AMPL after both versions conclude. This has already exceeded the Fund's inventory, which needs to be monitored.

Overall, AMPL's current performance across various indicators is relatively average, and the recent Geyser V2 version does not attract users as much as the initial V1 launch. Therefore, how AMPL will enhance its liquidity in the future remains to be observed.

Now, let's move on to governance. Since I have already stated that a true algorithmic stablecoin should not have a DAO-like governance organization, let's take a brief look.

Governance Level

As we all know, AMPLFORTH suddenly launched its governance token FORTH on April 22 this year and directly airdropped it to many old users' wallets. Since we mentioned earlier that algorithmic stability should pursue low governance or even no governance, we will not analyze the token. Instead, let's look at AMPL's DAO governance situation. First, let's see what the official governance steps are:

To assess the governance level of a project, we first look at how many proposals have been initiated on the Forum.

Quite a few; since June, there have already been 10 proposals. It seems that everyone is very interested in participating in AMPL's governance. However, let's see which of these proposals have entered the second step.

Here we see that although it appears that 6 proposals have entered the second step, which seems like a lot, we need to note that these proposals were all made before the governance token was launched. Because the API was originally the place for AMPL governance before FORTH was launched. In other words, as of now, there has not been a single proposal that has entered the actual discussion stage after FORTH was launched. Therefore, there is no need to look at the subsequent steps.

This also indicates that, at present, AMPL's governance level is not very high, but it is worth noting that there is a strong desire among users to participate in governance. So this will require continued observation.

Conclusion

After reviewing the above data, we can summarize AMPL itself: Whether it is an algorithmic stablecoin or other stablecoins, the most important goal is to allow it to circulate in a broader field. In other words, during the initial phase of rapid expansion, the surge in demand should lead to a surge in issuance. However, we can see that for AMPL, there are very few places where it is genuinely useful; currently, only the recent AAVE can be considered significant. So how does AMPL solve this problem? How to allow more AMPL to enter the market without collateral?

The answer is very clear. By being listed on various exchanges. As mentioned earlier, the AMPL rebase adjusts the "balance" in wallets and does not adjust the "application" part. What does this mean? I think we should clarify this statement: The rebase essentially adjusts the circulation of AMPL within the entire economic system. The AMPL listed on exchanges belongs to the application part and does not count as "circulating" AMPL.

By listing on a large number of exchanges, a significant amount of AMPL is locked up. This causes the circulating AMPL in the market to decrease rapidly, leading to an increase in AMPL's price. According to the rebase mechanism, the contract will issue a large amount of AMPL to offset this increase. Of course, some may argue: "The AMPL on exchanges can be withdrawn, and with the Geyser rewards, doesn't that increase the total circulation? Why do you say the circulation is decreasing?"

Everyone should think about it a bit; the AMPL listed on exchanges will not all be withdrawn in a short time and enter users' wallets, but rather it is a relatively slow process. Especially for DEXs, where liquidity pools do not have the concept of "drying up," it is even more difficult to see a large amount of AMPL in the pool flowing into the external market in a short time.

Meanwhile, Rebase is typically completed within 10 days. The time in between is not negligible. The Geyser rewards also generally last for more than a month, which is much longer than the rebase time. Therefore, in the short term, the total circulation decreases, leading to more AMPL being issued into the market through rebase. Ultimately, this results in an increase in the circulating AMPL in the market.

At the same time, the emergence of AMPL raises a new question about stablecoins: Does a stablecoin refer to price stability or purchasing power stability? In other words, what exactly is stable about value stability? Clearly, AMPL's answer is purchasing power stability. This is completely different from the understanding of other existing stablecoins. Other stablecoins still pursue price stability, much like the "central banks" of traditional industries. The ability of AMPL to propose this idea is inherently very innovative.

However, as I mentioned, its liquidity currently remains quite average, meaning not many people are using it. Therefore, whether it will truly become a "stablecoin" in the future remains to be seen. Of course, whether there will be a more effective algorithmic stability mechanism in the future is also uncertain. However, even if a new algorithmic stability mechanism emerges, AMPL's status will remain akin to the "Progenitor Giant" in "Attack on Titan," marking the beginning of true algorithmic stablecoins and destined to leave an indelible impact on blockchain history.