The Predictive Power of Systemic Risk Indicators Under Extreme Scenarios

Abstract

Against the backdrop of an increasingly complex global financial environment and persistently rising market uncertainty, the importance of accurately identifying systemic risk and strengthening macroprudential supervision has become ever more prominent. Using data on China’s listed financial institutions from 2008 to 2025, this paper focuses on systemic risk indicators of three types—risk contagion, tail risk, and capital fragility—and examines their predictive power under extreme scenarios from three perspectives: macro-level timeliness, financial institutional market performance, and early warning of bank distress. The findings are as follows: (1) The fluctuations of systemic risk indicators are closely associated with periods of financial stress, demonstrating good macro-level timeliness. (2) Systemic risk indicators can effectively predict the decline in stock returns and the increase in stock return volatility of financial institutions during stress periods, offering forward-looking information for financial institutional market performance. (3) Systemic risk indicators can effectively predict bank distress. The results of multi-horizon prediction show that risk contagion indicators exhibit strong cross-horizon stability, tail risk indicators provide the best medium-term early warning, and capital fragility indicators are more suitable for contemporaneous warning. The findings suggest that systemic risk indicators possess stable forward-looking information value under extreme scenarios, thereby providing empirical support for improving macroprudential supervision and early-warning systems for financial risk.