Do bank CEOs really increase risk in vega? Evidence from a dynamic panel GMM specification

Previous executive compensation studies find that firm risk increases in the risk-taking incentive (vega) of CEOs' compensation packages. However, the standard methodology of two-stage least squares (2SLS) regression can suffer from invalid instruments. Using a dynamic panel generalized method of moments (GMM) specification to control for dynamic endogeneity, unobserved heterogeneity, and simultaneity (Wintoki, Linck, & Netter, 2012), we find no evidence of a positive relationship between risk and vega for banking firms. Furthermore, across institutions, CEOs' pay-performance sensitivity (delta) positively relates to the risk. Finally, high-leverage banks and commercial banks seem less prone to risk increases in delta relative to the entire sample of financial institutions. These results are important to investors, boards, regulators, and creditors, as they are all concerned with the risk of the financial institution.