Director Liability Protection and the Quality of Independent Directors

Abstract

We study whether legal liability protection helps companies to recruit and retain high-quality independent directors. We conduct difference-in-differences analyses exploiting the 1999 Ninth Circuit Court of Appeals Ruling on the Silicon Graphics case, which substantially raised the bar for filing securities class action (SCA) lawsuits as a shock. We document supporting evidence for the talent attraction hypothesis by showing improvements in newly recruited independent director quality following the ruling, but only for candidates who are previously not exposed to SCA litigation risk. The effects are stronger for firms facing greater litigation risk ex ante or smaller local supplies of director candidates. Results are more evident for experience-based quality dimensions. We also analyze a sample of voluntary independent director departures and find little support for the talent retention hypothesis, suggesting that more complex factors enter into a director’s continuation decision once a director is already exposed to SCA litigation risk. A policy implication is that liability protection can be useful in attracting more unexposed high-quality candidates to the pool of public boards but does little to attract high-quality candidates who are already in the pool of public firms. This paper was accepted by Gustavo Manso, finance. Funding: H. Zou acknowledges the financial support from a competitive General Research Fund (GRF) grant from the Research Grants Council of the Hong Kong Special Administrative Region, China [Project No. HKU 17513816]. S. Shen acknowledges the financial support from the National Natural Science Foundation of China [Grants 72202164 and 72332003]. The work was partially carried out while R. W. Masulis visited the University of Hong Kong as an FBE visiting researcher, and he thanks the Faculty of Business and Economics of the University of Hong Kong for its sponsorship. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2020.03743 .