Earnings management and instances of material weaknesses reported under Sections 302 and 404

ABSTRACT

Sections 302 and 404 of the 2002 Sarbanes-Oxley Act require firms to disclose material weaknesses (MWs) in internal controls. We investigate the association of single and multiple instances of MWs with earnings management. Multiple instances are positively associated with earnings management for Section 302 and 404 firms, whereas single instances have no such association for either firm type. Thus, MWs that are not remediated within one quarter or year tend to be positively associated with earnings management and, in particular, with income-decreasing earnings management. Multiple instances of MWs for such firms can therefore provide critical information for auditors and investors.

KEYWORDS:

• Internal control weakness
• material weakness
• earnings management
• Sarbanes-Oxley Act (SOX)
• multiple instances

Disclosure statement

This work was supported by the research fund of Hanyang University (HY-2019-00000002813).

Notes

1. This study focuses on internal control deficiencies, the third part of Section 302.

2. Multiple instances of MWs can be defined as MWs that are not remediated or as more than one MW instance in consecutive periods.

3. Hermanson and Ye (2009) show that only 27 percent of accelerated filers disclosed MWs under Section 302 earlier in the fiscal year than they disclosed MWs under Section 404.

4. Using all firms with no internal control problems in COMPUSTAT, that is, the entire population, avoids the problems of biased parameters and probability estimates (Healy and Palepu 2001).

5. The descriptive statistics and correlation coefficients for multiple and single instances of MWs under Section 404 are similar to those under Section 302. Thus, we do not provide the corresponding table for Section 404 firms here.

6. For all models, Firm_Dummies are included to control firm fixed effects, and the standard errors are clustered at the firm level to avoid overstating t-statistics due to multiple MWs of the same firms in different time following Gow, Omiazabal, and Taylor (2010) and Petersen (2009).

7. For Section 302 firms, firms with multiple instances of internal control weakness include those that report an internal control problem in at least two consecutive quarters in a given year, and firms with a single instance include those that report such a problem in only one quarter. For Section 404 firms, firms with multiple instances are those that report internal control problems in at least two consecutive years, whereas those with single instances include those that report such a problem in only one year.

8. Note that in the models used in Tables 4, 5, and 6, the two indicator variables, MULTIPLE and SINGLE (FIRST and OTHERS) are actually only from MW firms. Since we compare the effect of multiple and single (first and others) MWs on earnings management with the one of control firms, the two indicator variables with three categories can be used simultaneously. In other words, we compare the effect of all MWs on earnings management with the one of control firms in the first column of each table and separate all MWs into multiple and single (first and others) MWs and compare the effects of those on earnings management again with the effect of control firms in the second column of those tables.

9. We are grateful to the anonymous reviewer who pointed out this line of inquiry.