THE ACCOUNTING REVIEW
天使之翼合唱团
V ol.83,No.1
伯努利家族2008
pp.217–250
The Effect of SOX Internal Control Deficiencies and Their Remediation on
Accrual Quality
Hollis Ashbaugh-Skaife
十三经University of Wisconsin–Madison
Daniel W.Collins
The University of Iowa
William R.Kinney,Jr.
The University of Texas at Austin
列宁的故事Ryan LaFond
Massachutts Institute of Technology and Algert Coldiron Investors
ABSTRACT:This paper investigates the effect of internal control deficiencies and their
remediation on accrual quality.Wefirst document thatfirms reporting internal control
deficiencies have lower quality accruals as measured by accrual noi and absolute
abnormal accruals relative tofirms not reporting internal control problems.Second,we
find thatfirms that report internal control deficiencies have significantly larger positive
and larger negative abnormal accruals relative to controlfirms.Thisfinding suggests
internal control weakness are more likely to lead to unintentional errors that add noi
to accruals than intentional misstatements that bias earnings upward.Third,we doc-
ument thatfirms who auditors confirm remediation of previously reported internal
control deficiencies exhibit an increa in accrual quality relative tofirms that do not
remediate their control problems.Finally,wefindfirms that receive different internal
control audit opinions in successive years exhibit changes in accrual quality consistent
with changes in internal control quality.Collectively,our cross-ctional and intertem-
poral change tests provide strong evidence that the quality of internal control affects
the quality of accruals.
Keywords:SOX;internal control;accrual quality;financial reporting reliability.
We thank minar participants at Emory University,University of Houston Accounting Rearch Conference,Penn State Rearch Conference,University of Technology–Sydney,Washington University in St.Louis,and the Uni-versity of Wisconsin–Madison and two anonymous reviewers for helpful comments on previous versions of this paper.
Editor’s note:This paper was accepted by Dan Dhaliwal.
Submitted June2006
Accepted August2007
217
218Ashbaugh-Skaife,Collins,Kinney,and LaFond
The Accounting Review,January 2008
I.INTRODUCTION
T he reliability of financial reporting is claimed to be a function of the effectiveness of a firm’s internal control (PCAOB 2004;Donaldson 2005).1However,rearch to date provides mixed evidence on whether internal control weakness adverly af-fect accrual quality,an important component of reliable financial statements (Hogan and Wilkins 2005;Bedard 2006;Doyle et al.2007).This paper us recently available data on the effectiveness of firms’internal controls mandated by the Sarbanes-Oxley Act (SOX)and the PCAOB’s Auditing Standards (AS)No.2to investigate how internal
喀什海拔control quality affects the reliability of financial information (U.S.Congress 2002;PCAOB 2004).We posit that if a firm has weak internal control,managers are less able to determine reliable accrual amounts,and a conquence of the unintentional misreprentations is that finan-cial information is more noisy and less reliable.In addition,managers of firms with weak internal control can more readily override the controls and intentionally prepare biad accrual estimates that facilitate meeting their opportunistic financial reporting objectives.Thus,regardless of whether misstatements are unintentional or intentional,the quality of accruals is likely diminished when firms have weak internal controls.
We u SOX-mandated internal control deficiency (ICD)disclosures and external au-ditor opinions on internal control to conduct both cross-ctional and within-firm intertem-poral change tests of whether the effectiveness of firms’internal control affect the quality of reported accruals.Specifically,we test whether:(1)firms that have ICDs exhibit noisier accruals and larger abnormal accruals (both absolute and signed)relative to firms not dis-closing ICDs;(2)firms who external auditors verify remediation of previously disclod ICDs subquently exhibit higher quality accruals relative to firms failing to remediate their control problems;and (3)firms that receive different SOX 404audit opinions in successive years exhibit improvements (declines)in accrual quality consistent with strengthened (weakened)internal controls indicated by the different opinions.
To conduct our tests,we identify a sample of firms that have at least one SOX 404audit opinion on internal control (i.e.,accelerated filers).We look back to determine whether a sample firm disclod an ICD under SOX 302and look forward to e if the firm received a cond SOX 404opinion.2Managements’prior internal control disclosures in combination with external audit reports allow us to conduct change analysis tests to provide strong evidence on whether weak internal control diminishes the quality of firms’reported accruals.
In cross-ction tests,we find that ICD firms have larger absolute,larger positive,and larger negative abnormal total and abnormal working capital accruals relative to non-ICD 1
In Auditing Standards No.2,the Public Company Accounting Oversight Board (PCAOB 2004)states:‘‘The Board believes that effective controls provide the foundation for reliable financial reporting.Congress believed this too,which is why the new reporting by management and the auditor on the effectiveness of internal control over financial reporting received such prominent attention in the [SOX]Act.Internal control over finan-cial reporting enhances a company’s ability to produce fair and complete financial reports.Without reliable financial reports,making good judgments and decisions about a company becomes very difficult for anyone,including the board of directors,management,employees,investors,lenders,customers,and regulators.The auditor’s reportin
g on management’s asssment of the effectiveness of internal control over financial reporting provides urs of that report with important assurance about the reliability of the company’s financial reporting ’’(para.E.5.,Appendix E,Background and Basis for Conclusions)(emphasis added).2Section 302of SOX,which became effective in August 2002,requires nior management to evaluate internal controls over financial reporting and certify the effectiveness of the controls (e Ashbaugh-Skaife et al.[2007]for a discussion of the SOX 302reporting requirements).Section 404of SOX requires that public company annual filings contain management’s asssment of the design and operating effectiveness of its internal control and that the external auditor provide an opinion on management’s asssment (Securities and Exchange Com-mission [SEC]2002).Auditing Standards No.2(AS No.2)requires that the auditor form and express a parate opinion on the auditor’s evaluation of internal control.We refer to the auditor’s own opinion as the ‘‘SOX 404opinion.’’
The Effect of SOX Internal Control Deficiencies and Their Remediation on Accrual Quality219firms.In addition,wefind that accruals of ICDfirms map less reliably to past,current,and future cashflows.All of the results are consistent with the accruals of ICDfirms being noisier and less reliable relative to accruals offirms with no reported internal control prob-lems.The results hold after controlling for conditional conrvatism andfirm character-istics that prior rearch has shown to be related to accr
ual quality.As to intertemporal changes in the effectiveness of internal controls,wefind thatfirms that disclo an internal control problem and subquently receive an unqualified SOX404audit opinion exhibit a decrea in absolute abnormal accruals relative to the year when internal control problems arefirst reported.Wefind no evidence that this decrea is due tofirms becoming more conrvative in their accounting choices.In contrast,firms identified as having internal control problems that subquently receive an adver SOX404opinion exhibit no signif-icant change in the magnitude of abnormal accruals.
Our analysis of successive year SOX404opinions reveals thatfirms who internal controls improve(going from adver to unqualified SOX404opinions)exhibit a modest increa in accrual quality,whilefirms who internal controls worn(going from un-qualified to adver SOX404opinions)exhibit a significant decrea in accrual quality. Firms with the same SOX404opinion in successive years(unqualified or adver in both years)exhibit no change in accrual quality.
Collectively,our cross-ctional and intertemporal change tests are consistent with the notion expresd by Congress and regulators that strong internal controls provide a signif-icant long-term benefit in improving the accuracy offinancial reporting that leads to higher quality information forfirms
’external stakeholders(Donaldson2005;U.S.Hou of Rep-rentatives2005;U.S.Senate2002,2004).
Our study makes veral contributions.First,we add to the literature that eks to identify the determinants of accrual quality,much of which focus on innatefirm operating characteristics as the primary determinants of accrual quality(Dechow and Dichev2002; Dechow et al.1995;Francis et al.2004;Kothari et al.2005)or intentional misu of accounting discretion to manipulate earnings through accruals(Dechow et al.1996).Our study is one of thefirst to investigate how internal control,which is intended to attenuate both intentional and unintentional misstatements,affects the quality of accruals and thereby the reliability offinancial statements.We empirically link the strength offirms’internal control overfinancial reporting to the quality of accruals through the u of SOX302and SOX404reporting.Most importantly,becau independent auditors’SOX404opinions provide unambiguous signals about changes in the effectiveness offirms’internal controls, our rearch design allows tests of the effects of changes in internal control quality on accrual quality in ways that minimize competing explanations for our results.
This study also contributes to the literature designed to asss thefinancial reporting conquences of SOX.While much of the prior literature address the costs of imple-menting and auditing SOX requirements(Li et al.2008;Zhang2007;Berger et al.2005), our study is among thefirst to provide evid
ence about potential benefits of strong internal control in terms of the quality of externally reportedfinancial information.
Our rearch is most cloly related to concurrent studies by Hogan and Wilkins(2005), Bedard(2006),and Doyle et al.(2007)that examine the relation between internal control weakness and earnings(accrual)quality.Hogan and Wilkins(2005)find no difference in the Dechow and Dichev(2002)measure of accrual noi between ICDfirms and a t of controlfirms and only modest differences in absolute abnormal total accruals using the modified Jones model(Dechow et al.1995).Bedard(2006)finds no difference in the absolute value of unexpected total accruals between ICD and controlfirms in the two years prior to the ICDfirms’disclosures and in the two years that follow disclosure.However,
The Accounting Review,January2008
220Ashbaugh-Skaife,Collins,Kinney,and LaFond The Accounting Review,January 2008
Bedard (2006)finds that ICD firms exhibit larger absolute abnormal accruals relative to control firms in the disclosure year.Doyle et al.(2007)find that firms that disclod ICDs under SOX 302(before required internal control audits)exhibit lower accrual quality rel-ative to control firms,but the differe
nces relate only to overall company level control problems and not to account specific weakness.Interestingly,Doyle et al.(2007)find no differences in accrual quality between ICD and non-ICD firms for internal control weak-ness disclod during the SOX 404regime.
产妇能吃海参吗Our study extends the prior studies in veral important ways.First,all three of the prior studies are restricted to cross-ctional tests of differences in accrual quality between ICD and non-ICD firms.Accordingly,as in all cross-ctional designs,there are potential concerns of endogeneity,lf-lection,and correlated omitted variables,which limit the ability to draw strong causal inferences from the results.3We address the concerns by focusing on firms that received one or more SOX 404opinions and identify which firms have continuously good,continuously bad,or changes in the quality of their internal control.
Incorporating SOX 404opinions is a central feature of our rearch design and is important becau:(1)it provides an unambiguous signal from an independent third-party about the effectiveness of a firm’s internal control;(2)it allows identification of which firms were able to remediate internal control weakness and when the improvement took place;and (3)it allows identification of subts of firms that experienced significant changes in internal controls from year to year.In contrast to the prior studies,we u the quence of ICD disclosures and SOX 404opinions to conduct within-firm and ac
显卡的作用是什么ross-firm testing of the effects of changes in the effectiveness of internal controls over time in ways that minimize competing explanations for obrved differences in the quality of accounting accruals.By linking changes in the strength or effectiveness of internal controls with changes in measures of accrual quality,we mitigate concerns about lf-lection,correlated omitted variables,and endogeneity that cloud the interpretation of results from studies that rely solely on cross-ctional tests,and we provide stronger tests of causal linkages between internal control weakness and the quality of external accounting numbers.
The remainder of our paper is organized as follows.Section II provides institutional background and provides examples of how internal control weakness are likely to intro-duce noi into accruals.Section III describes the accrual quality measures ud in our analysis,summarizes the determinants of accrual quality,and ts forth our predictions on the effects of internal control weakness and remediation on accrual quality.Section IV describes our sample and provides descriptive statistics.Section V prents our empirical findings and Section VI concludes.
II.INSTITUTIONAL BACKGROUND AND EXAMPLES
OF CONTROL WEAKNESSES
Auditing Standards No.2(AS No.2,PCAOB 2004,para.7)defines internal control over financial reporting (ICFR)as:
韩非子作品a process designed by ...the company’s principal executive and principal financial officers ...and effected by the company’s board of directors,management,and other personnel,to provide reasonable assurance regarding the reliability of financial report-ing and the preparation of financial statements for external purpos in accordance with generally accepted accounting principles.
3Indeed,Doyle et al.(2007,30),recognize this limitation and state:‘‘[O]ur ability to infer causality between internal control problems and accrual quality is limited.’’
The Effect of SOX Internal Control Deficiencies and Their Remediation on Accrual Quality 221The Accounting Review,January 2008An ICD exists ‘‘when the design or operation of a control does not allow management or its employees,in the normal cour of performing their assigned functions,to prevent or detect misstatements on a timely basis (AS No.2,para.8).Thus,ICFR is focud on timely prevention and detection of financial misstatements,whether unintentional or intentional.4
Internal Control—Integrated Framework (The Framework),the external standard or criteria for evaluating ICFR quality (corresponding to GAAP for financial statements),sug-gests five components
or elements of internal control:the control environment,risk asss-ments,control activities,information and communication,and monitoring (Committee of Sponsoring Organizations of the Treadway Commission [COSO]1992).5A high-quality control environment establishes the importance of internal control throughout the firm.Within such an environment,potential caus of financial misstatements are assd as to risk and control activities are designed and put in place to mitigate important risks,and exceptions that ari are communicated to parties who can take corrective actions.Finally,the entire internal control process is monitored for exceptions by management and the internal audit function.
ICDs can affect the noi and the magnitude of abnormal accruals in financial state-ments in two principal ways.One way is through random,unintentional misstatements due to the lack of adequate policies,training,or diligence by company employees.Examples are:inventory counting and pricing errors that misreport inventory on hand and related cost of sales,omission of items such as failure to record credit purchas,variation in revenue recording due to lack of specific policies (or employee discretion)for revenue recognition,expensing amounts that should be capitalized and vice versa ,inadequate basis for account-ing estimates such as the allowance of inventory obsolescence,and unreliable procedures for ‘‘rolling up’’amounts from gments and subsidiaries.The random misstatements can lead to increas or decreas in accruals and resulting net income.
A cond way that ICDs can adverly affect the quality of accruals is through inten-tional misreprentations or omissions by employees or by management.The non-random misstatements typically overstate earnings for the current period,but ‘‘big bath’’write-offs or cookie jar rerves result in opportunistic understatement of current earnings as well.For example,management’s exerci of discretion in accounting choices allows financial misreprentation through manipulation of accruals for recording important accounting es-timates such as warranty liabilities,rerves for sales returns,and allowance for uncollec-tible receivables.Furthermore,employee fraud is made possible by inadequate gregation of internal control duties.Weak internal control in the form of inadequate gregation of duties can allow the misappropriation of asts and alteration of recorded amounts by an employee that is not detected becau the company has inadequate staff for monitoring,or lack of action by top management becau of a lax control environment.In addition,mis-statements can be introduced into the financial reporting process through ‘‘lective over-sights or omissions’’in accumulating gment and subsidiary information for consolidated reports,as well as through management emphasizing earnings targets in instructions to employees.
4
There are inherent limitations in ICFR as noted in AS No.2,para.16.The limitations are due to laps in human judgment,compliance,and diligence,plus collusion and improper management override of ICFR.While the inherent limitations cannot be eliminated,AS No.2asrts that it is possible to design into the ICFR process safeguards that will reduce the risk that they will ari.5The control environment and monitoring may be uful in protecting against manipulation of records as a respon to management-prescribed targets bad on financial performance.
