Does increased board independence reduce earnings management? Evidence from recent regulatory reforms

While some studies (e.g., Beasley 1996; Klein 2002; Bédard et al. 2004) document a negative association between earnings management and the independence of board, the audit committee, or both, other studies (e.g., Vafeas 2005; Larcker et al. 2007) provide conflicting results. See Sect. 2.2 for detailed discussions.

See Larcker and Richardson (2004) for detailed discussions of the latent analysis.

We focus on board independence in the main analyses and examine audit committee independence in the additional analyses for the following reasons. First, while the audit committee directly oversees financial reporting, it reports to the board, which oversees all important decisions. Second, as discussed in detail in Sect. 2.1, regulatory reforms on audit committee independence spanned a longer period, making the difference-in-differences research design less powerful.

While regulatory reforms are not exogenous in the sense that they are responses to corporate scandals, the regulatory requirements are largely exogenous to individual firms.

Given that we find that increased board independence reduces earnings management for firms with better information environment, it is natural to ask why these firms do not choose to increase board independence before the reforms. There are several possibilities. First, managers may be entrenched and reluctant to do so. The market force may not be powerful enough to lead to the desired change in board structure due to free rider and information asymmetry problems (e.g., Adams and Ferreira 2007; Duchin et al. 2010). Second, even if managers are not entrenched, there are transaction costs associated with changing board structure, including costs of board selection, contract costs with directors, and legal costs (Coles et al. 2008). The transaction costs can explain firms’ deviation from optimal board structure for a prolonged period.

For example, in 1987 the NYSE required that the listed firms have an audit committee consisting solely of independent directors, and in 1989 the NASD required that the listed firms have an audit committee with the majority of the members being independent. Despite these rules, the definition of independent directors is not entirely clear, and firms often have affiliated directors sitting on audit committees (Vicknair et al. 1993).

Romano (2005) surveys the literature on the impact of audit committee independence on earnings management and finds that, while six studies find a negative association, 10 others fail to find a significant association.

Guthrie et al. (2012) find that the increase in board independence is not associated with a reduction in excessive executive compensation. While Chhaochharia and Grinstein (2009) document a reduction in excessive executive compensation for noncompliant firms after the reform, Guthrie et al. (2012) find that their results are driven by a few outliers. Our results are robust to outliers. See Sect. 4 for detail.

This is consistent with our private conversations with several directors, who indicate that they periodically obtain information from sources other than the management before board meetings. DeFranco et al. (2010) also argue that outside directors are more informed and are better monitors when firms make more public disclosures.

BoardEx covers S&P 1500 firms as well as smaller firms. The most important advantage of using BoardEx is to increase the generalizability of the results. Since the board data for smaller firms in BoardEx is somewhat limited for the period 2000–2002, we hand collect the information on board structure from proxy statements for the period 2000–2002 if such information is missing in BoardEx.

If more than 50 % of a firm’s voting power is held by an individual, a group of individuals who agree to vote together, or another firm, the firm is classified as a controlled firm. We use the proxy statements to identify the controlled firms.

According to the BoardEx, affiliated directors include those who are former employees of the firm, providers of professional services to the firm, customers or suppliers of the firm, or family members of an employee.

Noncompliant firms can increase board independence by hiring new independent directors, replacing old affiliated or insider directors, or both. Overall, our sample firms do not experience a significant change in board size, with a mean (median) board size of 8.51 (8.00) in 2000 and 8.58 (8.00) in 2005 (untabulated).

Based on our reading of firms’ proxy statements, the primary reason why some firms did not comply with the requirement by 2005 based on the BoardEx’s definition is the difference in the definition of independent directors between the firm and BoardEx. BoardEx’s definition tends to be stricter. For example, while the regulatory definition of past employees refers to employees within the last 3 years, BoardEx tends to regard directors with any employment history with the company as past employees. In some cases, the firm applies the rule loosely. For example, some firms classify directors with business relationships with the firm as independent directors, arguing that these relationships do not interfere with such directors’ independence. BoardEx classifies all such directors as affiliated.

Note that we are interested in increases in board independence in general. We use the noncompliance indicator as the instrument for increases in board independence. We are not implying that 50 % is the critical threshold for effective board monitoring.

Industries are defined as in Fama and French (1997). We note that industry composition is similar between compliant and noncompliant firms.

We formally compare the change in firm characteristics between the compliant and noncompliant firms (untabulated). We find that these two groups do not differ except that noncompliant firms experience a larger increase in institutional ownership than compliant firms. This finding helps alleviate the concern that systematic differences between the two types of firms may drive our results. Note that we control for the change in firm characteristics and industry fixed effects in the regressions to capture any remaining differences between the two groups.

One would expect the results to be stronger after excluding such firms. We find that the magnitude of the coefficient on the interaction term (Non_Compliance × I_Score/∆Board_Ind × I_Score) indeed becomes larger; however, the difference is too small to be statistically significant. Specifically, the coefficient on Non_Compliance × I_Score is −0.010 (Appendix 3, Panel A, Column (6)) vs. −0.009 (Table 4, Column (1)), and the coefficient on ∆Board_Ind × I_Score is −0.045 (Appendix 3, Panel A, Column (7)) vs. −0.041 (Table 4, Column (2)).

Another commonly used proxy is restatements. However, this proxy is inappropriate in our setting for the following reasons. First, the nature of restatements differs between the post-SOX and pre-SOX period. Recent studies (Plumlee and Yohn 2008; Scholz 2008; DeFond 2010) argue that post-SOX restatements often result from confusion over the interpretation of new accounting standards and have little impact on firm value. Second, we find that firms with restatements before SOX, particularly those with accounting irregularities as classified in Hennes et al. (2008), are unlikely to have restatements afterward. Thus one cannot practically use the change specification to study the impact of the change in board independence on the change in restatements. Lastly, prior research (e.g., Dyck et al. 2010) finds that analyst coverage can increase the likelihood of accounting frauds being detected. One can make similar arguments for board independence. Thus an analysis of the impact of increased board independence on restatements can be confounded by its impact on the detection of misstatements.

We do not use the abnormal level of operating cash flows because both Roychowdhury (2006) and Zang (2012) argue that real earnings management activities have an ambiguous effect on operating cash flows.

We also replicate the analyses in Table 3 using Non_Compliance_Audit and ∆Audit_Ind. We find that both variables are insignificant in explaining the change in |DA|; we do not tabulate the results to save space.

Opportunistic earnings management can lead to poor future performance for two reasons. First, it can disguise managers’ rent extraction and waste of firm resources. Second, it can lead to less informative earnings, reducing contracting efficiency and investment efficiency. For example, Beyer et al. (2014) show analytically that, when the cost of earnings management is higher (and the extent of earnings management is lower), the board can better motivate executives to work for the interest of shareholders, leading to better firm performance.

Note that this result appears to contrast with the latent analysis (Sect. 5.1). For the latent analysis, we find that, while the newly appointed directors are generally similar between the two clusters, they are more experienced for cluster I, where we find a negative association between increases in board independence and decreases in earnings management (Table 5, Panel B). This inconsistency is likely due to the different research methods used.