Vulnerability in global financial networks

Vulnerability in the financial system leads to economic instability. One way to reduce economic uncertainty and ensure sufficient capital flows is by detecting, monitoring and responding to the transmission of shocks in the financial network. Interconnectedness among integrated financial markets not only provides opportunities for economic expansion, resource allocation and improved risk-sharing, it also creates channels through which vulnerabilities spread across these financial markets. This thesis is motivated by the reoccurrences of crises such as the global financial crisis of 2007 - 2009, and the gaps in the literature to discover means to detect, assess and respond to extreme events that lead to financial instability. Objectively, we present empirical assessments of vulnerability in financial networks and employ different approaches to detect vulnerability in the global financial system. Combined Granger causality and the Diebold-Yilmaz approach are used to measure interconnectedness. A spatial autoregressive framework in capital asset pricing model is utilised to study the impact of network exposure on structural model, and generalised historical decomposition in vector autoregression is applied to measure signed spillovers. A portfolio mimicking framework using moments conditions is used to detect contagion, while generalised historical decomposition in Markov-switching vector autoregressions is employed to detect signed spillovers under different market conditions. Empirical evidence in this thesis reveals the changing structure of the financial network across different markets. Over time, new links are formed, and others removed due to changing relationships among different markets. These interconnections become larger and more complex acting as a channel through which shocks are transmitted and amplified throughout the entire financial system. Moreover, a growing influence of the Asian markets in spreading shocks is observed. Transitions of the network structure of different markets are discussed with emphases on individual Asian markets. In studying the dynamics of the network exposure on equity markets, our findings highlight the role of network exposure in increasing vulnerability, with both interconnectedness and network intensity playing key roles in monitoring these exposures. Our findings also indicate evidence of changing spillovers and contagion at a given time point, implying that vulnerability in the entire financial system changes over time. By assessing various forms of vulnerabilities in the global economy under different economic conditions, our findings indicate that vulnerability in the financial system changes depending on the state of the market. For instance, crises are associated with intense spillover regimes while normal times are associated with moderate spillover regimes. Transmission channels of shocks become more complex with the increasing integration and globalisation within the financial system. Our empirical findings assess different levels of vulnerability with the aim of providing guidance in making policy decisions to monitor the financial system, thereby promoting financial stability.