Partnered With:

Ouyi OKEx Research Institute: Discuss several controversial issues of Uniswap V3

Read this article in 18 Minutes

The high-profile star project in the DeFi field --- Uniswap has also recently released a new version of the plan.

Original title: "Ouyi OKEx Research Institute: Discussing Several Controversial Issues of Uniswap V3"

Original source: Ouyi OKEx Research Institute

Just like the Apple conference called "Technology Feast", The high-profile star project in the DeFi field --- Uniswap has also recently released a new version of the plan. Different from the classic V1 and V2 versions, Uniswap's latest V3 version introduces several new features: liquidity aggregation, multi-level fee rates, and oracle machine upgrades, etc.

From Judging from the actual situation after the release of Uniswap V3, the outside world has mixed opinions and opinions. Some praised it, while others expressed deep disappointment. Here, we have deliberately selected several issues worthy of market attention and discussion for discussion. It should be emphasized that in many cases, due to different definitions and standards, even for the same problem, the final conclusions may be different. Therefore, we might as well have a more comprehensive understanding of Uniswap V3 from different angles with the mentality of seeking common ground while reserving differences.

1. Is Uniswap V3 an order book model?

After the release of Uniswap V3, many opinion leaders in the DeFi community regarded it as an order thin gradient model, because Uniswap V3 introduced the "Range Orders" function, allowing liquidity providers (LP) Deposit digital assets within a custom price range above or below the current price.

However, if only Based on this, Uniswap V3 is regarded as the degradation of the order book model, which is too underestimating the innovation of Uniswap V3. In fact, if you study the design concept of Unsiwap V3 carefully, you can find that Uniswap V3 is not an order book model at all --- Uniswap V3 introduces a "price scale" for AMM (tick), LPs can place assets in different price ranges according to market conditions, and this does not change the operation mode of automatic market makers (AMMs) at all!

In traditional order book transactions, whether it is a bidding system or a market maker system, both parties need to place orders (including price and target quantity) on the order book, and use the price as a signal to trade. Among them, in the bidding system, matching is required to buy and sell assets at the price that is most beneficial to both parties to the transaction; similarly, in the market maker system, market makers are required to quote the buying and selling prices in advance, and investors can only place orders after seeing the quotations. Transaction entrustment. From the above, we can see that in the order book, liquidity providers all actively provide quotations, and the market trades according to the principle of "price priority", thus realizing the price discovery function.

Figure 1. Schematic diagram of the order book under the bidding system

< p align="left" line="X5NZ">

But in Uniswap V3, LPs did not take the initiative to provide quotations at all, they just sex placed in different price ranges! The transaction price on Uniswap V3 is the same as that of Uniswap V2. It is still driven by liquidity (adding asset x or asset y to the contract) rather than order instructions. The transaction price is determined by the ratio of the two assets in the asset pool, so There is still no price discovery feature either.

More vividly, if the liquidity design in Uniswap V2 is compared to a water tank, the liquidity provided at all price points is the same; then Uniswap V3 in terms of design principles, It is more like installing a card slot in a water tank. LP can place liquidity in the "card slot" of different price ranges according to the market price, as shown in the figure below.

Therefore, Unsiwap V3 did not degenerate into an order book model, it just introduced a "price scale" and became a Just a wolf in sheep's clothing. For the Range Orders function, the words in the Uniswap V3 white paper are more accurate: "Range Orders" (Range Orders) is more similar to a A price limit pending order strategy --- when the price crosses a given limit price range, the position held by the liquidity provider will completely change from one asset to another.

2. Has the capital efficiency of Uniswap V3 really improved?

capital utilization= Income/ Assets

On Uniswap, assets are the liquidity placed by LPs on Uniswap, and the income comes from transaction fees. Among them, the transaction fee is related to the number of transactions, the average transaction amount and the fee rate. Without considering asset price changes and LP entry/exit, within a certain period of time, the capital utilization rate can be expressed by the following formula:

(1)

Among them, k is the number of transactions within the period, m is the average amount of each transaction, and is the handling fee rate, x is the quantity of asset X in the asset pool, y is the quantity of asset Y in the asset pool, and P is the price of X relative to Y.

From the formula (1), it can be seen that if you want to increase the capital utilization rate on AMM, you must either increase the transaction amount or handling fee, or reduce the asset value in the asset pool.

So how does Uniswap V3 improve capital utilization? A case is given on the Uniswap official website:

Both Alice and Bob want to provide liquidity in Uniswap v3's ETH/DAI pool. Each of them has $1 million. In the case of reaching the same liquidity depth, Uniswap V2 needs to deposit 500,000 DAI and 333.33 ETH (a total value of 1 million US dollars); while in Uniswap V3's price range of $1,500-$1,750, only 91,751 DAI needs to be deposited and 61.17 ETH (total value ~$183,500), thereby improving capital efficiency.

You can see from above It is pointed out that the key to Uniswap V3's increase in capital utilization is to reduce the total value of the capital pool

However, when we read the above cases, we can always vaguely feel that something is wrong. Does Uniswap V3 really improve the capital utilization of AMM?

One of the most obvious Case: Assume that Uniswap V2 and Uniswap V3 have a transaction fee of 1,000 DAI for a period of time. Similarly, all LPs on the platform provide liquidity worth 100,000 for Uniswap V2 and Uniswap V3, then according to According to formula (1), the capital utilization ratio of Uniswap V2 and Uniswap V3 is the same.

It can be seen from the above that judging the capital efficiency of Uniswap V3 involves issues of position and perspective.

From an individual point of view, if a certain individual concentrates his liquidity within a certain price range, his personal capital utilization rate can be improved; however, the market as a whole capital efficiency has not improved as a result.

In addition, Uniswap V3 only considers the individual in terms of capital efficiency optimization design, while ignoring the influence of other people in the market on the individual, which is called in economics externalities.

< p align="left" line="BEw2" linespacing="200">What is an externality? Still take the case on the Uniswap V3 official website as an example: the designers believe that under the V3 version, if a person has $1 million in liquidity, he only needs to deposit 91,751 DAI and 61.17 ETH (a total value of about $183,500). But the actual situation is that in the case of a total liquidity of 1 million US dollars in the market, if people believe that the future price will only change in the price range of $1,500-$1,750, then starting from rational decision-making, each LP will put Liquidity is placed in the price range of $1,500-$1,750, so Unsiwap V3 will still be deposited with a total value of $1 million in liquidity. At this time, there is no essential difference between Uniswap V2 and Uniswap V3, and the capital efficiency is still the same.

More importantly, Uniswap V3 will raise fairness issues. In Uniswap V2, all LPs have equal status, and the handling fees obtained are equally distributed. However, Uniswap V3 aggregates liquidity through the setting of "Range Order" to improve capital efficiency. At the same time, Uniswap V3 stipulates that when the market price is within the specified price range, LP can obtain transaction fees; and when the market price moves outside the price range, this part of liquidity will no longer earn transaction fees.

This is invisible Introduced a liquidity competition mechanism for Uniswap V3---organized and professional LPs will adjust the price range in which they store liquidity in real time according to changes in market prices in order to obtain greater benefits; while ordinary LPs It is difficult to adjust in time, so its capital efficiency and fee share are relatively lower.

Therefore, from the overall From the looks of it, the "liquidity aggregation" that Uniswap V3 attaches the most importance to has not fundamentally improved the overall capital utilization of the platform. On the contrary, the resulting liquidity competition will also cause fairness issues and gas loss caused by LP adjustment positions, which is not worth the candle.