RELATED LITERATURE REVIEW

2.1 Association between Equity-based Compensation and Audit Committee Effectiveness:

Early studies have examined the association between audit committee independence and earnings

management using pre-SOX data. However, the empirical results are mixed. For example, Klein (2002a) shows that the magnitude of abnormal accruals is negatively related to audit committee independence (measured by the percentage of outsiders on the audit committees) only when the majority, rather than 100 percent, of the members are independent. In a similar study, Bedard et al. (2004) finds that high discretionary accruals are negatively associated with 100 percent, rather than 50 percent, independent audit committee members (measured by the number of outsiders and whether members are compensated by options). In contrast to these studies, Xie et al. (2003) documents that the level of audit committee independence (measured by the percentage of outsiders) is not related to current discretionary accruals.

In a different strand, Carcello and Neal (2000) examines the relation between audit committee independence and auditors’ report types. They find that firms with greater percentage of affiliated directors on their audit committees are less likely to receive a going-concern opinion. Carcello and Neal (2003) further tests whether audit committees can protect the auditors from being dismissed following a new going concern opinion.

The empirical results indicate that audit committee members having more stock holdings are more likely to dismiss auditors after first-time going concern reports.

A potential problem with these pre-SOX studies is that they measure audit committee independence using the percentage of outsiders (who may or may not own firms’ shares) on the audit committees. This “insiders (or affiliated directors) vs. outsiders” dichotomy may be appropriate before SOX because the exchanges’ listing requirements allowed the appointment of inside-affiliated directors to the audit committees if it is in the best interests to the companies (Klein 2002b; NYSE Rule §303.01[B][3][b]; NASDAQ Rule 4310[c][26][B][ii]). Due to this flexibility in audit committee composition, it was common before 1999 that many audit committees did not have fully independent outside directors (Klein 1998, 2002b). This creates a setting in which audit committee members’ equity holdings changed not only because of the equity-based compensation they received, but also because of inside-affiliated directors’ own shares before they became audit committee members. Under this setting, an examination of whether equity-based compensation jeopardizes audit committee independence may be trivial.

In contrast, Section 301 of SOX requires that audit committees to be composed entirely of independent directors. This provision changes audit committee independence research in three substantial ways. First, it

creates a cleaner setting in which audit committee members’ equity ownership exists purely from the equity-based compensation. Second, the issue of whether earnings management is associated with the majority or 100 percent of the independent members is no longer important. Finally, even though SOX imposes stringent rules on audit committees’ composition, expertise, and duties, it is silent in how audit committees should be compensated to maintain their independence over time. Therefore, an investigation of the association between equity-based compensation and audit committee independence becomes important.

Three recent studies examine this issue and provide some new evidence. Archambeault et al. (2008) use pre-SOX data to test whether options for audit committees are associated with restatement likelihood. They find a positive relation for both short-term and long-term options because short-term options induce audit committee members’ short-term orientation that possibly undermines their independence, while long-term options are highly uncertain that the payoffs may be too small to induce audit committee members to monitor financial reporting effectively. Differently, Cullinan et al. (2010) use post-SOX data and shows that firms paying options to their audit committees are more likely to report internal control weaknesses. In contrast, Engel et al. (2010) use both pre- and post-SOX data and finds that firms are more likely to structure audit committee compensation toward a fixed pay when there is a high demand for monitoring. Specifically, the level of audit committee compensation increases after SOX.

Our study differs from these recent studies in four aspects. First, while Archambeault et al. (2008) and Cullinan et al. (2010) investigate whether options are associated with restatement likelihood and the incidence of internal control weaknesses, they do not consider both stocks and options to examine whether such compensation is associated with firms’ risk of incurring material misstatements due to fraud. This issue is important because regulators will know how to govern audit committees’ compensation practice only when academic research provides evidence about the association between equity-based compensation and firms’ fraud risk. In addition, since restatements may be initiated by external parties (Palmrose et al. 2004) and internal control weaknesses may result from audit committees’ effective oversight, Archambeault et al.’s (2008) and Cullinan et al.’s (2010) empirical results may not provide strong evidence that option compensation is really associated with audit committees’ oversight failure. Third, existing literature using the post-SOX data has lagged in examining the association between audit committees’ equity-based compensation and firms’ potential risk of

fraudulent financial reporting. This issue is important to the regulators because Cohen et al.’s (2008) finding that the level of post-SOX earnings management has returned to the pre-SOX level may be the result of audit committees’ oversight failure due to firms’ increased use of stocks and options to compensate their audit committees after SOX. This issue is also important to the shareholders because equity-based compensation may impair audit committees’ ability to detect the change of managers’ earnings management techniques from pre-SOX accruals-based method to post-SOX manipulation of real operating activities (as have been found in Cohen et al. (2008), Cohen and Zarowin (2010), and Eldenburg et al. (2011)).

Finally, Naiker and Sharma (2009) examine whether former auditor affiliation impairs audit committee independence, and find that such affiliation contributes to more effective monitoring of internal controls.

Different from this study, we show that equity-based compensation may potentially threat audit committee’s independence, giving rise to higher fraud risk.

2.2 Efficacy of the Clawback Provisions:

Only few studies have examined different aspects of the clawbacks. Chan et al. (2012a) report that restatement likelihood declines after companies voluntarily adopt their clawbacks. Also, investors and auditors regard the adoption of clawbacks as signal of increased financial reporting quality. Chan et al. (2012b) further show that, while firms adopting clawbacks reduce their accrual management, they tend to switch to real transaction management due to pressure to meet or beat earnings benchmarks. Similar to Chan et al. (2012a), Dehaan et al. (2012) find significant improvements in both actual and perceived financial reporting quality (proxied by restatement likelihood and ERC, respectively) following clawback adoption. In another study, Addy et al. (2011) indicate that companies adopt clawbacks to demonstrate vigilance and oversight. Also, these companies have smaller accruals and have directors sitting on other boards of other companies who have adopted clawbacks. Finally, Chan et al. (2013) report that, because clawbacks enhance financial reporting quality, the information uncertainty facing the banks reduces. Therefore, banks use more financial covenants and performance pricing provisions in the loan contracts and decrease interest rates after firms initiate clawbacks.

Importantly, banks increase loan maturity and require less loan collateral subsequent to clawback adoption.

Different from the above studies, our study reports that, given the overwhelming practice of U.S. public companies’ use of stocks and options to compensate their audit committee members, clawbacks may give rise to

unintended consequences in which the negative effect of equity-based compensation on audit committees’

independence is mitigated. Importantly, we test and show that the adoption of clawbacks is more effective than employing more capable managers in reducing firms’ fraud risk. To securities regulators and policy-makers, these consequences are much more important and desirable than Chan et al.’s (2012b) finding that clawbacks induce managers who suffer high pressure to meet or beat earnings benchmarks because effective audit committees shall restrain managers from doing so.

3. RESEARCH DESIGN