行政院國家科學委員會補助專題研究計畫成果報告
期末報告
重編公司及時評估內部控制之決定因子與後續財務報導改
善
計 畫 類 別 ： 個別型計畫 計 畫 編 號 ： NSC 101-2410-H-004-076- 執 行 期 間 ： 101 年 08 月 01 日至 102 年 07 月 31 日 執 行 單 位 ： 國立政治大學會計學系 計 畫 主 持 人 ： 周玲臺 計畫參與人員： 碩士班研究生-兼任助理人員：李奕萱 博士班研究生-兼任助理人員：邱献良 報 告 附 件 ： 出席國際會議研究心得報告及發表論文 處 理 方 式 ： 1.公開資訊：本計畫涉及專利或其他智慧財產權，2 年後可公開查詢 2.「本研究」是否已有嚴重損及公共利益之發現：否 3.「本報告」是否建議提供政府單位施政參考：否中 華 民 國 102 年 12 月 20 日
中 文 摘 要 ： (1)本研究旨在研究沙賓法 404 條及審計準則第 5 號對於內部 控制揭露錯誤的影響。本研究將重編公司視為內部控制較不 健全的公司，這些公司理應較可能會收到 ICFR 否定意見，反 之亦然。本研究假定若公司報導公司內部控制系統為有效但 後續卻發生重編，此時意味犯型二錯誤的機率較高，若公司 報導公司內部控制系統為無效但後續卻並未發生重編，此時 意味犯型一錯誤的機率較高。型二及型一錯誤皆代表內部控 制品質降低。本研究發現實際沙賓法 404 條後可以使型二錯 誤降低，並不會使型一錯誤增加。再者本研究也發現較有彈 性的審計準則第 5 號可以使型一錯誤降低，有助提高內部控 制審計效率。但審計準則第 5 號卻會使得型二錯誤增加。 (2)有學者研究主張公司之財務報導品質主要受到報導誘因影 響，而非會計準則差異所致。本研究採用 Daske et al. （2013）所發展之報導誘因綜合指標，探討其與公司報導及 修正內控缺失之關係。實證結果支持假說一，表示當公司報 導誘因越高，其在 2004~2011 期間越不易發生內部控制重大 缺失；在假說二方面，並未證實當缺失公司的報導誘因越高 時，較不易存在公司層級的重大缺失；假說三以 2004 年至 2009 年間已發生內部控制缺失的公司作為研究對象，結果發 現在 2004~2009 年期間相較於低報導誘因的缺失公司，當缺 失公司的報導誘因越高時，並未顯著地較易於後續年度中對 缺失進行修正。(3)有學者研究主張公司之財務報導品質主要 受到報導誘因影響，而非會計準則差異所致。本研究採用 Daske et al. （2013）所發展之報導誘因綜合指標，探討其 與公司盈餘品質及報導內控缺失之關係。實證結果支持假說 1，當公司報導誘因越高，公司的盈餘品質越佳；假說 2a 之 結果顯示在 2004~2011 期間報導誘因越高的公司越不易發生 內部控制重大缺失；假說 2b 則進一步證實當公司的報導誘因 越高時，較不易存在公司層級的重大缺失。本篇是首篇利用 受 SOX 規範的美國上市公司之公司層級報導誘因來探討盈餘 品質及內部控制缺失的文章。 中文關鍵詞： 沙賓 404 條、審計準則第 5 號、內部控制揭露、重編公司、 內部控制缺失、報導誘因、缺失修正
英 文 摘 要 ： (1)This study is to examine both of the impacts of SOX 404 and AS5 on ICFR disclosure errors. We regard restatement companies as those with weak internal controls and they should be more likely to receive adverse ICFR opinions and vice versa. We presume that deterioration in ICFR disclosure quality is shown by an increase in the likelihood of TypeⅡerrors that an
effective ICFR is concluded by a company which subsequently restates its financial reports, or an increase in the likelihood of TypeⅠerrors that an ineffective ICFR is concluded by a company which does not subsequently restate its financial reports. Our evidence indicates that the enactment of SOX 404 reduces TypeⅡerrors of ICFR disclosures without the side effect of increasing TypeⅠerrors. In addition, we find that while the more flexible and laxer AS5 can enhance the efficiency of ICFR audits by reducing TypeⅠerrors, it inadvertently lowers public ICFR disclosure quality measured by increased Type Ⅱ errors.(2)Prior researchers advocate that financial reporting quality is influenced mainly by firms＇ reporting incentives, but not due to the difference of accounting standards. We follow Daske et al.
(2013) and develop a comprehensive index of reporting incentives to examine its relationship with firms＇ reporting and remediation of their internal control weaknesses. The empirical result supports hypothesis 1, that is, the higher firms reporting incentives, the less likely report internal control weaknesses. We cannot find evidence that weaknesses firms with higher reporting incentives report less company-level weaknesses. We further use weaknesses firms during 2004-2009 to examine hypothesis 3. The results cannot support that weaknesses firms with higher reporting incentives are more likely to remediate weaknesses in two years.(3)Prior researchers advocate that
financial reporting quality is influenced mainly by firms＇ reporting incentives, but not due to the difference of accounting standards. We follow Daske et al.(2013)and develop a comprehensive index of reporting incentives to examine its relationship with firms＇ earnings quality and reporting of internal control weaknesses. We find that higher firm-level reporting incentives are associated with better earnings quality, and may restrain the occurrence of internal control weaknesses, especially company-level material weaknesses.
重編公司及時評估內部控制之決定因
子與後續財務報導改善
(1) Do Both SOX 404 and AS5 Reduce ICFR Disclosure Errors?
(2)公司報導誘因能否抑制與修正內部控制缺失?
(3)
Reporting Incentives, Internal Control Weaknesses Disclosure and Remediation
- 0 -
Abstract
This study is to examine both of the impacts of SOX 404 and AS5 on ICFR
disclosure errors. We regard restatement companies as those with weak internal
controls and they should be more likely to receive adverse ICFR opinions and vice
versa. We presume that deterioration in ICFR disclosure quality is shown by an
increase in the likelihood of TypeⅡerrors that an effective ICFR is concluded by a
company which subsequently restates its financial reports, or an increase in the
likelihood of TypeⅠerrors that an ineffective ICFR is concluded by a company which
does not subsequently restate its financial reports. Our evidence indicates that the
enactment of SOX 404 reduces TypeⅡerrors of ICFR disclosures without the side
effect of increasing TypeⅠerrors. In addition, we find that while the more flexible and
laxer AS5 can enhance the efficiency of ICFR audits by reducing TypeⅠerrors, it
inadvertently lowers public ICFR disclosure quality measured by increased Type Ⅱ
errors.
Keyword: SOX 404, AS5,ICFR disclosure errors, quality of internal control disclosure, internal control audit, restatement
- 1 -
Do Both SOX 404 and AS5 Reduce ICFR
Disclosure Errors?
1. INTRODUCTION
In order to assuage investors’ skepticism after a number of high-profile
accounting scandals, the Sarbanes-Oxley Act (SOX) of 2002 was enacted. One of its
main goals was to improve the reliability of financial information quality by
strengthening issuers’ assessment and reporting process of internal control over
financial reporting (ICFR)1. Section 404 of SOX (hereafter SOX 404) was considered
very controversial because implementation of its requirements, especially internal
control audits, resulted in substantial increase in audit fees (Eldridge and Kealey 2005;
Ettredge et al. 2007; Kinney and Shepardson 2010) and strong complaints by public
companies to the Securities and Exchange Commission (SEC) and Public Company
Accounting Oversight Board (PCAOB) (Johnson 2005).
To mitigate audit costs while still preserving the benefits of internal control
audits, PCAOB replaced Auditing Standard 2 (AS2) (PCAOB 2004) with Auditing
Standard 5 (PCAOB 2007), which adopted a top-down, risk-based approach for
auditors to focus on the most important areas and to eliminate unnecessary testing
procedures. After the implementation of AS5, the number of ineffective ICFR
disclosures declines significantly. SEC and practitioners questioned whether the
decline is due to failure to detect internal control weaknesses under AS5, which
1 Section 302 of SOX, effective on August 29, 2002 requires management to self-evaluate and conclude on the effectiveness of ICFR in periodic reports. Besides, Section 404 of SOX, effective on November 15, 2004, mandates that management prepare ICFR assessment in annual reports (Section 404a) and the assessment be attested by external auditors (Section 404b).
- 2 -
required less process-level testing than AS2 did (SEC 2009a; Whitehouse 2010).
While both of the impacts of SOX 404 and AS5 on audit costs have received attention
in prior studies (e.g., Kinney and Shepardson 2010; Wang and Zhou 2012), our
understanding of the impacts of these regulations on ICFR disclosure quality remains
limited.
The purpose of this study is to provide systematic evidence regarding the impacts
of SOX 404 and AS5 on ICFR disclosure quality. To overcome the difficulty that the
ICFR disclosure quality cannot be either observed or directly gauged, we measure
ICFR disclosure quality by examining the likelihood that ICFR disclosures consist of
reporting errors. We collect the data of accelerated filers after the enactment of SOX
404 and divide the sample into restatement and non-restatement companies. Since
misstatements are indicative of ineffective internal control systems (Kinney and
McDaniel 1989; DeFond and Jiambalvo 1991; McMullen et al. 1996), we regard
restatement companies as those which are more likely to have material weaknesses
and should be more likely to receive adverse ICFR opinions and vice versa. We
presume that deterioration in ICFR disclosure quality is shown by an increase in the
likelihood of TypeⅡerrors that an effective ICFR is concluded by a company which
subsequently restates its financial reports, or an increase in the likelihood of TypeⅠ
errors that an ineffective ICFR is concluded by a company which does not
subsequently restate its financial reports.
First, we focus our investigation on the impacts of SOX 404. In comparison to
SOX 302, additional management documentation and auditor scrutiny over ICFR
required by SOX 404 may help detect and disclose underlying internal control
problems and thus reduce TypeⅡerrors. On the other hand, required signatures of
auditors on opinions regarding ICFR may make auditors unduly cautious in
- 3 -
result in many reported material weaknesses that do not lead to misstatements (Doyle
et al. 2007a), which in turn may increase the likelihood of making TypeⅠerrors. The
results of our study show that in the restatement sample, companies subject to SOX
404 are less likely to conclude their ICFR as effective than companies subject to SOX
302. In the non-restatement sample, we do not find difference in the likelihood of
concluding ICFR as ineffective between companies subject to SOX 404 and 302.
Hence, the enactment of SOX 404 reduces TypeⅡerrors of ICFR disclosures without
the side effect of increasing TypeⅠerrors.
Next we turn our focus to the impacts of AS5 on ICFR disclosure quality. As
mentioned above, PCAOB amended AS2 and proposed AS5 for achieving an optimal
balance between costs and benefits of ICFR audits. By eliminating unnecessary
procedures and testing in audits of internal controls, AS5 stresses to reallocate
resources to really important and high-risk areas (SEC 2007). If it is the case, we
expect that AS5 improves not only the efficiency but also the efficacy of audits of
internal controls and therefore reduces ICFR disclosure errors, including both TypeⅠ
and TypeⅡerrors. On the other hand, there exist concerns over the controversial
risk-based audit approach adopted in AS5 since risk-based audit approaches give
auditors more latitude in their professional judgment. Risk-based audit approaches are
reported to be associated with highly visible audit failures (e.g., HealthSouth and
WorldCom) and are criticized by PCAOB (Bell et al. 2008; Berkowitz and Rampell
2002; Weil 2004). PCAOB also documented that after AS5 15 percent of the 309 audit
engagements inspected by the Board failed to obtain sufficient audit evidence to
support their ICFR audit opinions (PCAOB 2012). Thus, improper application of the
risk-based approach may lead to under-auditing and increase TypeⅡerrors. Whether
public ICFR disclosure errors increase or decrease after AS5 is an empirical issue and
- 4 -
We find that non-restatement companies are less likely to conclude their ICFR as
ineffective after the adoption of AS5. The results manifested by reduced Type I errors
show that AS5 can improve the efficiency of ICFR audits. However, restatement
companies are found to be more likely to conclude their ICFR as effective after the
adoption of AS5. While the more flexible and laxer auditing standard can enhance the
efficiency of ICFR audits, it inadvertently lowers public ICFR disclosure quality
measured by Type Ⅱ errors. The results echo the concerns raised by investors and
regulators about AS52 and the 2010 PCAOB inspection report, which identified
several significant deficiencies in audits of internal controls under AS5 (PCAOB
2012).
This study contributes to the literature in the following ways. First, our study
provides insights into the impact of SOX 404 on ICFR disclosure quality. The
objective of SOX 404 is to provide reliable information so that investors can have
confidence that all internal control weaknesses are publicly reported (PCAOB 2004).
To our best knowledge, the closest antecedents to our paper are Kinney and
Shepardson (2010) and Rice and Webber (2012), both of which investigate the impact
of SOX 404 on control ineffectiveness disclosures. Kinney and Shepardson (2010)
examine material weakness disclosure rates after the enactment of SOX 404 and they
find substantial increases in material weakness disclosure rates for small size
companies undergoing initial internal control audits. While Kinney and Shepardson
(2010) examine the change of control ineffectiveness disclosure rates, our paper goes
one step further by more directly and deeply measuring ICFR disclosure quality with
the respective likelihood of Type I and TypeⅡerrors. Besides, Rice and Webber (2012)
2 Leeway allowed by AS5 could open the door for more lax audits, noted by John Morrissey, operating controller at General Electric and a member of the PCAOB’s Standing Advisory Group. Similarly, Ed Trott, a then member of the Financial Accounting Standards Board wonders if AS5 is to achieve better efficiency by the cost of its effectiveness.
- 5 -
find that only a few restatement companies acknowledge ineffective controls during
their misstatement periods, indicating that ICFR reporting under SOX 404 are
noticeably Type Ⅱ error-prone. Our study differ from Rice and Webber (2012) in
that we examine not only Type Ⅱ but also Type I errors in ICFR reporting.
Additionally, we examine the effect of AS5 on ICFR audit during the SOX 404 era.
Second, prior research regarding AS5 mainly concentrates on its impact on audit
fees (Jiang and Wu, 2009; Doogar et al. 2010; Krishnan et al. 2011; Wang and Zhou
2012). Wang and Zhou (2012) examine whether audit quality has changed following
AS5 adoption and find no evidence of a decrease in audit quality. The main difference
between Wang and Zhou (2012) and this study is that we examine the first-order
effect of AS5, i.e. the effect of AS5 on ICFR disclosure quality and therefore we can
provide direct and clear evidence regarding the impact of AS5 adoption.
Third, since compliance with the requirements of SOX 404 is costly, especially
internal control audits (Hartman 2007; Eldridge and Kealey 2005; Kinney and
Shepardson 2010), whether the burdensome requirements can improve the quality of
public ICFR disclosures is a crucial emiprical research question. Moreover, although
AS5 prescribes a more flexible, top-down risk-based approach, concerns about
auditors’ misuse of their professional judgments regarding risk of material
misstatements led to the promise made by the PCAOB and SEC to monitor the impact
of AS5 (Olson 2007, Cox 2007). Our study sheds light on the impacts of SOX 404
and AS5 on public ICFR disclosures and points out that the enactment of SOX 404
can enhance ICFR disclosure quality by reducing TypeⅡerrors. We also indicate that
AS5, however, deteriorates ICFR disclosure quality by showing increased likelihood
of TypeⅡerrors. The evidence provided in this study can help policymakers and
standard setters formulate future ICFR audit-related rulings.
- 6 -
related literature and our hypotheses. Section 3 describes our research design and
sample selection. Our results are presented in Section 4. Section 5 reports the
sensitivity and additional tests. Finally, Section 6 concludes our research.
2. RELATED LITERATURE AND HYPOTHESIS
DEVELOPMENT
2.1 SOX 404 Regulation and Auditing Standards regarding ICFR
Prior to the passage of SOX, public companies were required to disclose
information about internal controls only when a Form 8-K was filed after an auditor
change. Sections 302 and 404 of SOX are the first statutory legislation that requires
public companies disclose the effectiveness of internal controls. Section 302, effective
on August 29, 2002, requires managers of public companies to evaluate and conclude
on the effectiveness of their internal control systems in periodic reports. Section 404
which became effective on November 15, 2004 contains two subsections. Under
Section 404(a) managers are mandated to prepare ICFR assessment in annual reports.
Under Section 404(b) public accounting firms that audit the issuers’ financial reports
shall also attest to the management internal control assessments. Moreover, AS2, An
Audit of Internal Control over Financial Reporting Performed in Conjunction With an Audit of Financial Statements, was enacted on November 15, 2004 to establish
requirements and provide auditors guidance in performing ICFR audits (PCAOB
2004).
Internal control audits under the directions prescribed in AS2 resulted in strong
complaints from public companies to the SEC and PCAOB about substantial increase
- 7 -
internal control audits, the PCAOB replaced AS2 with AS5, An Audit of Internal
Control over Financial Reporting that is Integrated with an Audit of Financial Statements (PCAOB 2007). AS5 uses a principle-based, fraud awareness focus and
adopts a top-down, risk-based approach, which allows auditors to use their
professional judgment and to reference the work of others, ex. management and
internal auditors. By eliminating unnecessary and detailed procedures required by
AS2, AS5 is less prescriptive and encourages auditors to focus on critical risks of
misstatements and related controls. The aim of AS5 is to alleviate the compliance
burden while still preserving the benefits of ICFR audits.
2.2 The Disclosure Quality of ICFR Reports
ICFR disclosure quality cannot be either observed or directly measured and the
preponderance of prior research regarding disclosure quality focus on the quality of
financial statements. Similar to Knechel and Vanstraelen (2007), we in this study
develop a measurement method of ICFR disclosure quality based on the likelihood of
making TypeⅠor Type Ⅱ errors in ICFR reports. We regard restatement companies
as those in which internal control weaknesses are more likely to exist in the period of
the misstated financial reports. Prior literature has supported the notion that
misstatements are indicative of internal control problems. For example, it is argued
that a breach of a company’s internal control system can be demonstrated by the need
to correct a misstatement (Kinney and McDaniel 1989; DeFond and Jiambalvo 1991;
McMullen et al. 1996). PCAOB (2004) also points out that restatements are high
supportive indicators that material weaknesses in internal control over financial
reporting exist. Even though not all misstatements necessitate the existence of
- 8 -
restatement (non-restatement) companies in this study are viewed as those which are
more (less) likely to have material weaknesses and ought to more (less) probably
receive adverse ICFR opinions. We presume that deterioration in ICFR disclosure
quality is shown by an increase in the likelihood of TypeⅡerrors that an effective
ICFR is concluded by a company which subsequently restates its financial reports, or
an increase in the likelihood of TypeⅠerrors that an ineffective ICFR is concluded by
a company which does not subsequently restate its financial reports.
2.3 SOX 404 and ICFR Disclosure Errors
Conceptually, to conclude ICFR as ineffective, three conditions must be met:
there exist control deficiencies; control deficiencies are discovered and managers
decide to disclose those deficiencies (Ashbaugh-Skaife et al. 2007). Following this
argument, both detection capability and disclosure incentives are important factors to
high-quality ICFR reports. Control weaknesses may not be discovered if managers are
not competent enough to apply adequate and sufficient assessment procedures in
evaluating internal control systems. SOX 404(a) requires a formal management
assessment of internal controls which needs to be prepared according to publicly
recognized internal controls frameworks, such as Committee of Sponsoring
Organizations of the Treadway Commission (COSO) or Control Objectives for
Information and Related Technology (COBIT) framework. Kinney and Shepardson
(2011) find substantial increase in material weakness disclosure rates for small
companies undergoing initial SOX 404(a) management self-assessment of ICFR. Thus,
the requirement of Section 404(a) may enhance detection capability of top
management and mitigate the problem of incompetence of managers in examining
- 9 -
SOX 404(b) requires that issuers’ internal control systems be evaluated by
external auditors who have professional knowledge, training and experience in
examining internal control systems. Most internal control weaknesses are detected by
auditors rather than by management and material weaknesses are more likely to be
disclosed at the fourth quarter when auditors are on-site and when auditors have
experience with internal control audits (Bedard and Grahma 2008; Hammersley et al.
2008). Besides, some studies indicate disclosure incentives of control systems are
associated with firm size, corporate governance, ownership structure, and firm growth
(Bronson et al. 2006; Deumes and Knechel 2008). Even if internal control weaknesses
are actually discovered, not all managers have equal level of incentives to report
internal control problems. Auditors are regarded as an effective monitoring
mechanism for corporate governance (Becker et al. 1998). A survey conducted by the
SEC indicates that investors regard auditor attestation of internal control as an
important measure in evaluation of internal control systems given the auditors’
expertise and independence (SEC 2009b). We argue that managers may feel
compelled to truthfully disclose weaknesses of their internal control systems if their
disclosures need to be attested by external auditors under SOX 404. As a result, the
requirement of auditors’ involvement in public ICFR reporting may help both
discover and disclose underlying internal control weaknesses.
In sum, additional management documentation and auditor scrutiny over ICFR
required by SOX 404 may help detect and disclose underlying internal control
problems and thus reduce TypeⅡerrors. This leads to our Hypothesis 1a:
Hypothesis 1a: The implementation of SOX 404 decreases the likelihood that a
company reports an effective internal control system but subsequently restates its
- 10 -
On the other hand, Doyle et al. (2007) find a negative relation between disclosed
internal control weaknesses and accruals quality under SOX 302 regime but they find
no such relation under SOX 404 regime. Beneish et al. (2008) find that unaudited
SOX 302 weakness disclosures are associated with negative abnormal returns while
audited SOX 404 disclosures are not. Both of the studies suggest that the auditor
attestation required by SOX 404 results in a lower threshold for disclosure under SOX
404 regime than SOX 302 regime. We argue that required signatures of auditors on
opinions regarding ICFR may make auditors unduly cautious in identifying material
weaknesses and their low effectiveness thresholds for material weaknesses could
result in many reported material weaknesses that do not lead to misstatements (Doyle
et al. 2007a), which in turn may increase the likelihood of making TypeⅠerrors. Our
Hypothesis 1b is stated as follows:
Hypothesis 1b: The implementation of SOX 404 increases the likelihood that a
company concludes its internal control system to be ineffective but does not
subsequently restate its financial statement.
2.4 AS5 and ICFR Disclosure Errors
PCAOB noted that although audits of internal control under AS2 improved audit
committee oversight, quality and transparency of financial reporting process, it also
incurred several significant costs (PCAOB 2006). For instance, some auditors
re-tested items which had been tested by management only to opine on management
assessments, or in some cases, auditors inappropriately dictated management to
- 11 -
business community about the onerous and costly requirement of internal control
audits, the PCAOB on December 15, 2007 suspended AS2 and enacted a new
simplified, relaxed standard for internal control audits, i.e. AS5. AS5 adopts a
top-down, risk-based approach, which includes several key steps: (1) identifying
significant financial reporting elements and associated material misstatement risks, (2)
determining appropriate entity and/or transaction level controls that can address these
risks with sufficient precision, and (3) determining the nature, extent and timing of
audit evidence which needs to be gathered to complete assessments of ICFR systems.
By eliminating of unnecessary procedures and testing in audits of internal controls,
AS5 is aimed to reallocate limited corporate resources to really notable, high-risk
areas (SEC 2007). Bell et al. (2008) argue that a risk-based audit approach which AS5
adopts may result in efficiency gains for less risky auditees and improvement in audit
efficacy for more risky auditees. Following this argument, it is expected that AS5
improves not only the efficiency but also the efficacy of audits of internal controls and
therefore reduces reporting errors of public ICFR disclosures, i.e. aforementioned
TypeⅠand TypeⅡerrors.
On the other hand, audits of internal controls prescribed by AS5 inevitably
involve auditor professional judgment which raises concerns that such leeway may
allow auditors to make excuses to cut back on work instead of improving the efficacy
of internal control audits3. PCAOB documented that after AS5 15 percent of the 309
audit engagements inspected by the Board failed to obtain sufficient audit evidence to
support their ICFR audit opinions (PCAOB 2012). Besides, risk-based audit approach
are related with the high-profile audit failures (e.g., HealthSouth and WorldCom) and
are previously criticized by PCAOB (Bell et al. 2008; Berkowitz and Rampell 2002;
- 12 -
Weil 2004). As a result, the implementation of AS5 may lead to under-auditing and
increase TypeⅡerrors if auditors misuse the risk-based approach.
Facing the two contradicting arguments, we are not sure whether TypeⅡ errors
of public ICFR disclosures increase or decrease after the AS5 implementation and it is
an empirical issue to examine in the study. In contrast, we expect that the efficiency of
ICFR audits improves and TypeⅠerrors decrease after the implementation of AS5.
Our Hypothesis 2a and 2b are stated as follows:
Hypothesis 2a: The substitution of AS5 for AS2 does not increase or decrease the
likelihood that a company reports an effective internal control system but
subsequently restates its financial statement.
Hypothesis 2b: The substitution of AS5 for AS2 decreases the likelihood that a
company concludes its internal control system to be ineffective but does not
subsequently restate its financial statement.
3. RESEARCH DESIGN
3.1 Sample
We begin with all company-year observations available in Audit Analytics’
disclosure control (SOX302)4 and internal control files (SOX404) for the period
2002–2010. We then merge the data with Audit Analytics restatement files and
4 SOX 302 requires management to conclude on the effectiveness of ICFR in periodic reports. In contrast, SOX 404 requires management assessments and audit opinions of ICFR disclosed in annual reports. To reconcile the inconsistency of reporting frequency, we consider only 302 disclosures made in annual reports.
- 13 -
Compustat Annual files, resulting in 50,304 company-year observations. Next, we
delete foreign companies, non-accelerated filers which are exempted from compliance
with SOX 404 and companies without necessary data for our empirical analysis.
Besides, prior studies show that in order to circumvent compliance with SOX 404,
companies may choose to keep their size small, go private or go dark (Gao et al. 2009;
Engel et al. 2007; Leuz et al. 2008). The incentives to circumvent the compliance of
SOX 404 may have an impact on ICFR disclosure quality, and we, therefore, drop
companies which had not continuously met the accelerated filer criteria (companies
whose public float is 75 million or more as of six months before the fiscal year-end)
for the period 2004–2010. Finally, our sample covers 12,627 company-year
observations. Table 1 details our sample selection procedures and the sample
composition.
[Insert Table 1 here]
3.2 Estimation Models
We use logistic regression to estimate the following models and predict the
likelihood of a company concluding its internal controls as effective:
(1) ) Prob( 11 10 9 8 7 6 5 4 3 2 1 0 INDUSTRY YEAR MARKETCAP AGLOSS FT RCP BIGN MB PE LEV ROA SIZE SOX404 EFFECTIVE (2) ) Prob( 11 10 9 8 7 6 5 4 3 2 1 0 INDUSTRY YEAR MARKETCAP AGLOSS FT RCP BIGN MB PE LEV ROA SIZE AS5 EFFECTIVE
- 14 -
where EFFECTIVE is coded 1 if a company concludes its internal control system as
effective in its10-K filing and 0 otherwise. Our experimental variables of interest are
SOX404 and AS5. SOX404 is a dummy variable with a value of 1 if a company is
subject to the SOX 404 regulation and 0 otherwise. AS5 is also a dummy variable
which is coded 1 if AS5 is applicable to a company and 0 otherwise.
We estimate Model 1 using the subsamples of restatement companies and
non-restatement companies separately. For the subsample of restatement companies,
based on H1a, we expect the coefficient on SOX404 to be negative, which would
indicate the implementation of SOX 404 reduces TypeⅡerrors in the sense that
restatement companies are less likely to conclude ICFR systems as effective under
SOX 404 regime. For the subsample of non-restatement companies, based on H1b, it
is expected that the coefficient on SOX404 is negative, which would indicate that
non-restatement companies are less likely to conclude ICFR systems as effective
under SOX 404 regime, resulting in a higher TypeⅠerror rate.
To test H2a and H2b, we draw a sample covering only companies subject to SOX
404 and estimate Model 2 in the same manner using the subsamples of restatement
companies and non-restatement companies separately. For the subsample of
restatement companies, based on H2a, we do not have directional expectation of the
coefficient on AS5. If the coefficient on AS5 is positive (negative), it indicates that
restatement companies adopting AS5 are more (less) likely to conclude ICFR systems
as effective, resulting in a higher (lower) TypeⅡerror rate. For the subsample of
non-restatement companies, based on H2b, it is expected that the coefficient on AS5 is
positive, which would indicate that non-restatement companies are more likely to
conclude ICFR systems as effective if they adopt AS5, resulting in a lower TypeⅠ
- 15 -
Prior studies indicate that the existence of internal control deficiencies is
associated with company size, operating loss, business risk, company growth, the
appointment of a Big 4 auditor, restructuring charges, and the incidence of foreign
transactions (Ge and McVay 2005; Ashbaugh-Skaife et al. 2007; Doyle et al. 2007b).
Consistent with prior studies, we control for companies size with natural log of total
assets (SIZE) and market capitalization (MARKETCAP). We proxy operating loss with
return on assets (ROA) and aggregate loss (AGLOSS). We proxy business risk with
leverage (LEV). Company growth is controlled with price-to-earnings (PE) and
market-to-book (MB) ratios. The appointment of a Big 4 auditor (BIGN), restructuring
charges (RCP) and the incidence of foreign transactions (FT) are also controlled in
our models. YEAR is a set of dummy variables separately representing each of the
fiscal years. Finally, we include a set of 13 industry indicator variables (INDUSTRY)
based on Frankel et al. (2002). To mitigate the effect of extreme values, we winsorize
all of the continuous variables at the 0.01 and 0.99 percentiles. Table 2 details our
variable definitions.
[Insert Table 2 here]
4. RESULTS
4.1 Descriptive Statistics
The descriptive statistics for all variables used in this study are presented in
Table 3. We further divide the sample into two categories: restatement and
- 16 -
companies with effective internal controls, respectively. As would be expected,
restatement companies are less likely to conclude their internal controls as effective
than their non-restatement counterparts (84.8 percent versus 98 percent). The two
groups are similar in price to book ratios, restructuring charges, and the incidence of
foreign transactions. Restatement companies, on average, are smaller in size and
market capitalization, have poor return on assets and aggregate loss, use less leverage,
and have lower market-to-book ratio than non-restatement counterparts. Additionally,
restatement companies are more likely to appoint Big 4 auditors.
[Insert Table 3 here]
Additional information about the ratios of effective ICFR disclosures among the
restatement and non-restatement samples is provided in Table 4. We can see from
Panel A that restatement companies subject to SOX 404 are significantly less likely to
conclude internal controls as effective compared to restatement companies subject to
SOX 302 (79 percent versus 97.6 percent, p-value < 0.000). Non-restatement
companies under SOX 404 regime, however, are also less likely to conclude internal
controls as effective compared to non-restatement companies under SOX 302 regime
(97.6% versus 99.7 percent, p-value < 0.000). The results in Panel A support our H1a
and H1b, indicating that the enactment of SOX 404 reduces the likelihood of TypeⅡ
errors but increases the likelihood of TypeⅠerrors in public ICFR disclosures.
Panel B presents the ratios of effective ICFR disclosures among the restatement
and non-restatement samples classified by the adoption of AS5. Restatement
companies are significantly more likely to conclude internal controls as effective in
- 17 -
71.5%, p-value < 0.000). This indicates that the implementation of AS5 results in a
higher Type Ⅱ error rate of public ICFR disclosures. On the other hand,
non-restatement companies are also significantly more likely to conclude internal
controls as effective in AS5 period than thier non-restatement counterparts in AS2
period (98.5percent versus 96.2%, p-value < 0.000), which supports our H2b and
indicates that the implementation of AS5 results in a lower TypeⅠerror rate.
[Insert Table 4 here]
Table 5 presents the Pearson correlation matrix for all variables used in the study.
The strongest pairwise correlations are -0.484 and the largest variance inflation
factors are less than 2.3. We do not find that there exist the problems of
multicollinearity in our sample.
[Insert Table 5 here]
4.2 Logistic Regression Results
The results of applying Model 1 to the full sample are presented in Table 6. For
the restatement sample, the coefficient on SOX404 is -0.932 with a p-value < 0.046,
indicating that the odds of concluding effective internal controls for restatement
companies subject to SOX 404 decrease by 60.62% compared to restatement
counterparts subject to SOX 302. It suggests that the enactment of SOX 404 reduces
Type Ⅱ errors of ICFR disclosure and is consistent with our H1a. For the
- 18 -
do not find evidence that the enactment of SOX 404 results in a higher TypeⅠerror
rate. Taken together, the enactment of SOX 404 reduces TypeⅡerrors of ICFR
disclosures without the side effect of increasing TypeⅠerrors.
For restatement companies, concluding internal controls as effective is more
likely for companies with more restructuring charges, companies without foreign
transactions, and companies having less aggregate loss. For non-restatement
companies, concluding internal controls as effective is positively associated with
market-to-book ratios, the appointment of a Big 4 auditor, restructuring charges and
market capitalization; it is negatively associated with price-to-earnings ratios, the
incidence of foreign transactions, and aggregate loss.
[Insert Table 6 here]
Table 7 presents the results of applying Model 2 to the subsample of SOX 404
era. For the restatement sample, the coefficient on AS5 is 1.143 with a p-value < 0.000,
indicating that the odds of concluding effective internal controls for restatement
companies adopting AS5 increase by 213.62% compared to their restatement
counterparts adopting AS2. It suggests that the implementation of AS5 results in a
higher TypeⅡerror rate. For the non-restatement sample, the coefficient on AS5 is
1.148 with a p-value < 0.000, indicating that the odds of concluding effective internal
controls for non-restatement companies adopting AS5 increase by 215.19% compared
to their non-restatement counterparts adopting AS2. It is consistent with our H2b and
suggests that implementation of AS5 can improve ICFR audit efficiency by reducing
- 19 -
AS5 can enhance the efficiency of ICFR audits, it inadvertently lowers public ICFR
disclosure quality measured by Type Ⅱ errors.
The significance of the control variables in Model 2 is quite similar with that in
Model 1 except some differences. For the restatement companies, while FT is
significant in Model 1, it is insignificant in Model 2. On the contrary, while
MARKETCAP is insignificant in Model 1, it is significant in Model 2. For the
non-restatement companies, RCP and FT are insignificant in Model 2 while they are
significant in Model 1.
[Insert Table 7 here]
5. SESITIVITY AND ADDITIONAL TESTS
5.1 Different Way to Identify Companies with Ineffective ICFR
Although prior studies suggest that misstatements are indicative of ineffective
internal control systems (Kinney and McDaniel 1989; DeFond and Jiambalvo 1991;
McMullen et al. 1996), it is arguable that there may exist some extenuating or unique
situations in which a misstatement does not always parallel to ineffective internal
controls (PCAOB 2004). In this case, applying our research method could not exactly
measure the likelihood of TypeⅠand TypeⅡerrors.
Ashbaugh-Skaife et al. (2006) and Doyle et al. (2007a) suggest that internal
control problems are associated with lower accruals quality because weak internal
controls are reasonably possible to fail to detect intentionally biased accruals through
earnings management or unintentional accrual estimation errors. As a result, low
- 20 -
internal controls exists. We measure accrual estimation error based on the method
developed by Dechow and Dichev (2002) and modified by McNichols (2002) and
Francis et al. (2005). We divide our sample into two categories: (1) low-AQ sample
which covers observations whose residual is above the median and (2) high-AQ
sample which covers observations whose residual is below the median. Except the
different way to identify companies with weak ICFR, we conduct the analyses with
the same research method described above. The results are presented in Table 8 and
Table 9 and are very similar with our main results. Our results are consistent across
the different ways to identify companies with weak internal controls.
[Insert Table 8 here]
[Insert Table 9 here]
5.2 Periods of Financial Restatements
Doyle et al. (2007a) argue that material weaknesses, on average, have existed for
several years before being reported. Due to the long duration of material weaknesses,
weak internal controls may increase the likelihood of misstatements in not only
current but also future financial reports. In this sensitivity test, we regard companies
which restate financial statements of the current or next period as those which are
more likely to have material weaknesses and should be more likely to receive adverse
ICFR opinions in the current period. We replicate the analyses using the alternative
definition of restatement companies. The results are presented in Table 10 and Table
11 and are very similar with our main results, indicating that our main results are
- 21 -
[Insert Table 10 here]
[Insert Table 11 here]
6. CONCLUSION
In this study, we examine both of the impacts of SOX 404 and AS5 on ICFR
disclosure errors. We find that the enactment of SOX 404 results in a lower TypeⅡ
error rate without the side effect of increasing TypeⅠerrors. Moreover, we document
that even though the more flexible and laxer AS5 can enhance the efficiency of ICFR
audits manifested by reduced TypeⅠerrors, the leeway provided by AS5 allows
auditors to misuse their professional judgment and cut back on necessary testing
procedures in the audits of internal controls, which inadvertently results in lower
public ICFR disclosure quality measured by increased Type Ⅱ errors. Our results
echo the concerns raised by investors and regulators about AS5. A potential limitation
of this study is the restricted sample which we can draw because only large companies
(i.e. accelerated filers) are confirmed to be subject to SOX 404(b) after the passage of
Dodd-Frank Wall Street Reform and Consumer Protection Act in 2010. It is unclear
whether our evidence can be generalized to other small companies. Our paper is
among the first studies which provide insights into both of the impacts of SOX 404
and AS5 on ICFR disclosure quality and may help policymakers and standard setters
- 22 -
REFERENCES
Ashbaugh-Skaife, H., D. Collins, W. Kinney, and R. LaFond. 2006. The effect of SOX internal control deficiencies and their remediation on accrual quality. The
Accounting Review 83 (1): 217–250.
Ashbaugh-Skaife, H., D. Collins, and W. Kinney. 2007. The discovery and reporting of internal control deficiencies prior to SOX-mandated audits. Journal of
Accounting and Economics 44 (1-2): 166–192.
Becker, C. L., M. L. DeFond, J. Jiambalvo and K. R. Subramanyam. 1998. The effect of audit quality on earnings management. Contemporary Accounting Research 15 (1): 1–24.
Bedard, J. C., and L. Graham. 2008. Archival evidence on processes and outcomes of internal control assessment under Section 404 of the Sarbanes-Oxley Act. Working paper, Bentley University.
Bell, T. B., R. Doogar, and I. Solomon. 2008. Audit labor usage and fees under business risk auditing. Journal of Accounting Research 46 (4): 729–760. Berkowitz, A., and R. Rampell. 2002. Drive-by audits have become too common and
too dangerous. Wall Street Journal (August 9). Available at:
http://online.wsj.com/article/SB1028822538710052160.html.
Bronson, S., J. Carcello, and K. Raghunandan. 2006. Firm characteristics and voluntary management reports on internal control. Auditing: A Journal of
Practice & Theory 25 (2): 25–39.
Cox, C. 2007. Testimony, Sarbanes-Oxley Section 404: new evidence on the cost for small companies, before the U.S. House of Representatives Committee on small business. Available at:
http://www.sec.gov/news/testimony/2007/ts121207cc.htm.
Dechow, P., and I. Dichev. 2002. The quality of accruals and earnings: The role of accrual estimation error. The Accounting Review 77 (Supplement): 35–59. DeFond, M. L., and J. Jiambalvo. 1991. Incidence and circumstances of accounting
errors. The Accounting Review 66 (3): 643–655.
Deumes, R., and W. R. Knechel. 2008. Economic incentives for voluntary reporting on internal risk management and control systems. Auditing: A Journal of
Practice & Theory 27 (1): 35–66.
Doogar, R., P. M. Sivadasan, and P. I. Solomon. 2010. The regulation of public company auditing: Evidence from the transition to AS5. Journal of Accounting
Research 48 (4): 795–814.
Doyle, J., W. Ge, and S. McVay. 2007a. Accruals quality and internal control over financial reporting. The Accounting Review 82 (5): 1141–1170.
Doyle, J., W. Ge, and S. McVay. 2007b. Determinants of weaknesses in internal
control over financial reporting. Journal of Accounting and Economics 44 (1–2): 193–223.
Eldridge, S., and B. Kealey. 2005. SOX costs: auditor attestation under Section 404. Working Paper.
Engel, E., R. M. Hayes, and X. Wang. 2007. The Sarbanes-Oxley Act and firms’ going private decisions. Journal of Accounting and Economics 44 (1–2): 116–145. Ettredge, M., L. Chan, and S. Scholz. 2007. Audit fees and auditor dismissals in the
Sarbanes-Oxley era. Accounting Horizons 21 (4): 371–386.
Francis J., R. LaFond, P. Olsson, and K. Schipper. 2005. The market pricing of acruals quality. Journal of Accounting and Economics 39 (2): 295–327.
- 23 -
Frankel, R., M. Johnson, and K. Nelson. 2002. The relation between auditors’ fee for nonaudit services and earnings management. The Accounting Review 77 (Supplement): 71–105.
Gao F., J. S. Wu, and J. Zimmerman. 2009. Unintended consequences of granting small firms exemptions from securities regulation: evidence from the Sarbanes Oxley Act. Journal of Accounting Research 47 (2): 459–506.
Ge, W., and S. McVay. 2005. The disclosure of material weaknesses in internal control after the Sarbanes-Oxley Act. Accounting Horizons 19 (3): 137–158.
Hartman T. E.. 2007. The cost of being public in the era of Sarbanes-Oxley. Foley & Lardner LLP.
Hammersly, J.S., L.A. Myers., and C. Shakespeare. 2008. Market reactions to the disclosure of internal control weakness and to the characteristics of those weaknesses under section 302 of the Sarbanes Oxley Act of 2002. Review of
Accounting Studies 13 (1): 141–165.
Jiang, W., and J. Wu. 2009. The impact of PCAOB Auditing Standard 5 on audit fees. The CPA Journal 79(4): 34–38.
Johnson, C. 2005. Public companies complain to SEC about audit costs. Washington Post (April 9): E01.
Kinney, W., and L. McDaniel. 1989. Characteristics of firms correcting previously reported earnings. Journal of Accounting and Economics 11 (1): 71–93. Kinney, W. R., and M. L. Shepardson. 2010. Do control effectiveness disclosures
require SOX 404(b) internal control audits? A natural experiment with small U.S. public companies. Journal of Accounting Research 49 (2): 413–448. Krishnan, J., J. Krishnan, and H. Song. 2011. The effect of Auditing Standard No. 5
on audit fees. Auditing: A Journal of Practice & Theory 30 (4): 1–27.
Knechel W. R., and A. Vanstraelen. 2007. The relationship between auditor tenure and audit quality implied by going concern opinions. Auditing: A Journal of
Practice & Theory 26 (1): 113–131.
Leuz, C., A. Triantis, and T. Wang. 2008. Why do firms go dark: Causes and economic consequences of voluntary SEC deregistrations. Journal of
Accounting Research 45 (2–3): 181–208.
McMullen, D., K. Raghunandan, and D. Rama. 1996. Internal Control Reports and Financial Reporting Problems. Accounting Horizons 10 (4): 67–75. The
Accounting Review 77 (Supplement): 61–69.
McNichols, M. 2002. Discussion of the quality of accruals and earnings: The role of accrual estimation errors.
Olson, M. W. 2007. Remarks at the compliance week annual conference on governance, risk & compliance. June 7, 2007. Available at:
http://pcaobus.org/News/Speech/Pages/06072007_MarkWOlsonRemarks.aspx. Public Company Accounting Oversight Board (PCAOB). 2004. An Audit of Internal
Control over Financial Reporting Performed in Conjunction with an Audit of
Financial Statements. Auditing Standard (AS) No. 2. Washington, D. C.: PCAOB.
Public Company Accounting Oversight Board (PCAOB). 2006. An Audit of Internal
Control Over Financial Reporting That Is Integrated with An Audit of Financial Statements and Related Other Proposals. Washington, D. C.: PCAOB.
Public Company Accounting Oversight Board (PCAOB). 2007. An Audit of Internal
Control Over Financial Reporting That Is Integrated with An Audit of Financial Statements. Auditing Standard (AS) No. 5. Washington, D. C.: PCAOB.
- 24 -
2010 inspections of domestic annually inspected firms regarding deficiencies in audits of internal control over financial reporting. Release No. 2012-006. Washington, D. C.: PCAOB.
Rice, S. C. and D. P. Weber. 2012. How effective is internal control reporting under SOX 404? Determinants of the (Non-) Disclosure of Existing Material Weaknesses. Journal of Accounting Research 50 (3): 811–843.
Securities and Exchange Commission (SEC). 2007. SEC approves PCAOB Auditing Standard No. 5 regarding audits of internal control over financial reporting; adopts definition of “significant deficiency.” News Release 2007-144. Washington, D.C.: Government Printing Office.
Securities and Exchange Commission (SEC). 2009a. Remarks before the 2009 AICPA national conference on current SEC and PCAOB developments. (Besch, D.) December 7. Washington, D.C.: SEC
Securities and Exchange Commission (SEC). 2009b. Study of the Sarbanes-Oxley Act
of 2002 section 404 internal control over financial reporting. Washington, D. C.:
SEC.
Wang D., and J. Zhou. 2012. The impact of PCAOB Auditing Standard No. 5 on audit fees and audit quality. Auditing: A Journal of Practice & Theory 26 (3): 493– 511.
Weil, J. 2004. Missing numbers–Behind wave of corporate fraud: A change in how auditors work. Wall Street Journal (March 25): 1.
Whitehouse, T. 2010. SEC Curious about drop in material weaknesses. Compliance Week 7: 32–63.
25
TABLE 1
Sample Selection and Sample Composition
Panel A: Sample Selection
Initial sample of company-year observations covered by both Analytics and Compustat from the period 2002–2010 50,304
Less: Foreign company observations (1,220)
Less: Companies observations without necessary data for our empirical analyses (7,381)
Less: non-accelerated filers and companies which had not continuously met the accelerated filer criteria from the period 2004– 2010
(29,076)
Final Sample 12,627
Panel B: Sample Composition
Company-year observations under SOX 302 regime 2,722
Company-year observations under SOX 404 regime 9,905
Total 12,627
Company-year observations to which AS2 is applicable 4,245
Company-year observations to which AS5 is applicable 5,660
26
TABLE 2 Variable Definitions
EFFECTIVE = dummy variable coded 1 if a company concludes its internal control system as effective in its10-K filing and 0 otherwise;
SOX404 = dummy variable with a value of 1 if a company is subject to the SOX 404 regulation and 0 otherwise; AS5 = dummy variable coded 1 if AS5 is applicable to a company and 0 otherwise;
SIZE = natural log of total assets in the unit of million dollars;
ROA = net income before extraordinary items divided by total assets; LEV = total liabilities divided by total assets;
PE = year-end share price divided by earnings per share; MB = year-end market value divided by book value;
BIGN = dummy variable coded 1 if the appointed auditor is a Big 4 auditor and 0 otherwise; RCP = aggregate restructuring charges scaled by market capitalization;
FT = indicator variable coded 1 if a company has a non-zero foreign currency translation and 0 otherwise;
AGLOSS = dummy variable coded 1 if earnings before extraordinary items in years t and t-1 sum to less than zero and 0 otherwise;
MARKETCAP = natural log of share price * number of outstanding shares;
YEAR = a set of year dummy variables;
27 TABLE 3 Descriptive Statistics Variables Means Total Sample Restatement Sample (n=1853) (1) Non-Restatement Sample (n=10774) (2) Difference (2) – (1) p-value EFFECTIVE 0.961 0.848 0.980 0.132 0.000 SIZE 7.256 7.048 7.292 0.244 0.000 ROA 0.183 0.004 0.021 0.017 0.000 LEV 0.572 0.560 0.575 0.015 0.054 PE 14.825 14.102 14.898 0.796 0.623 MB 2.775 2.594 2.806 0.212 0.044 BIGN 0.861 0.890 0.856 -0.034 0.000 RCP -8.636 -7.980 -8.749 -0.769 0.269 FT 0.244 0.251 0.243 -0.008 0.416 AGLOSS 0.191 0.228 0.185 -0.043 0.000 MARKETCAP 4267.138 3074.847 4472.198 1397.351 0.000
28
TABLE 4
Ratio of Effective ICFR Disclosures
Panel A: SOX 302 and SOX 404
SOX302 Sample (1) SOX404 Sample (2) Difference (2) - (1) p-value Restatement Companies 0.976 0.790 -0.186 0.000 Non-Restatement Companies 0.997 0.976 -0.020 0.000
Panel B: AS2 and AS5
AS2 Sample (1) AS5 Sample (2) Difference (2) - (1) p-value Restatement Companies 0.715 0.886 0.171 0.000 Non-Restatement Companies 0.962 0.985 0.023 0.000
29
TABLE 5
Pearson Correlation Matrix
SIZE ROA LEV PE MB BIGN RCP FT
AGLO SS MARK-ETCAP SOX 404 AS5 VIF SIZE 1.000 2.30 ROA 0.219 1.000 1.52 LEV 0.331 -0.246 1.000 1.43 PE 0.036 0.159 -0.058 1.000 1.08 MB -0.124 0.095 -0.128 0.027 1.000 1.08 BIGN 0.112 0.013 -0.142 -0.010 0.002 1.000 1.07 RCP -0.334 0.024 -0.102 0.027 -0.008 -0.105 1.000 1.24 FT 0.060 0.023 -0.099 0.005 -0.008 0.121 -0.164 1.000 1.06 AGLOSS -0.211 -0.484 0.082 -0.262 -0.025 0.023 -0.059 0.045 1.000 1.46 MARKET CAP 0.604 0.137 0.023 0.048 0.090 0.132 -0.370 0.106 -0.147 - 1.000 1.85 SOX404 0.099 0.030 0.016 0.006 0.003 0.030 -0.012 0.079 0.105 0.048 1.000 1.31 AS5 0.099 -0.023 0.039 -0.040 -0.050 0.020 -0.066 0.084 0.141 0.020 0.473 1.000 1.32
30
TABLE 6
Logistic Regression for Restating and Non-restating Samples (Dependent Variable = EFFECTIVE)
Restatement Sample
Non-restatement Sample
Coefficient p-value Coefficient p-value
SOX404 -0.932 0.046 -0.682 0.204 SIZE 0.075 0.306 0.008 0.918 ROA 0.224 0.697 -0.466 0.351 LEV -0.225 0.395 -0.146 0.579 PE 0.000 0.914 -0.003 0.077 MB 0.010 0.565 0.046 0.013 BIGN 0.283 0.259 0.592 0.003 RCP 0.006 0.062 0.007 0.023 FT -0.356 0.051 -0.285 0.096 AGLOSS -0.503 0.010 -0.715 0.000 MARKETCAP 0.000 0.246 0.000 0.000 CONSTANT 2.689 0.034 18.164 0.982
YEAR (include) (include)
INDUSTRY (include) (include)
LR Chi Squared 271.78 0.000 272.12 0.000
Pseudo R2 0.172 0.132
31
TABLE 7
Logistic Regression for Restating and Non-restating Samples (Dependent Variable = EFFECTIVE)
Restatement Sample
Non-restatement Sample
Coefficient p-value Coefficient p-value
AS5 1.413 0.000 1.418 0.000 SIZE 0.095 0.217 -0.005 0.950 ROA 0.242 0.690 -0.565 0.285 LEV -0.199 0.469 -0.183 0.500 PE 0.000 0.968 -0.003 0.094 MB 0.009 0.622 0.043 0.023 BIGN 0.327 0.198 0.578 0.004 RCP 0.007 0.050 0.006 0.101 FT -0.300 0.115 -0.277 0.112 AGLOSS -0.434 0.031 -0.752 0.000 MARKETCAP 0.000 0.077 0.000 0.001 CONSTANT -0.292 0.816 16.321 0.985
YEAR (include) (include)
INDUSTRY (include) (include)
LR Chi Squared 128.71 0.000 213.85 0.000
Pseudo R2 0.099 0.111
32
TABLE 8
Logistic Regression for High-AQ and Low-AQ Samples (Dependent Variable = EFFECTIVE)
Low-AQ Sample High-AQ Sample Coefficient p-value Coefficient p-value
SOX404 -0.935 0.095 0.661 0.207 SIZE 0.070 0.388 -0.044 0.569 ROA 0.043 0.906 -0.622 0.293 LEV -0.192 0.366 -0.387 0.173 PE -0.001 0.444 0.001 0.554 MB 0.027 0.069 0.011 0.512 BIGN 0.370 0.106 0.540 0.039 RCP 0.012 0.000 0.002 0.533 FT -0.151 0.387 -0.556 0.001 AGLOSS -0.309 0.137 -0.988 0.000 MARKETCAP 0.000 0.002 0.000 0.003 CONSTANT 17.383 0.982 18.451 0.984
YEAR (include) (include)
INDUSTRY (include) (include)
LR Chi Squared 234.42 0.000 322.23 0.000
Pseudo R2 0.143 0.162
33
TABLE 9
Logistic Regression for High-AQ and Low-AQ Samples (Dependent Variable = EFFECTIVE)
Low-AQ Sample High-AQ Sample Coefficient p-value Coefficient p-value
AS5 1.205 0.001 2.395 0.000 SIZE 0.024 0.790 -0.044 0.591 ROA 0.073 0.850 -0.615 0.312 LEV -0.174 0.438 -0.400 0.169 PE -0.001 0.579 0.000 0.831 MB 0.024 0.113 0.010 0.572 BIGN 0.347 0.137 0.550 0.036 RCP 0.013 0.000 0.001 0.778 FT -0.116 0.520 -0.574 0.001 AGLOSS -0.288 0.174 -0.944 0.000 MARKETCAP 0.000 0.001 0.000 0.002 CONSTANT 14.215 0.974 15.934 0.979
YEAR (include) (include)
INDUSTRY (include) (include)
LR Chi Squared 182.21 0.000 267.99 0.000
Pseudo R2 0.123 0.148
34
TABLE 10
Logistic Regression for Restating and Non-restating Samples (Dependent Variable = EFFECTIVE)
Restatement Sample (Current & Next Period)
Non-restatement Sample (Current & Next Period) Coefficient p-value Coefficient p-value
SOX404 -0.959 0.034 -0.564 0.300 SIZE 0.059 0.398 0.029 0.721 ROA 0.200 0.675 -0.609 0.249 LEV -0.156 0.521 -0.140 0.611 PE -0.000 0.770 -0.003 0.093 MB 0.009 0.616 0.046 0.014 BIGN 0.218 0.350 0.620 0.003 RCP 0.007 0.034 0.008 0.016 FT -0.380 0.026 -0.258 0.144 AGLOSS -0.493 0.007 -0.795 0.000 MARKETCAP 0.000 0.307 0.000 0.001 CONSTANT 3.202 0.010 16.971 0.972
YEAR (include) (include)
INDUSTRY (include) (include)
LR Chi Squared 273.85 0.000 267.03 0.000
Pseudo R2 0.156 0.135
35
TABLE 11
Logistic Regression for Restating and Non-restating Samples (Dependent Variable = EFFECTIVE)
Restatement Sample (Current & Next Period)
Non-restatement Sample (Current & Next Period) Coefficient p-value Coefficient p-value
AS5 1.298 0.000 1.453 0.000 SIZE 0.078 0.291 0.016 0.847 ROA 0.196 0.691 -0.726 0.193 LEV -0.139 0.579 -0.178 0.533 PE -0.001 0.710 -0.003 0.115 MB 0.008 0.673 0.043 0.024 BIGN 0.252 0.286 0.607 0.004 RCP 0.008 0.030 0.006 0.077 FT -0.328 0.065 -0.249 0.168 AGLOSS -0.431 0.021 -0.838 0.000 MARKETCAP 0.000 0.105 0.000 0.001 CONSTANT 0.428 0.727 15.229 0.977
YEAR (include) (include)
INDUSTRY (include) (include)
LR Chi Squared 133.36 0.000 213.95 0.000
Pseudo R2 0.090 0.117
1
公司報導誘因能否抑制與修正內部控制缺失？
摘要
有學者研究主張公司之財務報導品質主要受到報導誘因影響，而非會計準則差 異所致。本研究採用 Daske et al. （2013）所發展之報導誘因綜合指標，探討其與 公司報導及修正內控缺失之關係。實證結果支持假說一，表示當公司報導誘因越高， 其在 2004~2011 期間越不易發生內部控制重大缺失；在假說二方面，並未證實當缺 失公司的報導誘因越高時，較不易存在公司層級的重大缺失；假說三以 2004 年至 2009 年間已發生內部控制缺失的公司作為研究對象，結果發現在 2004~2009 年期 間相較於低報導誘因的缺失公司，當缺失公司的報導誘因越高時，並未顯著地較易 於後續年度中對缺失進行修正。 關鍵詞：報導誘因、內部控制缺失、缺失修正2
Do Firms’ Reporting Incentives Restrain and Remediate
Internal Control Weaknesses?
Abstract
Prior researchers advocate that financial reporting quality is influenced mainly by firms’ reporting incentives, but not due to the difference of accounting standards. We follow Daske et al. (2013) and develop a comprehensive index of reporting incentives to examine its relationship with firms’ reporting and remediation of their internal control weaknesses. The empirical result supports hypothesis 1, that is, the higher firms reporting incentives, the less likely report internal control weaknesses. We cannot find evidence that weaknesses firms with higher reporting incentives report less company-level weaknesses. We further use weaknesses firms during 2004-2009 to examine hypothesis 3. The results cannot support that weaknesses firms with higher reporting incentives are more likely to remediate weaknesses in two years.
