The determinants of funding liquidity risk in decentralized lending

Abstract

Decentralized lending in the DeFi ecosystem mirrors traditional financial intermediation but poses significant risks, particularly funding liquidity risk, due to the volatility and composbility of digital assets, high leverage, and the absence of regulatory protections. This study applies traditional financial intermediation theories to DeFi lending and empirically test which internal factors such as interest rates and user market power, as well as external factors like the USD Index, influence funding liquidity risk in DeFi lending. Analyzing high-frequency blockchain data using the ARDL model and a novel dynamic ARDL simulation from major pools such as Wrapped Bitcoin (WBTC) and Wrapped Ethereum (WETH), the research finds that current algorithmic interest rate models fail to function as effective self-stabilization mechanisms. Additionally, lower deposit concentration in these pools may exacerbate, rather than mitigate, funding liquidity risk.