Impermanent loss has become one of the important problems of decentralized exchanges. Many projects have proposed different solutions for this. Recently, I saw a project solve the problem of impermanent loss by coupling two AMMs of uniswap and balancer. The idea of reducing impermanent loss by dynamically adjusting the weight of tokens seems to have been firstly proposed by bancor before they released the V2 version. However, they did not choose this technical route in the end and launched a impermanent loss protection by inflation tokens compensation. The reason is very simple. Although dynamic weight adjustment can reduce impermanent loss, However, it increases trading slippage and hinders market activity. It also can be seen from the data in their white paper. Compared with uniswap, sync-AMM with dynamic weights increases the maximum slippage from 1.5% to tripled 4.5% under the same condition. This increase in trading costs is fatal to traders. Although their backtest data shows that LP returns have also increased compared to uniswap or balancer, the test data does not consider traders’ trading costs. The problem is that most traders will not choose it unless it is profitable to arbitrage. However, the transaction will become very inactive if just only by the promotion of arbitrage, and the income of LP fees will also be greatly reduced. Therefore, through dynamic weights to reduce impermanent loss is not an efficient solution.
