Time-Varying Reproduction Number as a Monitoring Tool for Bankruptcy Risk Assessment

This study introduces a novel application of the time-varying reproduction number , a core concept of modeling infectious disease in epidemiology, to the domain of financial distress and bankruptcy contagion. By modeling firm bankruptcies as a transmissible process across interconnected economic agents, we estimate -based on the temporal incidence of bankruptcy events and a calibrated serial interval distribution derived from empirical lag structures. We apply this framework to major financial crises—including the dot-com bubble, the 2008 global financial crisis, the shale oil shock, and the COVID-19 pandemic—and demonstrate that consistently exceeds 1 prior to the surge in bankruptcies, signaling early contagion dynamics. A case study of South Korea further confirms the model's ability to differentiate between systemic contagion phases and structurally stabilized periods. Compared to traditional financial ratios or ex-post econometric models, the -based approach offers a lightweight, ex-ante early warning system that is both theoretically grounded and empirically practical. However, its limitations—such as sensitivity to structural breaks, oversimplified variance assumptions, and the need for integration with macro-financial indicators—highlight avenues for methodological refinement. Our findings suggest that the framework can serve as a complementary monitoring tool for policymakers and financial regulators aiming to detect and mitigate systemic risk.