Restraining Overconfident CEOs through Improved Governance: Evidence from the Sarbanes-Oxley Act

Behavioral CEOs: The Role of Managerial Overconfidence

“…On the other hand, measures to improve active monitoring of the firm-to the extent that the monitors themselves are rational-may address both concerns. Recent work has begun to look seriously at how and when governance can mitigate managerial biases (Banerjee, Humphery-Jenner, and Nanda 2015b;Kolasinski and Li 2013), which seems to offer a promising direction for additional research. Ultimately, behavioral biases like CEO overconfidence matter not just for the choices and outcomes of the agents who are subject to them, but also for the (potentially rational) agents with whom they interact, transact, and contract in the market place.…”

Litigation risk: Measurement and impact on M&A transaction terms

“…This variable is introduced only once an acquisition has been announced. –Overconfident CEOs can potentially overpay in takeovers, triggering litigation. We have calculated the overconfidence dummy variable following the methodology implemented by Banerjee, Humphery‐Jenner, and Nanda () and Malmendier and Tate (). This variable refers to the first quartile of overconfident CEOs identified by the holding of unexercised but exercisable stock options.…”

“…total assets minus all debt; LN_ACQ_TOT_ASS : Logarithm of total assets of the acquirer (US$ million); LN_SALES : Logarithm of the acquirer's total sales in the year preceding the transaction; LN_TARG_TOT_ASS : Logarithm of total assets of the target the year before the transaction (US$ million); LN_TRANS_VAL : Logarithm of transaction value (US$ million); OVER_CONF : Dummy for overconfident CEOs measured using the in‐the‐money stock options methodology of Banerjee et al. (); ROA_N‐1 : Acquirer's pre‐tax return on assets, calculated the year previous the transaction year; REL_ACQ_PPE : Ratio of properties, plant, and equipment, divided by total assets; SQ_LN_ACQ_NET_ASS : square of LN_ACQ_NET_ASS ; SQ_LN_SALES : square of LN_SALES ; SAME_SECTOR : Dummy equal to 1 if the acquirer and the target belong to the same industry sector. Thomson Financial classification in 74 sectors is used; TARG_ N : Number of citations of the target during the year of the transaction ending at the announcement date; TARG_SALES : Target sales over the last year preceding the transaction (US$ million); TARG_OV_ACQ : Relative litigation risk of the target compared with the acquirer.…”

New Entry Threats and Firm Performance in the IT Industry: The Moderating Role of Board Independence

“…One of the key requirements of SOX was that a majority of directors on a firm's board should be outsiders, or independent directors who has 'no material relationship' (either directly or as a partner, shareholder or officer of an organization that has a relationship with the company) with the listed company (Banerjee et al 2015;Chhaochharia et al 2016). The enactment of SOX was thus associated with a significant, exogenously mandated increase in the number of independent directors for non-compliant firms (i.e., firms with fewer than 50% independent directors prior to SOX), but did not significantly affect the board independence of compliant firms (i.e., firms with greater than 50% independent directors prior to SOX).…”

“…Whereas SOX was enacted in mid-2002, we also use an alternative cutoff to construct the timing indicator that equals one if the observation occurs in year 2003 or later as a robustness check (Chhaochharia et al 2016). The instrumental variable for Board Independence is thus defined as the interaction of the non-compliant group*post-SOX dummy (Banerjee et al 2015). For models with interaction NET × Board Independence, we further interact this instrument with NET to address the potential endogeneity of NET × Board Independence in the full model.…”