Essays on financial contagion and financial crises

Financial crises spread across countries through a variety of channels. A crisis originating in one market may propagate in an alternative form and have different impact in another market. The spread of crises through existing trade and financial linkages may be anticipated; however, during periods of stress, transmission in excess of these linkages is frequently observed. It is this excess transmission, known as financial contagion, which is the focus of this dissertation. Contagion is one of the important mechanisms by which a financial crisis may become systemic. Consequently it is important for policy makers and market participants alike to identify the channels by which crises transmit and to assess the extent of contagion risk in global markets. Understanding contagion risk helps frame policies to reduce the immediate impact of a crisis and to undertake long-term structural reform policies for financial stability. This dissertation examines contagion between international equity markets during the global financial crisis (GFC) of 2007-2009. It identifies channels (common factor, idiosyncratic factor, structural shift and volatility spillovers) by which crisis shocks transmit across borders and uses a variety of econometric techniques to test for the existence and extent of contagion during this period. The channels of contagion identified are shown to relate to the probability and severity of banking crises. While existing research largely focuses on tests for contagion at aggregate market level, this thesis compares aggregate results with those for the financial sector, and finds – perhaps surprisingly – that there is somewhat less evidence at the financial sector level. The empirical evidence in the dissertation shows that both advanced and emerging equity markets are exposed to contagion risk through the idiosyncratic shocks emanating from the crisis originating market (here the US). While emerging markets are as affected as developed markets at aggregate market level, the financial sectors of advanced economies are less exposed to the idiosyncratic shock channel than the emerging markets. Banking sectors across the world are shown to be significantly exposed to contagion. Most have evidence of contagion from some or all four channels. However, contagion associated with the idiosyncratic shock channel has the largest impact – it increases the probability of a banking crisis by 27 percent, suggesting policy makers may wish to design policies aimed at mitigating these effects. Finally, the dissertation empirically relates the evidence for systematic risk for large US financial firms with existing empirical measures of systemic risk. Using high frequency equity data to separate the responsiveness of these firms to jumps and continuous movements in market prices reveals that firms have higher jump beta than continuous beta. Further, the aggressiveness of the jump beta response of the institutions is positively related to systemic risk measured by capital shortfall and negatively to systemic risk measured by interconnectedness. The evidence compiled in this thesis confirms the existence and extent of contagion effects across international equity markets, including their banking sectors, focusing on the crisis of 2007-2009. It offers empirical evidence on the relative importance of considering the source and channel of the crisis transmissions as well as the structure of the domestic banking sector in formulating policy and investment portfolio responses to periods of stress