The market consensus during the financial crisis was that financial sector CEOs were engaged in excessive risk taking induced by compensation practices. Thus, the primary focus of this paper is to determine whether empirical evidence supports this assertion. As such, I examine bank CEO compensation, bank risks, and the relation between bank CEO risk taking incentives and bank risks and the effect of the 2007/9 financial crisis on this relation. I find that banks on average reduced their exposure to credit, capital, total, and unsystematic risks, and increased their exposure to liquidity, portfolio, off-balance sheet and (accounting) foreign exchange risks, from 2003 to 2006. These trends largely reversed during 2007 to 2009. During the 2007/9 financial crisis, banks experienced significant structural shifts in all risk indicators (except for capital and foreign exchange risks) which increased significantly consequent on the economic downturn. I also find that banks remained highly sensitive to changes in short- and long-term interest rates and foreign exchanges rates throughout the period. My findings also support a bank size effect. I observe consistent real growth in CEO base salary annually, from 2003 to 2009, which suggests that there is resilience in this form of compensation to the financial crisis. However, only small banks paid significantly higher base salary during the financial crisis to offset the similar decline in annual bonus payments caused by deteriorating financial and market performances during that time. I find that CEO portfolio option values were more responsive to changes in total risk during the pre-financial crisis period (2003 to 2006) than during the financial crisis (2007 to 2009). Also, I find evidence of banks size effects in compensation components, compensation structure and compensation sensitivity. My results are robust to other sample formations and statistical indicators. After adjusting for the simultaneity bias between bank CEOs' risk taking incentives (measured by the sensitivity of CEO option portfolio and pay for performance sensitivity) and bank risks (using accounting and market based measures), my findings reveal significant shifts in the relation between compensation and bank risks during the financial crisis. Specifically, during the financial crisis, CEOs with more sensitive pay for performance were related to banks with greater capital risk, and banks with higher portfolio risk had CEOs with more sensitive pay for performance. Also, banks with greater total and unsystematic risks during the financial crisis had CEOs with less risk taking incentives. Other indicators during the financial crisis show that less stable banks had CEOs with less risk taking incentives, while banks with greater asset return risk had CEOs with less sensitive option portfolios. Overall, these results do not support the risk inducing incentives of bank CEO compensation especially during the financial crisis.