Self-organised criticality and other phenomena in bank failures.
Banking failures propagate through financial links in the interbank money market. The phenomenon of these failures is investigated using a stylized interbank network model inspired by the dynamical rules of the Bak–Tang–Wiesenfeld (BTW) sandpile. The key modification made was to introduce features o...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
2012
|
Subjects: | |
Online Access: | http://hdl.handle.net/10356/48721 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | Banking failures propagate through financial links in the interbank money market. The phenomenon of these failures is investigated using a stylized interbank network model inspired by the dynamical rules of the Bak–Tang–Wiesenfeld (BTW) sandpile. The key modification made was to introduce features of interbank loans and debt. We found that participation in the interbank market is the most efficient solution in increasing the shock resilience of the network. Despite this, we demonstrated how liquidity hoarding can arise in the market when agents are left to choose a perceived optimal strategy and explained this observation by employing the analogue of the prisoner’s dilemma. We also showed how this problem is exacerbated when massive liquidity injection is utilised as the primary solution to arrest large-scale collapses. From these, we concluded that a more efficient solution is to also oblige banks to maintain participation on the interbank market during a crisis. Additionally, we justified the presence of SOC dynamics in banking failures by positively identifying scale-free behaviour of avalanche statistics on our network model even when it was subjected to modifications of increasing complexity and realism. We broke away from the lattice architecture of the BTW sandpile as well and applied our model on a scale-free network. In the critical state, we found that large-scale banking contagions can be predicted and attenuated if timely measures are taken at the onset of the first collapses. |
---|