LIQUIDITY RISK MANAGEMENT IN DECENTRALIZED FINANCE PROTOCOLS UNDER CONDITIONS OF CRYPTOASSET MARKET VOLATILITY

Abstract

The rapid penetration of decentralized financial mechanisms into the structure of Ukraine's cryptocurrency market, where the volume of DeFi transactions consistently exceeds that of centralized platforms, highlights the need for effective tools to maintain the instant solvency of lending protocols during periods of sharp price fluctuations in digital assets. The purpose of this article is to systematize liquidity risks in decentralized financial systems, conduct a comparative assessment of algorithmic strategies for their minimization, and identify ways to enhance the stress resilience of protocols. The methodological basis of the study consists of a taxonomic analysis for classifying types of risks, a comparative analysis of the effectiveness of key liquidity management strategies, economic-mathematical modeling of cascading liquidation processes, and a correlation analysis of the relationship between the magnitude of cryptoasset price declines and the frequency of protocol failures. The empirical basis consists of on-chain data on the transaction activity of leading DeFi protocols for the period 2024–2026. The results show that hybrid configurations–which combine dynamic interest rate regulation with overcollateralization and decentralized oracle networks–demonstrate the highest resilience to extreme volatility. It was found that increasing collateral requirements proportionally reduces the probability of cascading liquidations, but simultaneously limits the protocol's capital efficiency, highlighting the need to optimize these parameters. It has been established that compositional links between protocols create a domino effect: a local liquidity shortage in one pool can trigger a chain of forced liquidations in adjacent systems within a critically short time frame. The scientific novelty lies in the development of a typologized scheme for neutralizing liquidity threats, which, unlike existing ones, integrates sentinel oracle, execution liquidation, and reserve insurance instruments into a unified protocol risk management system. The method for estimating margin call thresholds has been improved to account for the historical volatility of specific cryptoassets. The practical significance of the obtained results lies in their potential use by DeFi protocol architects and smart contract developers when designing risk management systems, configuring liquidation auction parameters, and selecting the optimal configuration of oracle networks for the Ukrainian crypto market.