A Dynamic Liquidity-Proportionate Fee Model for Mitigating Impermanent Loss in Decentralized Automated Market Makers

Decentralized Automated Market Makers (AMMs) have transformed financial markets by enabling permissionless, peer-to-contract trading. However, Liquidity Providers (LPs) in constant-product AMMs remain highly exposed to Impermanent Loss (IL), a systematic risk arising from price divergence between pooled assets. This paper introduces a novel Dynamic Liquidity-Proportionate Fee (DLPF) model designed to mitigate IL by dynamically adjusting transaction fees in response to real-time price volatility and pool imbalance. Unlike static fee structures, the DLPF model increases transaction costs during periods of high divergence to extract premium from toxic arbitrage flows, redistributing these gains to LPs. Conversely, the model reduces fees during low-volatility periods to encourage organic trading volume. We evaluate the DLPF model using a high-fidelity agent-based simulation calibrated with historical high-frequency data from the ETH/USDT and WBTC/ETH pools. Our results demonstrate that the DLPF model reduces impermanent loss by up to 42.4% while maintaining 88.7% of the baseline trading volume. The findings suggest that dynamic fee scaling offers a robust, oracle-free solution to the LP profitability dilemma, contributing to the long-term stability and sustainability of decentralized liquidity pools.