Tail Risk Spillover Effect in a New Financial Institution Network Based on Extreme Value and Entropy Theory

This article takes 57 listed financial institutions in China from 2014 to 2024 as the research sample. Combining extreme value theory and the TENET method, it constructs a tail risk spillover network and analyzes the network structure and relationships from three levels: system, industry, and institution. Node entropy indicators are also used to help assess the level of institutional risk spillovers. The study found that the tail risk connectivity in China’s financial system shows cyclical fluctuations, significantly affected by extreme events, international economic frictions, and regulatory policies. It rises during crisis periods and gradually recedes afterward. Risk spillover effects within each industry are significant; fintech institutions have the strongest risk absorption capability and are the most sensitive to external shocks. Their risk spillover effect on traditional financial sectors is stronger, making them a key hub for systemic risk. Large state-owned banks and leading fintech institutions, as core nodes in the network, dominate risk accumulation. By the end of 2023, various risk events significantly intensified risk transmission in the insurance and fintech sectors. Securities institutions have the weakest risk resistance and are most sensitive to market fluctuations, while the risk trends in banking and insurance are relatively stable. Influenced by both their network hub positions and technological innovation characteristics, fintech institutions have significantly higher inherent risk overall.