Systemic Risk -- Identification, Assessment and Monitoring based on Eigenvector Centrality Analysis of Thai Interbank Connectivity Matrices

This paper presents a coherent framework for identifying Systemically Important Financial Institutions (SIFI) and assessing the overall banking system's vulnerability to systemic phenomena from the viewpoint of banking system as a complex network of interconnected nodes generating pair-wise financial activities. The proposed parameter-free methodology comprises two parts, Systemic Importance Analysis (SIA), which rates and ranks individual financial institutions relative to one another in terms of systemic influence each exerts over the banking system overall, and Systemic Vulnerability Analysis (SVA), which tracks the banking system over time in terms of how the very pattern of connectivities (inter-connectedness) somehow enable 'small players' to exert 'large leverage' on system (in)stability. SIA & SVA both utilizes Bonacich's Eigenvector Centrality measure, essentially the principal eigenvector associated with a matrix of pair-wise connection strengths. Empirical results presented herewith are based on actual records (2009-) of interbank transactions between financial instiutions operating in Thailand (actual bank names masked by aliases)