Estimating risk propagation between interacting firms on inter-firm complex network.

Area of Science:

• Economics
• Financial Risk Management
• Network Science

Background:

• Financial systems are susceptible to cascading failures, where one bankruptcy can trigger others.
• Understanding the dynamics of risk propagation is crucial for economic stability.

Purpose of the Study:

• To empirically derive a stochastic function for risk propagation in chain-reaction bankruptcy events.
• To identify key factors influencing the probability and spread of financial distress.

Main Methods:

• Analysis of over 5,000 chain-reaction bankruptcy events in Japan (2006-2015).
• Formulation of a probability function based on firm size and interaction.
• Development of a numerical model to simulate firm ecosystems and risk spreading.

Main Results:

• A stochastic function for risk propagation was empirically derived.
• The probability of cascading failure is proportional to the product of the first bankrupt firm's size (α-th power, α > 0) and the chain-reaction bankrupt firm's size (β-th power, β < 0).
• Larger initial bankruptcies pose a higher risk to smaller trading firms.

Conclusions:

• The interaction kernel is a key factor in modeling the complexity of bankruptcy risk spreading.
• Findings provide insights into the mechanisms of financial contagion and systemic risk.