Does network topology and structure affect the contagion effect of a collapse of a financial institution in an interbank market? A case study on the Barabási–Albert model, Watts-Strogatz model and Erdős–Rényi model

• Main Author: Chew, Wei Jian
• Other Authors: Fedor Duzhin
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Science::Mathematics::Discrete mathematics; Social sciences::Economic theory
• Online Access: https://hdl.handle.net/10356/148506
• Institution: Nanyang Technological University

This report takes inspiration from the 1998 Asian Financial Crisis, and the resulting cascade of insolvencies of banks in Thailand and Russia in the aftermath. We analysed the contagion effect a collapse of a financial institution would have on other financial institutions in an interbank market. Due to the vast difficulties in acquiring real-world data, a theoretical simulation-based approach was favoured. Graph theory is commonly applied to many different fields to investigate different forms of relationships. In an interbank market, nodes represent financial institutions, while the edges represent the lending relationships. Significant negative impact on a bank's collateral could provoke adverse effects on other banks in the network structure. In order to evaluate whether the type of network structure would impact the contagion effect differently, we simulated the collapse of banks in different forms of networks. The following network structures are used for the graphs: the Erdős–Rényi model, the Watts--Strogatz model, and the Barabási–Albert model. We investigate whether the contagion effect differs according to network topology and structure. Empirical evidence suggests that the Barabási–Albert model is more vulnerable to contagion effect in an interbank market with high banking system stability and low debt intolerance, whereas Watts-Strogatz model and Erdős–Rényi model are more likely to have longer-lasting contagion effect in an interbank market with low banking system stability and high debt intolerance.