Author: Kylo@Foresight Ventures
Tips:
AMM and RFQ actually represent the difference between DeFi and TradFi thinking modes
AMM improves capital utilization efficiency through LP leverage
The RFQ model has a natural advantage for cross-chain transactions
The introduction of Core Pools will significantly change Balancer’s revenue structure
Liquidity incentives regarding price ranges can reshape the shape of Liquidity Position
This article mainly discusses the development history and possible development direction of DEX. The research on Perpetual Trading DEX has been described in detail in "Perpetual DEX: The Road to LP Productization". This article focuses on the spot DEX field, covering directions including aggregators, veToken mode and Uni V3 mode DEX.
1. Spot Trading Aggregator
1) AMM V.S. RFQ
Before entering the main text, we need to review the history of AMM V.S. RFQ. Today, when the various basic frameworks of DeFi are relatively complete, we seem to have become accustomed to AMM DEX. But the original form of DEX is not what we see now. The order book system and RFQ are the earliest ways to trade assets on the chain. In the iterative changes of DEX, RFQ was surpassed by AMM in the competition with AMM. The event of AMM V.S. RFQ actually reflects the difference between the thinking logic of DeFi and the thinking logic of TradFi. From the logic of TradFi, the essence of finance is the process of improving the efficiency of capital utilization. However, compared with the RFQ model, the capital utilization efficiency of the AMM mechanism is extremely low, and it is easy for assets to face price fluctuations when pricing assets, which means that AMM should be inferior to the RFQ model. But the actual result is just the opposite. The AMM model goes further and further on the road of DeFi, and eventually even affects the model of Perp Trading.
This actually shows that it is a deviation to think about the development model of DeFi from the logic of TradeFi. As a financial system native to the blockchain, DeFi must have its own underlying logic for finance, and this underlying logic determines the development of DeFi. We can roughly summarize it as follows:
LP Market Making Democratization
Transparency of pricing power
LP Leverage
Except for the first two of the above three characteristics, which are characteristics given to DeFi by Web3 ideology, LP leverage can actually be combined with the logic of TradFi. In the logic of TradFi, improving capital utilization is the essence of finance. Since the TradFi system has now developed to a relatively mature state, the composability between different TradFi applications is not that strong. Therefore, for TradFi, improving capital turnover means improving capital utilization efficiency. The RFQ model is a typical trading model that improves capital utilization through high capital turnover. But for DeFi, there are two ways to improve capital utilization: improving capital turnover within a single project and improving asset composability between multiple projects. This means that DeFi can not only improve capital utilization efficiency through capital turnover, but also leverage assets through composability and realize capital reuse through LP leverage.
Therefore, if you think about the above logic clearly, you can easily deduce that P2Pool in Perp Trading will become a trading model that goes hand in hand with the order book - GMX and GNS are just extensions of Uniswap in the development logic of DeFi.
Generally speaking, AMM and RFQ achieve efficient use of funds from two perspectives, the former through composability and the latter through high turnover. Since the entire endogenous DeFi architecture has not been built in the early stage of DeFi development, it is more meaningful to improve the utilization rate of funds through the DeFi composable model than to simply improve the turnover rate of funds. This is why AMM is better than RFQ in the early development of DeFi. But the development logic of things is circuitous, and the dominance of AMM in Trading is not absolute. At present, the reuse of funds through composability seems to have reached the end of leverage, and AMM DEX has begun to make a fuss from the perspective of improving capital turnover. At this time, the RFQ model representing high capital utilization has been put on the table again, and the focus of the narrative is still the issue of capital utilization.
2) RFQ model complements AMM
As mentioned above, the advantages of the RFQ model are high capital turnover and lower slippage. The RFQ model can be seen in various current trading aggregators, such as 1inch, Paraswap, etc. The RFQ model was introduced into the aggregator as a supplement to AMM trading, and it also contributed to increasing the complexity of the aggregator token model. The economic model of the trading aggregator is difficult to design. If a transaction commission is implemented, it is easy for other aggregators that engage in price wars to seize the market, so trading aggregators basically do not charge commissions. The introduction of the RFQ model provides some uses for protocol tokens. Taking 1inch as an example, since the RFQ model requires the introduction of market makers, market makers need to pledge a certain amount of Para before entering 1inch to perform the RFQ function. The distribution ratio is determined by the $Para voting ratio.
3) RFQ’s Supplement to Cross-Chain Transactions
At present, the most common cross-chain trading mode is still AMM, but the RFQ mode has a natural advantage for cross-chain assets. The AMM cross-chain liquidity solution has problems with liquidity fragmentation and efficiency: each additional chain requires adding duplicate liquidity pools to the chain, and the exchange rate will also be greatly affected by the depth of the pool. However, using the RFQ model with market makers as intermediaries will not have the above liquidity fragmentation problem. The specific solution is to introduce Brokers on different chains, and a public ledger manages the accounts of Brokers on different chains. When users have the need for cross-chain assets or cross-chain swaps, they only need to connect with the Brokers on the corresponding chain.
Of course, the above description of Broker’s participation in cross-chain transactions is just a simple description, and the actual process will be more complicated. Currently, projects that use similar methods to achieve cross-chain transactions or cross-chain assets include WooFi DEX and UxUy. WooFi DEX has been connected to multiple networks and allows cross-chain swaps between multiple currencies, while UxUy is still in the testnet stage.
2. Changes in the revenue structure of the veToken model — the introduction of the Core Pool
The veToken model is a synonym for a type of DEX, representing a series of DEXs that use Gauge Voting to determine the release ratio of tokens in different pools, including Curve, Balancer, and various Ve(3,3) DEXs. Taking Curve as an example, the closed-loop logic of the veToken model is that veCRV holders can obtain discounts on veCRV bribery from all future B-side protocols. The continuous issuance of $CRV is essentially just overdrawing future bribery income. However, as DEXs, Curve and Balancer themselves do not seem to play the role of DEX transactions. They are more of a liquidity protocol, accumulating liquidity for the market through continuous subsidies of tokens, and the tokens are then enabled through B-side bribery. This model actually causes a waste of liquidity, and most of the liquidity "lying" in Curve is not circulated in the market.
This issue is actually a controversial point in the veToken model: it attracts a large amount of liquidity through high subsidies, but these liquidity do not play the role of high-speed liquidity, and thus cannot generate external benefits. Therefore, for the entire Curve system, the only external benefits that the system can obtain, in addition to a very small swap fee, are the bribery of veCRV by the B-side protocol.
In an economic sense, token subsidies can be understood as the system's overdraft of future external revenue debt issued to users. This debt can only be repaid by the external revenue obtained by the protocol. In the example of Curve, the debt issued by the system is the $CRV reward issued in each period, and the external income is the B-side’s bribery for veCRV and part of the swap fee. If Curve's external sources of income are added to Curve's entire economic design, Curve's model of DEX will significantly improve the debt level of the system.
Therefore, the current status of Curve-based DEX can be summarized as follows:
Liquidity in Curve is not fully utilized
Adding external revenue sources to Curve could significantly improve the system's debt levels
The solution is obvious: leverage the underutilized liquidity within the Curve model DEX to add external revenue sources to Curve.
1) Introduction of Boosted Pools
The first DEX to propose the above solution is Balancer, whose Boosted Pool mechanism aims to increase LP returns. Boosted Pool is a more advanced form of liquidity formed by combining Composable Stable Pools, Weighted Pools and Linear Pools. Weighted Pools are weighted AMM pools. The common AMM weight is 50/50. Weighted Pools can freely adjust the weight when the pool is set up. For example, Radiant Capital's dLP is a RDNT/ETH trading pair set with an 80/20 weight; Linear Pools are liquidity pools specifically set up for a certain asset and its derivatives, such as DAI and aDAI. Usually, Weighted Pools will work with Linear Pools to form Nested Linear Pools. Taking DAI and aDAI as an example, the Nested Linear DAI Pool is a DAI-aDAI liquidity pool composed of 20/80 weights, and the LP token issued by it is represented as bb-a-DAI. Similarly, USDC and USDT also have corresponding bb-a-USDC and bb-a-USDT. At this time, if bb-a-DAI, bb-a-USDT and bb-a-USDC form an LP pool, the pool is called Composable Stable Pool, and the LP token issued by it is also called bb-a-USD.
Although bb-a-USD is an LP token, it is still in the form of ERC-20 token standard, so it can be paired with various assets in the AMM pool as a paired asset. Among them, the more typical one is the bb-a-USD/ETH trading pool. The characteristics of this trading pool include:
The pool can perform asset swaps between ETH, DAI, USDT, USDC, aDAI, aUSDT, and aUSDC.
Due to the existence of aToken, bb-a-USD generates interest at all times
The above-mentioned AMM trading pool that generates interest all the time is called Boosted Pools in Balancer's definition.
The main reason for Boosted Pools' high capital utilization efficiency is that it leverages assets by issuing LP tokens. The initial AMM pool needs to maintain sufficient liquidity depth to ensure price stability when swapping assets. This means that most of the assets in the AMM pool do not perform the function of asset swaps but the function of stabilizing the coin price. The innovation of Boosted Pools is to split the assets with these two functions. Some assets still perform the function of asset swaps, while the other assets used to stabilize the coin price are taken to external protocols for income, and LP tokens are issued through external protocols to perform the function of stabilizing the coin price of the original AMM pool.
However, the high capital utilization efficiency of Weighted Pools can only benefit LPs, and cannot improve the debt level of the entire system from the protocol level. Therefore, after the BIP-19 proposal, Balancer proposed Core Pools, the main purpose of which is to correct the defect that Boosted Pools can only benefit LPs, so that the Balancer protocol itself can also benefit from Boosted Pools.
2) Introduction of Core Pools
The core proposals of BIP-19 include the following:
Select some Weighted Pools to form Core Pools
50% of the interest-bearing income in Core Pools is distributed to LP, 17.5% is allocated to BalancerDAO, and 32.5% is used as Bribes to bribe veBAL holders to vote for Core Pools to increase the TVL of Core Pools
25% of Bribe funds are allocated directly to Aura Finance’s vote buying market
The rest of the bribery funds were distributed to Hidden Hands, a subsidiary of the Redacted Cartel.
The introduction of the Core Pools mechanism is actually a guide for the future development trend of Balancer, guiding other capital pools to transform to Boosted Pools as much as possible. Since the BAL released by Balancer for liquidity incentives every week is shifted to Core Pools, the $BAL left for liquidity incentives in other liquidity pools will be greatly reduced. Therefore, for these liquidity pools, joining Core Pools as much as possible is the most profitable choice.
Prior to BIP-19, protocol revenue was distributed among stakeholders as follows:
50% of transaction fees go to LP, 17.5% to BalancerDAO, and 32.5% to veBAL holders
100% of Bribe is allocated to veBAL holders
After BIP-19, the protocol revenue distribution will be as follows:
50% of transaction fees go to LP, 17.5% to BalancerDAO, and 32.5% to veBAL holders
100% of Bribe is allocated to veBAL holders
50% of the interest-bearing income in Core Pools goes to LP, 17.5% is allocated to BalancerDAO, and 32.5% is used as Bribes to bribe veBAL holders to vote for Core Pools
The interest-bearing income of Core Pools is diverse, coming from lending protocols such as AAVE, as well as LSD assets such as stETH. From the perspective of asset replacement, the LP of Core Pools uses 50% of the interest-bearing income in exchange for the token incentive bribed by 32.5% of the interest-bearing income. As long as the value of this part of BAL is higher than the original 50% interest-bearing income, then the LP is profitable overall. As for veBAL, due to the real value of BAL.
In general, Core Pools has changed the revenue structure of Balancer and added external interest-bearing asset income as a source of Balancer system revenue. Compared with other veToken-based DEXs, it is more conducive to reducing the debt of the entire system.
3. Uni V3 with Liquidity Incentive Model — Reshaping Liquidity Position
During the Arbitrum airdrop, TraderJoe captured a large amount of $Arb trading volume due to the particularity of its V2 algorithm, making the Uni V3 narrative a hot topic in the short term DeFi, and also hyping Timeless Finance with Uni V3 liquidity incentive attributes to the forefront. Therefore, there is a voice that "the Uni V3 model with liquidity incentives will lead the next DEX development narrative." There are currently two forms of liquidity incentives and vote buying on the chain. One is about the LP liquidity depth incentive, and the other is price-related, about liquidity incentives in a specific price range. There are big differences between these two liquidity incentive models. The former is mainly to increase the LP liquidity depth, regardless of price; while the latter requires price guidance, which is mainly achieved by changing the market-making income in different price ranges.
From the perspective of market size, deepening liquidity is a demand that all protocols need to face, while protocols that have the need to guide prices are limited to certain anchored assets, such as ETHLSD and stablecoins. Therefore, from the perspective of the breadth of demand, the demand for deepening liquidity is broader and the demand for price guidance is more limited. According to the classification of the above models, the liquidity incentive model of Curve and Timeless Finance for Uni V3 can be summarized as the former, focusing on providing deeper liquidity; while Maverick Protocol should be classified as the latter, providing more efficient and accurate liquidity.
Currently, there are two pain points in the liquidity incentives of the AMM model:
For the AMM model of Uni V3, only the liquidity near the price range is active, and the liquidity outside the price range is hardly used. However, in terms of the incentives for liquidity depth, these two types of liquidity with different activity levels receive the same rewards.
For the Curve model, incentives for the liquidity depth of the anchored asset may aggravate the depegging status of the anchored asset.
The second point may be difficult to understand. Take stETH as an example. Before the Shanghai upgrade, since the ETH deposited in the beacon chain could not be withdrawn, stETH had a liquidity discount and the price was about 0.98 ETH. At this time, if you want to get the reward of stETH - ETH Curve LP, you must add liquidity to the Curve pool at the exchange rate of 0.98 ETH = stETH. The added liquidity did not improve the depegging status of stETH, but added deeper liquidity to the original price. This will hinder the price of stETH from returning to the anchored state.
Maverick Protocol has provided its own solution to the above two demand pain points, which is called Boosted Position. The official uses "surgical" to vividly describe the guiding role of Boosted Position on liquidity. Taking stETH as an example, the "surgical" function of Boosted Position on stETH is mainly reflected in two aspects:
Give additional incentives to LPs that automatically adjust their positions to the right following the price of stETH
Give additional incentives to LPs who make markets in the 0.98 - 1 range
Therefore, it can be seen that Maverick's liquidity guidance is mainly reflected in two aspects: price change direction guidance and liquidity market making range guidance. Through additional income incentives, the form of Liquidity Position is reshaped, and at the same time, liquidity incentives are all around the spot price, which improves the efficiency of fund utilization used for incentives.
The next DEX that may introduce specific price range liquidity incentives is Traderjoe, mainly because its special V2 mechanism can seamlessly integrate interval liquidity incentives. Traderjoe's special mechanism includes the following points:
The price range is planned in advance, and LP can only add LP within the planned range.
The interval between two adjacent scales is called 1 bin. Each bin corresponds to a single point of liquidity. The transaction price will not change before the liquidity in the bin is consumed.
Due to TraderJoe’s clear division of price ranges and the vertical integration of LP liquidity in a single bin range, the distribution of liquidity incentives among LPs will be greatly simplified. The simple distribution of liquidity incentives means that introducing vote buying in TraderJoe may only require adding a third-party component.
Overview
The examples of DEX listed above are descriptions of the changes and development possibilities of DEX from a relatively macro perspective. In fact, DEX has some interesting small innovations, such as Crocswap's mechanism to automatically adjust the swap rate according to the transaction volume, Cowswap's Batch Auction for MEV prevention, Cow Protocol's Coincidence of Wants protocol, and 1inch's anti-MEV rabbit hole function jointly launched by Metamask. These small innovations of DEX are tools that users can choose to use when executing swaps. In addition, studying the profit models of some free protocols, such as 1inch and WOOFi, is also an interesting direction. "We all thought that the development of DeFi has come to an end, but in fact it is constantly trying to explore new mechanisms and business models." Whether it is RFQ, the new Balancer revenue model or Uni V3-style vote bribery, it seems that only time can tell whether it is a gimmick or the future.
