Solvency Distress Contagion Risk: Network Structure, Bank Heterogeneity and Systemic Resilience

Abstract

We systematically analyse how network structure and bank characteristics affect solvency distress contagion risk in interbank networks. As interbank networks become more connected and more regular in structure, the contagion risk of system-wide shocks and individual bank defaults initially increases and then decreases, all else being equal. The low density heterogeneous network structures that typify real interbank networks are particularly prone to solvency distress contagion risk, when banks are similar in balance sheet size and capitalisation. However, when networks are calibrated to UK data, the higher capitalisation of large, highly-connected banks relative to their interbank exposures significantly increases the resilience of the system and reduces the importance of network structure. These findings reinforce the importance and effectiveness of imposing higher capital buffers on systemically important banks and of policies that limit interbank exposures. We also demonstrate that for real-world interbank networks, simple network metrics other than individual bank connectedness do not provide robust indicators for monitoring solvency contagion risk, suggesting that policymakers should continue efforts to model these risks explicitly rather than rely on simple aggregate indicators.

Macro-Prudential Stress Test Models: A Survey

“…This is in line with the findings of Nier et al (2007), which find that the better capitalised banks are, the more resilient is the banking system against contagious defaults and this effect is non-linear. Abduraimova and Nahai-Williamson (2021) confirm this finding for the UK interbank system building on the model of Bardoscia et al (2019). Similar findings have been found for the Brazilian interbank market in Souza et al (2015), and for the Euro Area when considering direct contagion risk related to the bail-in of other banks holding securities issued by banks entering resolution (Hüser et al (2018)).…”

A Hybrid Network Analysis and Machine Learning Model for Enhanced Financial Distress Prediction

“…The use of complex network theory to model the structure of the financial system and analyze risk contagion, particularly in banking systems, has been widely adopted by [57], [58]. This approach has allowed for the systematic analysis of how network structure and bank characteristics affect solvency distress contagion risk in interbank networks [59]. As the economic and financial system becomes more complex, the use of complex networks to study systemic risk and risk contagion has become increasingly important [60].…”