Material weakness disclosures and restatement announcements: The joint and order effects

• Author: Li He,Bharat Sarath,Nader Wans
• Date: 01 January 2019
• DOI: http://doi.org/10.1111/jbfa.12363
• Published date: 01 January 2019

DOI: 10.1111/jbfa.12363

Material weakness disclosures and restatement

announcements: The joint and order effects

Li He1Bharat Sarath2Nader Wans3

1SouthwesternUniversity of Finance and Eco-

nomics,555 Liutai Avenue, Chengdu, P.R.China

611130

2RutgersBusiness School, 1 Washington Park,

Newark,NJ 07102, USA

3Facultyof Business Administration, Memorial

Universityof Newfoundland, St. John's, NL A1B

3X5,Canada

Correspondence

BharatSarath, Rutgers Business School, 1 Wash-

ingtonPark, Newark, NJ 07102, USA.

Email:bsarath@andromeda.rutgers.edu

JELClassification: M40, G14, K22

Abstract

We examine differences in stock price, option volatility, and litiga-

tion reactions to restatement announcements that are associated

with a material weakness (MW) disclosure. Contrasted with restate-

ments that are not associated with any MW disclosure, our analy-

ses reveal that firms that announce both a restatement and an asso-

ciated MW experience significantly more negative market returns,

greater implied volatility, and higher likelihood of class action law-

suits. Separating the restatements into timely reporters, where the

MW precedes the restatement, and non-timely reporters, where the

MW is concurrent with or follows the restatement, we find that

timely reporters experience more negative returns at the time of

the restatement, relative to non-timely reporters, suggesting that

investors perceive the early MW disclosure to signal more perva-

sive control-related problems. Interestingly, we find that timelyand

non-timely reporters are equally likely to be sued, consistent with

theargument that wrongdoing (through either a timely or non-timely

MW disclosure) provides stronger grounds for establishing scien-

ter. However, timely reporters appear to secure more favorable lit-

igation outcomes: they face higher likelihood of lawsuit dismissals

and pay much lower settlements, compared to non-timely reporters.

Overall, our evidence provides new insights into how market par-

ticipants incorporateinformation about internal control weaknesses

into their perceptions regarding the economic implications of finan-

cial restatements, and financial reporting quality.

KEYWORDS

class actions, financial reporting quality, market reaction, material

weakness disclosure, restatements

1INTRODUCTION

Sections 302 and 404 of the Sarbanes-Oxley Act (hereafter, SOX 302; SOX 404) include keyprovisions that require

both managers and auditors to periodically evaluate the quality of internal controls overfinancial reporting (ICFR) and

to report any material weakness (MW) that is detected. Therefore, MW disclosures provide early warning to investors

68 c

2018 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/jbfa JBus Fin Acc. 2019;46:68–104.

HE ET AL.69

that internal controls are not designed or operating in a manner to prevent or detect material misstatements aris-

ing in the financial statements (PCAOB, 2004). Prior research has examined the capital marketconsequences of MW

disclosures and restatement announcements in isolation, finding consistently that these disclosures lead to a nega-

tive stock price reaction (Hammersley,Myers, & Shakespeare, 2008; Palmrose, Richardson, & Scholz, 2004). We con-

tribute to extantresearch by examining how markets react to restatement announcements that are associated with an

MW disclosure made contemporaneously with, before, or after the restatement. We argue that a restatement with an

associated MW disclosure would provide informational value to guide investors to assess the implications of financial

restatements, as compared with a restatement that is not connected to any MW disclosure.

Examining the consequences of restatements and associated MW disclosures in combination is important for sev-

eral reasons. First, very generally, information about an underlying economic eventc ouldelicit different responses

based on whether it is conveyed in a single disclosure or multiple disclosures (Gonedes, 1978).1More specifically,

timely reporting of MWs(i.e., before restatements occur) is important in assessing the quality of financial reporting. For

example, Moody's expresses concerns that late reporting of MWs(i.e., concurrent with, or subsequent to the restate-

ment) could be lagging indicators of reporting quality problems, arguing that late MW reporting undermines the use-

fulness of these reports (Jonas, Rosenberg, & Gale, 2006). Further,the Public Company Accounting Oversight Board

(PCAOB), in its recent concept release on developinga portfolio of quantitative audit quality indicators, lists “the num-

ber of restatements for errors resulting from fraudor other financial reporting misconduct with no previously reported

material weakness in internal control”as a negative indicator of audit quality (PCAOB, 2015, p. A-21), suggesting that

the PCAOB views timely MW reporting as a signal of higher audit quality. Finally,a large number of restatements in

the post-SOX era have been made to correct immaterial misstatements (Burks, 2011), leading to concerns that they

“mayreduce confidence in reporting, and may create confusion that reduces the efficiency of investor analysis” (ACIFR,

2008, p. 78). Assessing restatements in light of the effectiveness of the underlying ICFR can shed light on restatements

that are associated with greater adverse effects. Our study contributes to prior research by analyzing differences in

stock price, option volatility,and litigation reactions to restatement announcements that are associated with MW dis-

closures contrasted with restatements that are not associated with any MW disclosure, and whether the reaction to

MW-associated restatements vary with the timeliness of the MW disclosure.

Auditing Standard No. 5 characterizes MWs as deficiencies “such that there is a reasonable possibility that a mate-

rial misstatement of the company'sinterim or annual financialstatements will not be prevented or detected on a timely

basis.”This argument is consistent with both prior studies that firms with ineffective control procedures are more likely

to have material misstatements (e.g., Nagy, 2010), and the PCAOBargument that restatements imply that MWs are

likely to exist (PCAOB, 2007).2Hence, ideally,restatements should be preceded by a related MW disclosure (Glass

Lewis & Co., 2007). However,some firms only make these disclosures after they are compelled to restate their finan-

cials; still others do not make these disclosures at all (Besch, 2009; IMA, 2008; Rice & Weber,2012).3,4 The variation in

the timeliness of MW reporting in relation to restatements leads restating firms to fall into one of three groups: (i) firms

1Kaznikand Lev (1995) examine the effect of a voluntary disclosure that may precede a negative earnings shock that will be reported in a 10-Q.While an MW

isnot avoluntary disclosure, the fact that it is expected to precede a restatement leads to some economic similarities between our setting and that of Kaznik

andLev (1995). A recent survey of the many studies concerning the use of single or multiple signals regarding some underlying news may be found in Connelly,

Certo,Ireland, and Reutzel (2010).

2The argument that restatements have typically arisen from weak internal controls has been documented in the literature. Kinney and McDaniel (1989)

arguethat revision of previously issued financial statements implies a breach in internal control systems. Turner and Weirich (2006, p. 17) argue that “If these

companies’internal controls really had no weaknesses, why did the companies have errors that required restatements?”

3Forexample, Glass Lewis & Co. (2007) documents that in 2005, 57% of the MW disclosures were disclosed subsequent to restatement filings and were more

ofa “formality” than “a warning flag”.

4There is always the possibility that some restatements do not relate to MWs. First, firms maycorrect minor errors that do not necessitate classifying the

relatedICFR weakness as material. For example, out-of-period adjustments are error corrections made through a onetime charge in the current period, and do

notrequire a restatement (Choudhary, Merkley,& Schipper, 2016). Second, the PCAOB allows for “extenuatingor unique circumstances” where a misstatement

doesnot necessitate the existence of an MW (PCAOB, 2004, para. E97)—for instance, if it “reflected the SEC's subsequent view of an accounting matter, when

the auditor concluded that management had reasonable support for its original position” (PCAOB, 2004, para. E99). Finally,there are restatements that are

technical,where the restatement does not imply that a misstatement existed in the original filing; Audit Analytics excludes both out-of-period adjustments and

technicalrestatements from their restatement database and thus are not included in our sample.

70 HE ET AL.

that disclose MWs before restatements (timely MW reporters); (ii) firms that disclose MWs at the time of,or follow-

ing, restatements (non-timely MW reporters); and (iii) firms that do not disclose an MW either before or after restate-

ments (non-MW reporters). Our aim is to examine (i) how market participants’ reactions to restatements vary across

MW reporters and non-MW reporters (i.e., the joint effect: contrasting timely and non-timely reporters against non-

MW reporters), and (ii) within the group of MW reporters, whether timeliness has a significant impact (i.e., the order

effect: contrasting timely against non-timely reporters). Wematch restatement and MW disclosures by the underlying

accounting issue to ensure that all restatements in our sample are related to, or havearisen from, an MW.

Our first set of hypotheses compares investors’ responses to restatements that accompanyan MW disclosure (the

joint MW/restatements consisting of timely and non-timely reporters) to restatements not associated with an MW

disclosure (non-MW reporters). Restatements by non-MW reporters may be perceived as isolated events that do not

result from weak internal control system, and thus might not trigger a strong negative reaction. This prediction is

consistent with regulators’ concerns over the large number of restatements made by public firms as a result of the

increased scrutiny under SOX and which might not necessarily reflect bad news (Burks, 2011).5In contrast, a restate-

ment that is accompanied by an MW disclosure potentially implies that the underlying deficiencies in the firm'sICFR

may be systemic and, consequently, may lead to stronger negative reactions as the restatement could be one mani-

festation of several other problems yet to be uncovered.6Our first testthus investigates whether investors perceive

the joint MW/restatementdisclosure (a restatement preceded by, disclosed concurrently with, or followed by an MW)

differently from the single disclosure of a restatement (we call this test the “joint effect” test).

Our second set of hypotheses focuses on the timeliness of MW Reporting. Prior research documents that MW dis-

closures serve as early warnings that internal controls are not designed or operating in a manner to prevent or detect

financial misstatements, and they are thus perceived to have informational value to investors (Hammersley et al.,

2008). Restatements also lead to adverse market consequences as investors perceive the revisions of prior financial

statements as a sign of poor financial reporting quality (e.g.,Palmrose et al., 2004). Understanding how investors incor-

porate MW disclosures into their interpretation of subsequent restatements is important since internal control pro-

visions were mainly promulgated as a mechanism to communicate to investors the quality of financial reporting on a

timely basis.7Tothe extent that MW warnings raise a red flag about potential future misstatements, investors would

react more negatively to restatements that were not preceded by an MW than to those that were forewarned through

an MW disclosure.8An alternativetheory still posits that an MW that is readily apparent or is more serious will be iden-

tified and disclosed promptly,whereas less serious ones may not be detected or delayed. From this perspective, timely

disclosures would indicate more pervasive internal control issues, which in turn, signal more adverse future outcomes,

compared to non-timely reporters. Additionally,auditors are expected to adjust their audit procedures in response to

the elevated control risk when MWsare detected; therefore, a restatement following a timely MW indicates that audi-

tors did not modify their substantive tests adequately to reduce detection risk, which can then raise users’ concerns

about audit quality. Our second set of predictions examines whether a prior MW disclosure augments or dampens

reactions at the time of the restatement relative to delayedMW disclosures (we call this test the “order effect” test).

5Regulatorsargue that many restatements in the post-SOX era are made to correct immaterial errors. Further,the Securities and Exchange Commission'sAdvi-

soryCommittee on Improvements to Financial Reporting (ACIFR) emphasized the need to “reduce the number of restatements that do not provide important

informationto investors,” noting that these restatements may create confusion among investors (ACIFR, 2008, p. 78).

6Although a current MW disclosure may,by itself, indicate the potential for future problems, we argue that an MW alone does not necessarily imply impair-

mentof future financial reporting quality, as firms usually (disclose that they) initiate remediation efforts to fix detected MWs. Therefore, if managers’ remedi-

ationefforts are successful, investors should not expect restatements to arise in the future. On the other hand, a subsequent restatement would indicate that

theMWs have not been remediated, and therefore indicate that more problems are yet to be uncovered.

7The internal control provisionsof SOX require managers and auditors to report MWs in the periods “in which they exist.” In this regard, the PCAOB argues

that “for the implementation of Section 404 of the Act to achieveits objectives, the public must have confidence that all material weaknesses that exist as of

thecompany's year-end will be publicly reported” (PCAOB, 2004, para. E94). In Moody's words, “creditors would benefit were management, audit committees

and auditors to devote more attention to controls that preventand detect fraudulent reporting, and to not wait for disasters to occur before flagging these

criticalcontrol problems” (Jonas et al., 2006, p. 5).

8Given Moody's argument that late MW reporting undermines the usefulness of ICFR reporting and the PCAOB view that timely reporting is indicative of

higheraudit quality compared to non-timely reporting, MW disclosures should convey potentially adverse information ahead of the restatement and provoke

apartial reaction in advance of the actual restatement.