Three essays on financial risk

The thesis contributes to the literature on behavioural finance and price jumps by moving beyond the traditional framework of the efficient market hypothesis. The aim of the thesis is to deepen the understanding of actual financial markets rather than the ideal world described by that hypothesis. This thesis empirically investigates three related topics: managerial traits; the incentives for risk-taking provided by managerial compensation packages; and price jumps. The first empirical chapter of the thesis examines the varying effect of CEO overconfidence after the experience of a severe financial shock. It investigates the association between CEO overconfidence and a firm’s risk-taking behavioural during pre- and post-crisis periods. We find that the association between CEO overconfidence and firms’ risk-taking becomes weaker during the post-crisis period. This work contributes to the literature on the state dependence of executive managerial traits. The second empirical chapter examines the association between components of managerial compensation packages that provide incentives for firms to undertake risky investments and default risk in the banking industry. It investigates the association between CEO vega (the sensitivity of CEOs’ wealth to stock price volatility) and default risk in the US banks. We find that the positive association between CEO vega and default risk is mitigated by high leverage ratios. This work contributes to the literature on managerial compensation in different financial conditions. The third empirical chapter examines the association between price jumps and market liquidity in the context of cross-listed stocks. It investigates intraday patterns in market liquidity and price jumps for cross-listed stocks during overlapping and non-overlapping trading hours. We find that the improved market liquidity during overlapping hours is associated with fewer price jumps. Moreover, we find that overlapping-hours phenomenon is sensitive to varying overlapping hours caused by daylight-saving time. This work contributes to the literature on the incidence of price jumps for cross-listed stocks. Collectively, the three studies empirically explore biased behaviours of CEOs and non-random-walk movements of stock prices, which cannot be well explained by the classical efficient market hypothesis