A Study of Systemic Risk in Economic System: Interbank System and Interindustry System Perspectives

Abstract

Nowadays, Systemic risk is an issue of great concern in economic systems, since these systems have grown ever larger and more complex, especially after the global economic crisis of 2008-2009, which has made it clear that a critical need for comprehensive study of systemic risk by focusing on the interconnectedness of economic systems. However, the related researches about the role of system architecture on systemic risk remain inadequate and imperfectly understood. This is not only due to the intertwined and sophisticated interdependencies among individual entities in the economic system, but also due to the lack of convincing empirical evidence on the mechanism of systemic risk. This dissertation takes up these challenges by introducing the techniques of network analysis into the setting of economic systems. In particular, we studies systemic risk in economic systems by taking into account two common relationships—the lending-borrowing and input-output relationships. So this dissertation is divided into two parts. The first part focuses on systemic risk in an interbank system which is determined by the lending-borrowing relationships among individual banks; the second part concentrates on systemic risk in an interindustry system where its architecture is represented by a web of input-output relationships among industries.

In the first part, we study the topic of systemic risk in an interbank system. In detail, we firstly focus on how to model this kind of system and the mechanism of systemic risk under the framework of lending-borrowing relationships and the stylized Balance Sheet. Based on this modeling, we propose a simple contagion algorithm to study the role of the heterogeneity of interbank system on systemic risk. The heterogeneity is measure by the diversification of interbank connections (reflected by the mean and variance of degree) and interbank exposure (reflected by the mean and variance of exposure). We further focus on how to reduce systemic risk to enhance the stability of interbank system by propose several intervention policies based on two widely used prudential approaches—forced merger and capital injection. The performance of the intervention policies are evaluated by comprehensive numerical experiments.

In the second part, we concentrate on how to measure the systemic risk in a particular interindustry system by viewing this economic system as a weighted directed network of interconnected industry. In particular, we firstly investigate the influence of the crisis of 2008-2009 on this interindustry system to witness the existence of systemic risk. We further measure the systemic risk by measuring the contribution of a given individual industry to the overall risk of the interindustry system—the systemic importance of individual industry— by a combined hyperlink-induced topic search (HITS) algorithm which considers both market information and network information among industries. Based on the stress testing, the performance of the combined HITS is compared with the purely market-based systemic risk measurement. The result shows that the combined HITS outperforms the baseline in find the top N systemically important industries.

This dissertation presents a better understanding to the systemic risk from both the theoretical and empirical perspective. In theoretical study, we partly clarify the interplay between the heterogeneity of system and systemic risk, which improve the management of the stability of interconnected system, especially the interbank system. The proposed intervention policies also help the managers to regulate a particular economic system. In the empirical perspective, the using of combined HITS algorithm can help to measure systemic risk, which have real policy relevance in terms of identifying entities (e.g. industries in the interindustry system) that are potentially systemically important entities for the entire network in case of economic and financial crisis. So our work has significant implications on the regulation and surveillance of an interconnected economic system, we hope this work can be a supplement for the vibrant emergent literature using network analysis to study systemic risk.

Date of Award	25 Oct 2016
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Shaoyi Stephen LIAO (Supervisor)