On one level, it is surprising to see the recent surge in cases of bank misconduct. One explanation holds that when a CEO has too much authority within the firm, misconduct is but one potential outcome (Khanna et al. 2015). However, by most accounts, oversight of CEO decision-making has improved markedly in recent years. Data from Riskmetrics show that eight out of ten US bank board members are classified as independent in 2012, up from around half in 2000 (see Figure 1). With increasing levels of independence, one would expect bank boards to be more effective in preventing misconduct. However, far from a declining trend, the number of enforcement actions has increased from 5 to 28 over the same time period.
Figure 1. The percentage of directors classified as independent on US banks, and the number of enforcement actions received by US banks
Source: Riskmetrics and Nguyen et al. (2015)
The rise in cases of bank misconduct under increasingly more independent boards is consistent with the view that true board independence is difficult to achieve. Board independence can be undermined if CEOs exert intangible influence over those charged with monitoring them. One way in which a CEO could wield soft power is by ‘capturing’ the board through director appointments (Khanna et al. 2015). Since the CEO is typically involved in the process of recommending directors to the board, directors appointed during the tenure of the current CEO have an incentive to return the favour (Coles et al. 2014, Khanna et al. 2015). Even independent directors may reciprocate the CEO’s requests and side with the CEO to support, engage in, or conceal wrongdoing.
Thus, only directors appointed before the current CEO’s tenure are free from this type of intangible influence, and are therefore capable of objectively monitoring the CEO. In the case of Wells Fargo, around half of all current members of the board have been appointed under the bank’s current CEO. This ratio is not unusual for US banks, but it raises questions over how independent, assertive, and effective Wells Fargo’s and the boards of other US banks really are.
In a recently published study, we employ a unique dataset of regulatory enforcement actions issued by the three US supervisory bodies – the Federal Reserve Board (FRB), the Federal Deposit Insurance Corporation (FDIC) and the Office of the Comptroller of the Currency (OCC) – against banks that engage in unsafe, unsound and illegal banking practices (Nguyen et al. 2016). Generally, one can only observe detected misconduct (once an enforcement action has been issued), but not the set of all committed cases of misconduct. That is, even in the absence of enforcement actions, a bank may still be engaged in undetected misconduct. To address this problem of partial observability, we follow Wang (2013) and Wang et al. (2010) in employing a bivariate probit model that disentangles committing misconduct from the detection of misconduct, conditional upon misconduct having occurred.
We find that a bank in which monitoring quality is high (i.e. all directors have been appointed before the CEO takes office) has a 27% lower probability of committing misconduct and a 35% higher probability of detection (conditional upon misconduct having occurred) compared to a bank where all directors have been appointed under the current CEO. Further, in all specifications, we control for the proportion of independent directors and the number of directors with financial expertise. We find that these traditional measures of board monitoring and advising have little or no power to prevent misconduct in the banking sector.
Interestingly, we also show that CEO bonus and CEO option payments are positively related to the probability that misconduct is committed. This is consistent with the argument that some CEOs commit wrongdoing in order to boost stock prices and enjoy higher pay-outs.
How do board monitoring and advising prevent bank misconduct? We examine two channels that help explain the results. First, many enforcement actions are issued when bank fundamentals indicate increased bank risk. Our results show that better monitoring prevents enforcement actions because these boards are associated with higher bank capital cushions, lower portfolio risk, and fewer non-performing loans. Second, CEOs will be deterred from committing wrongdoing if they know ex ante that a board will penalise them for instances of misconduct. We find that boards not captured by the CEO are more willing to impose heavier penalties on them following detected misconduct. That is, after misconduct is detected, better monitoring quality is associated with a larger reduction in (i) the level of CEO pay, (ii) the level of CEO pay relative to the other top executives at the same bank, and (iii) the value of CEO risk-taking incentives.
Our study offers novel insights for how to structure bank boards to prevent misconduct. These include important lessons for the ongoing case at Wells Fargo and other banks. We show that conventional board measures such as board independence and financial expertise have no measurable impact on misconduct being committed or detected. By contrast, the board metrics we study related to monitoring and advising are important predictors of misconduct. Overall, our study illustrates the importance of board governance in banking. Our findings demonstrate that governance metrics revolving around CEO connections warrant more attention from regulators, investors, and governance activists.