kong - 2005 - an empirical study of audit quality in china

An Empirical Study of Audit Quality in China From the Perspectives of Regulators and the Accounting Profession by Wong Man Kong, Raymond Doctor of Philosophy The effective functioning

AN EMPIRICAL STUDY OF AUDIT QUALITY IN CHINA FROM THE PERSPECTIVES OF REGULATORS AND THE ACCOUNTING PROFESSION

by Wong Man Kong, Raymond

A thesis submitted in partial fulfillment

of the requirements for the Degree of

Doctor of Philosophy

School of Accounting and Finance The Hong Kong Polytechnic University

September 2005

UMI Number: 3205525

INFORMATION TO USERS

The quality of this reproduction is dependent upon the quality of the copy submitted Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction

In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted Also, if unauthorized copyright material had to be removed, a note will indicate the deletion

®

UMI

UMI Microform 3205525 Copyright 2006 by ProQuest Information and Learning Company All rights reserved This microform edition is protected against unauthorized copying under Title 17, United States Code

ProQuest Information and Learning Company

300 North Zeeb Road P.O Box 1346

CERTIFICATE OF ORIGINALITY

I hereby declare that this thesis is my own work and that, to the best of my knowledge and belief, it reproduces no material previously published or written, nor material that has been accepted for the award of any other degree or diploma, except where due acknowledgement has been made in the text

WONG Man Kon£ (Name of student)

An Empirical Study of Audit Quality in China From the Perspectives of

Regulators and the Accounting Profession

by Wong Man Kong, Raymond

Doctor of Philosophy

The effective functioning of securities markets depends on the quality of audits and the integrity of auditors However, the recent tide of corporate scandals and audit failures has shocked the public, and both audit quality and auditor independence have been questioned It is therefore in the interests of both auditors and regulators to understand the main causes of audit failure, and to investigate the factors that drive audit quality While much of the focus in the accounting literature has been on audit quality in the U.S and other developed countries, a comprehensive study of the audit quality in developing countries is of equal significance given the increasing importance of international investment The increasingly vital role that is played by China in the world economy and its unique institutional setting provide a motivation for the examination of the specific factors that drive audit quality

In this thesis, I first evaluate audit quality directly through an examination of the cases of audit failure that have been discovered by the Chinese regulator, the China Securities Regulatory Commission (CSRC) I attempt to show the extent to which the regulator perceives auditors to be responsible for their failure to detect fraud Then I use auditor independence as a measure of audit quality to investigate two specific factors that drive auditor independence from the perspective of the accounting profession I hypothesize that regulatory sanctions serve as an incentive to improve auditor independence, and further posit that the organizational form of CPA firms affects the independence level of auditors

The empirical results indicate that audit failures that involve material misstatements by clients are more likely to lead to enforcement actions against auditors than disclosure fraud Misstatement fraud that involves accounting earnings

is more likely to lead to enforcement actions against auditors than misstatement fraud that involves balance sheet figures The results reveal a strong association between

several types of fraud and audit failures from the regulator’s perspective To examine the factors that drive auditor independence, I use the issuance of modified audit opinions as a proxy for auditor independence, and demonstrate that sanctioned auditors act more independently after being subject to regulatory sanctions and that CPA firms that are registered as partnership firms are more independent than firms that are registered as limited liability firms This implies that formal regulatory sanctions and the organizational form of CPA firms are important drivers of auditor independence and audit quality in China.

Overall, the results make several contributions to regulators, the accounting profession, and academia First, the analyses on audit failures provide direct and useful information for the improvement of audit quality In particular, regulators and the accounting profession can make use of my findings in their policy making Regulators and policy makers may consider devising new accounting and auditing standards to govern substandard audits to detect specific types of fraud The accounting profession may also consider providing technical training for staff to identify various prominent types of fraud Second, the evidence of an improvement

in auditor independence that is motivated by regulatory sanctions implies that formal sanctions serve as an effective deterrent to the unethical behavior of auditors This result supports the view that more resources should be devoted to regulatory supervision to improve audit quality and auditor independence Finally, the results on the organizational form of CPA firms contribute to the literature by empirically demonstrating the link between auditor independence and the organizational form of CPA firms This finding is also useful for regulators and the accounting profession as they attempt to evaluate the optimal balance in implementing a limited liability regime without diminishing audit quality and auditor independence

ACKNOW LEDGEMENTS

I wish to acknowledge the help of many people during my study The first and foremost thanks go to my two supervisors, Professor Phyllis L L Mo and Professor Michael A Firth This dissertation would not have been possible without their supports and encouragements throughout my study I owe my great deal to my chief supervisor, Professor Phyllis Mo, to her dedication in offering me direction and penetrating criticism, as well as guiding me through draft after draft over the past several years My gracious thanks also go to my co-supervisor, Professor Michael Firth, who has been so generous with his time in providing valuable comments on

my thesis and is a wonderful coach in guiding me throughout this dissertation Their advice on both substance and style are palpable on every page of this dissertation

Thanks are also due to many faculty members at The Hong Kong Polytechnic University I am greatly indebted to Professor Ferdinand Gul, Dr Peter Cheng, Dr Thomas Lau, Dr Ben Bao, and Dr Sandra Ho, for their insightful comments on various chapters of my thesis I also thank many academic colleagues and professionals in Hong Kong, China, and the United States who provided valuable suggestions, comments, and insights on audit quality and auditors’ practices in China

Last but not least, I would like to thank my family members whose constant love and encouragement enable me to persist throughout my study Especially, my wife, Agnes, is my source of inspiration and impetus for pursing my study and finishing this dissertation

An Empirical Study of Audit Quality in China From the Perspectives of

Regulators and the Accounting Profession

Chapter 3 Audit Quality from the Perspective of Regulators

3.5 Empirical Results 463.5.1 Sanctions on CPA Firms and Audit Partners 463.5.2 Fraud Characteristics and Univariate Tests 47

Chapter 4 Audit Quality from the Perspective of the Accounting

Profession - Formal Sanctions as an Effective Incentive for Auditor Independence

4.2.1 Chinese Institute of Certified Public Accountants (CICPA)

Chapter 5 Audit Quality from the Perspective of the Accounting

Profession - An Empirical Examination of the Organizational Form of CPA Firms and Auditor Independence

5.5.3 Results for Testing the Association between Auditor Switch

and Audit Opinions (Hypothesis H2) 1095.5.3a Univariate Tests and Correlation Matrix 109

5.5.4 Results for Testing the Association between Incorporation ofPartnerships and Audit Opinions (Hypothesis H3) 1115.5.4a Univariate Tests and Correlation Matrix 111

LIST OF TABLES

Table 1-1 Table 1-2 Table 2-1Table 3-1

Table 3-2 Table 3-3

Table 3-4 Table 3-5Table 3-6Table 4-1Table 4-2

Table 4-3 Table 4-4

Table 4-5 Table 4-6Table 5-1Table 5-2 Table 5-3Table 5-4

Table 5-5

Average Annual Percentage Growth of Real GDP Development of the Chinese Stock Market Development of the Accounting Profession for the Audits of Listed Companies

Sample Collection Procedure for the CSRC Enforcement Releases

Descriptive Statistics of Enforcement Releases Analysis of the Punishments Enforced in the Enforcement Actions

Detailed Description of Fraud Descriptive Statistics and Univariate Tests for Financial Statement Fraud and Auditor Sanctions

Multivariate Analysis Results for Financial Statement Fraud and Auditor Sanctions

Sample Descriptions for Testing the Effectiveness of Auditor Sanctions

Market Reaction to Sanctioned Clients with Sanctioned Auditors versus Sanctioned Clients with Non-sanctioned Auditors

Consequences for Sanctioned CPA Firms and Audit Partners after Enforcement Action

Univariate Tests for Sanctioned CPA Firms’ Market Share, Audit Partners’ Share of the Client Portfolio, and the Percentage of Modified Opinions Issued before and after the Enforcement Actions

Descriptive Statistics of Auditor and Client Characteristics Multivariate Results for the Effectiveness of Auditor Sanctions

Sample Descriptions for Testing the Association between the Legal Form of CPA Firms and Audit Opinions

Descriptive Statistics of the Sample Firms Univariate Tests for the Association between the Legal Form

of CPA Firms and Audit Opinions (full sample) Pearson (Spearman’s rho) Correlations for the Association between the Legal Form of CPA Firms and Audit Opinions (full sample)

Multivariate Tests for the Association between the Legal

123124125

126127

128129130131133

134135

136137138

139140141

142

Table 5-6 Table 5-7

Table 5-8 Table 5-9

Pearson (Spearman’s rho) Correlations for the Association between Auditor Switch and Audit Opinions (auditor-switch subsample)

Multivariate Tests for the Association between Auditor Switch and Audit Opinions (auditor-switch subsample) Univariate Tests for the Association between Incorporation

of Partnerships and Audit Opinions (incorporation subsample)

Pearson (Spearman’s rho) Correlations for the Association between Incorporation of Partnerships and Audit Opinions (incorporation subsample)

Multivariate Tests for the Association between Incorporation of Partnerships and Audit Opinions (incorporation subsample)

Sensitivity Tests for the Association between the Legal Form

of CPA Firms and Audit Opinions

143144

145146

147

148

149150

CHAPTER 1

INTRODUCTION

1.1 MOTIVATION FOR THE RESEARCH

The demand for auditing services arises from a desire to reduce the divergence

of interests and information asymmetry between the owners (the principal) and managers (the agent) in a principal-agent relationship (Jensen and Meckling 1976; Watts and Zimmerman 1986) Managers can voluntarily increase the transparency of their actions by hiring independent auditors to monitor their behavior

Recent research provides empirical evidence that high-quality, independent audits are used as monitoring and bonding mechanisms to alleviate agency costs (DeFond 1992; Fan and Wong 2005; Francis and Wilson 1988) In addition, high-quality auditors give greater credibility and better quality to financial statements (Khurana and Raman 2004; Teoh and Wong 1993), are more likely to constrain the opportunistic behavior of management (Becker et al 1998; Francis et al 1999; Kim

et al 2003), experience a higher incidence of auditor-client disagreement over income-increasing accounting methods (DeFond and Jiambalvo 1993), report more conservatively by issuing more modified audit reports (Francis and Krishnan 1999;

DeFond et al 2000; Chen et al 2001), and have a greater efficiency in technologically specialized industries (Craswell et al 1995) than low-quality auditors

As a consequence, high-quality auditing services are of particular importance in lending credibility to financial statements and increasing investor confidence Auditors specialize in the supply of various levels of quality (product differentiation), and clients demand different levels of audit quality because of the differential agency costs across firms (DeAngelo 1981; DeFond 1992) Understanding the factors that are associated with audit quality is a vital concern for investors, policy makers, regulators, the accounting profession, and the general public alike

1.1.2 Importance of Audit Quality in China

Table 1-1 reports the annual growth in GDP for the world and selected developed and developing economies According to the International Monetary Fund (IMF) (2003; 2005), the ten-year average growth in GDP for the period of 1995 to

2004 for China is 9.5 percent, which is significantly greater than the growth of the United States (3.7 percent), the Euro area (2.2 percent), Japan (1.3 percent), and India (6.8 percent), and is double that of the world average (4.2 percent) With the rapid growth in its economy, China is playing an increasingly important role in world economic trade Some analysts are predicting that China’s GDP will surpass that of Germany by 2010 and Japan by 2015, and will equal the GDP of the United States by

around 2040 (see, for example, Kren 2005) In addition, the expansion of the domestic stock market has aroused the interest of foreign investors to invest in China Table 1-2 documents the development of the Chinese domestic stock market Since the establishment of the two stock exchanges in the early 1990s, the stock market has grown rapidly (Haw et al 2003) In 1991, thirteen firms were listed on the two stock exchanges (eight on the Shanghai Stock Exchange and five on the Shenzhen Stock Exchange) By the end of 2004, 1,373 firms were listed on the exchanges, and the total market capitalization of the listed firms reached RMB3,706 billion, which is equivalent to approximately US$440 billion China is increasingly seen as an interesting country for foreigners who want to make portfolio investments, and to facilitate this demand the authorities have opened up avenues for foreigners to invest

in Chinese firms For example, the recent Qualified Foreign Institutional Investors (QFII) initiative allows foreigners to directly invest in Chinese domestic equity stocks, and many U.S., European, and other international funds have taken advantage

of this opportunity Furthermore, the Chinese authorities have granted more domestic firms permission to raise funds from abroad and to list their securities in Hong Kong and in foreign markets Many Chinese firms have taken up these opportunities

[Insert Tables 1-1 and 1-2 About Here]

China is undergoing a transition from a planned economy to a market economy,

in which various market segments (including insurance, banking, and auditing markets) are opening up to foreign investors Although the lower economic costs of production attract substantial capital inflows to the economy, the quality of financial statements is an impediment to foreign investment in, and transactions with, Chinese firms The absence of an efficient legal environment and corporate governance regimes, which are characteristics of many transitional and developing economies, means that the quality of audit services plays an important role in ensuring the proper functioning of the financial reporting systems, and helps safeguard the assets of investors (Fan and Wong 2005) It is therefore highly instructive for foreign and domestic investors and the government to learn more about the quality of financial statements and the mechanisms that are used to ensure quality in such a fast growing emerging economy Previous literature affirms the role of audit quality and auditor independence in China (Chen et al 2001; Gul et al 2003), and provides some evidence of improved auditor independence following regulatory changes (DeFond

et al 2000; Yang et al 2001) However, there is limited research that examines other factors that drive audit quality and auditor independence in China Given the importance of quality auditing services in the Chinese economy, this study hopes to shed some light on these issues

1.2 OBJECTIVES OF THE THESIS

The primary objective of my thesis is to provide a comprehensive study to understand audit quality in China from the perspectives of regulators and the accounting profession Recently, the rising tide of fraudulent financial reporting in major corporations has shocked the public and tarnished the reputation of the auditing profession The first aim of my thesis is therefore to directly measure audit quality by studying audit failures in listed companies Through an in-depth examination of formal regulatory sanctions against auditors, I attempt to show the types of fraud that are more likely to lead to poor audit quality (audit failures), and investigate the extent to which the regulators perceive auditors to be responsible for the detection of these types of fraud

Second, recent policy debates on the regulatory supervision of the accounting profession have raised the concern of the profession, policy makers, and the general public alike There is a growing consensus that regulatory intervention in audit practices, such as the promulgation of the Sarbanes-Oxley Act in the U.S., is necessary to improve auditor independence and audit quality, and, in particular, to restore investor confidence in financial statements Previous studies suggest that auditors that have been criticized by regulators have experienced economic losses in terms of market share, audit fees, and the market value of their client firms (see

Davis and Simon 1992; Dopuch and Simunic 1982; Firth 1990; Moreland 1995; Wilson and Grimlund 1990) Therefore, auditors that suffer economic losses should

be motivated to act more independently to repair their tarnished reputation Recent research suggests that formal sanctions are an effective incentive to improve auditor independence (Shafer et al 1999), but these studies use surveys or experiments to examine the perceived risk of sanctions, and, though they provide some insight, are

no substitute for the measurement of the actual effects of formal sanctions on audit quality and auditor independence Thus, the second objective of my study is to investigate whether regulatory supervision through the execution of formal regulatory sanctions helps to improve auditor independence and audit quality in the Chinese securities market

Finally, there are theoretical models that demonstrate that increased liability protection for auditors results in reduced audit quality (Dye 1995; Chan and Pae 1998) To date, however, there has been little or no research to empirically test this theory In the late 1990s, China underwent an accounting reform that required all accounting firms to disaffiliate themselves from their government-sponsoring agents and form independent CPA firms Independent CPA firms can be registered as unlimited liability partnerships' or as limited liability firms if certain requirements

1 H ereafter, I use the term s “partnership” , "audit partnership”, o r “ partnership firm ” for an unlim ited

are met Since the reform, the levels of auditors’ risk and liability exposure have substantially increased, and the liabilities are not covered or guaranteed by the government This unique institutional characteristic provides an excellent setting in which to empirically investigate whether an auditor’s intrinsic liability exposure that

is dependent on the legal organizational form of the CPA firm affects that auditor’s propensity to issue modified audit reports (which is a proxy for auditor independence and audit quality)

1.3 CONTRIBUTIONS OF THE THESIS

My thesis contributes to the literature in the following ways Previous studies show some evidence of the motivation effect of formal sanctions on audit quality, but their results are largely based on analyses of survey responses in experimental settings My study presents empirical results that are based on the effectiveness of the

actual regulatory sanctions on auditor independence and audit quality Second, my

research is the first to empirically demonstrate the association between the organizational form of a CPA firm and auditor independence

Based on an analysis of audit deficiencies, the results of my study provide a direct measure of poor audit quality, and provide some useful information for the profession to develop various auditing procedures for the detection of egregious

fraud For example, CPA firms could design more comprehensive audit procedures to verify eamings-related items The analysis also gives some insights into the regulator’s views on the responsibility of auditors in detecting fraud This knowledge will be helpful to the accounting profession and policy makers when they devise new accounting and auditing standards that will help uncover these specific types of fraud.

My thesis also has significant implications for policy makers, regulators, and the accounting profession in their continuous effort to structure accounting reforms in China The results on the effectiveness of regulatory sanctions as incentives for auditor independence suggest that more resources should be allocated to regulatory supervision Without effective regulatory supervision, auditors are likely to act less independently, and clients thus have more power over auditors (DeFond et al 2000)

My research also provides some directions for the design of regulatory procedures in other transitional economies Furthermore, the evaluation of the impact of the organizational form of a CPA firm on auditor independence suggests that partnership firms are more independent and report more conservatively than limited liability firms Regulators and policy makers could consider more stringent regulations on the establishment of limited liability audit firms to better control audit quality Accounting professional bodies could also consider the setting up of a peer review

regime to self-reguiate the behavior of auditors in China.

1.4 ORGANIZATION OF THE THESIS

Chapter 2 provides a literature review of audit quality measures and presents the research to date on audit quality in developed countries and in China An investigation into audit quality from the regulator’s perspective is presented in Chapter 3, and an investigation into audit quality from the perspective of the accounting profession is documented in Chapters 4 and 5 Specifically, Chapter 4 examines the effectiveness of formal sanctions as incentives for auditor independence, and Chapter 5 empirically investigates the impacts of the legal form of CPA firms on auditor independence and audit quality Finally, Chapter 6 concludes the thesis, documents the implications of the study, and discusses future research opportunities

As audit quality cannot be observed directly except in ex-post audit failures, previous studies use different proxies as measures of audit quality and investigate the impact of various economic and environmental factors that are associated with audit quality Some examples of these measures are auditor size (DeAngelo 1981; Francis 1984; Francis and Wilson 1988; Johnson and Lys 1990), auditor’s brand name reputation or industry specialization (Francis and Simon 1987; Davidson and Neu 1993; Craswell et al 1995), the issuance of modified audit opinions (Francis and Krishnan 1999; Chen et al 2000), and auditor economic dependence on clients

(Abbott et al 2003; Ashbaugh et al 2003; Frankel et al 2002) Recent research also reveals that audit quality is strongly associated with the credibility of reported earnings (Teoh and Wong 1993), audit price premiums (Francis 1984; Baber et al 1987; Francis and Simon 1987), the resolution of agency costs (DeFond 1992; Fan and Wong 2005), the constraint of earnings management (Becker et al 1998; Francis

et al 1999; Kim et al 2003), the threat of sanction risk (Shafer et al 1999), the propensity of auditors to issue modified reports (Morris and Strawser 1999), and the protection of auditor reputation over economic fee dependence (Reynolds and Francis 2001)

This chapter provides a literature review of audit quality and the various factors that affect audit quality in developed countries and in China The organization of the chapter is as follows Section 2.2 describes some of the common measures of audit quality in the literature, and section 2.3 provides a review of the literature that studies audit quality in developed countries and in China

2.2 VARIOUS MEASURES OF AUDIT QUALITY

A direct measure of audit quality is the ex-post audit failures that reveal “bad”

or poor-quality audits that are supplied by auditors during an audit engagement A

number of studies that examine audit failures involve direct investigations of regulatory enforcement actions (Feroz et al 1991; Rollins and Bremser 1997; Bonner

et al 1998; Carcello and Nagy 2004) and litigation against auditors (St Pierre and Anderson 1984; Palmrose 1988) By and large, these studies reveal a positive association between poor audit quality and audit failures (Palmrose 1988), and report the main problems that result in poor-quality audits (St Pierre and Anderson 1984; Rollins and Bremser 1997; Bonner et al 1998) A common limitation of these analyses is the small numbers of audit failure cases that are examined For example, Palmrose (1988) examines a sample of 472 U.S legal cases that were brought against auditors between 1960 and 1985, which represents just 18 litigation cases per year approximately Feroz et al (1991) summarize 188 firms that were subject to the U.S SEC enforcement actions between 1982 and 1989 Of the 188 enforcement action cases, there were 85 cases where the auditors were sanctioned by the SEC, which represents around 10 enforcement cases against auditors per year

Due to the limited observations of audit failures, researchers have developed various indirect measures of audit quality that are based on technical competence and auditor independence

Auditor size is one of the most common proxies for audit quality Auditor size

can be measured as the audit fees that are charged to clients, the total number of clients, total client assets or sales, which are continuous measures, and Big N versus non-Big N auditors, which is a dichotomous measure DeAngelo (1981) argues that larger auditors (as measured by number of clients) have more client-specific quasi-rents that serve as collateral against losses that are incurred through the discovery of a lower quality audit than was promised This collateral motivates larger auditors to behave less opportunistically, and thus to supply a perceived higher quality audit and to display a higher level of independence Subsequently, many researchers have used auditor size as a measure of audit quality (Francis 1984; Francis and Wilson 1988; Johnson and Lys 1990; DeFond 1992; DeFond et al 2000; Chen et al 2001; Gul et al 2003) Francis and Wilson (1988), Johnson and Lys (1990), and DeFond (1992) use the sales of clients as a proxy for audit quality, because client sales revenues should be highly correlated with client-specific quasi-rents, and greater client sales signify a larger auditor size DeFond et al (2000), Chen et al (2001), and Gul et al (2003) also use the measure of the financial situation of clients as a surrogate for auditor size in China They use the total assets

of clients as a surrogate of quasi-rents, and employ a dummy variable of TOP 10 to indicate whether an auditor is a top ten auditor (large auditor) according to the total assets of its clients

Another example of a measure of auditor size is that of the Big Eight (or Big 7/6/5/4) versus non-Big Eight firms, in which researchers generally argue that Big Eight audit Firms have “more to lose” from discovery of substandard audits, in terms

of loss of reputation or loss of future economic rents from clients (Francis 1984; Baber et al 1987)

Based on the quality differentiation argument of DeAngelo (1981), auditors voluntarily specialize in different (albeit uniform) quality levels for the purpose of capturing higher fees (product differentiation) Big Eight auditors generally have a brand name reputation to the extent that they charge a higher price premium for higher quality audits (Francis 1984; Francis and Simon 1987; Simon and Francis 1988; Davidson and Neu 1993) The brand name reputation effect also motivates auditors to supply high-quality auditing services, as they are subject to the threat of loss of reputation if they provide substandard audits Francis and Simon (1987) analyze the small client segment of the U.S audit market for publicly traded companies, and find that a Big Eight price premium exists relative to non-Big Eight auditors, which are characterized by national firms or local/regional firms Other research studies similarly report that Big N auditors charge higher premiums These findings imply the existence of Big N product differentiation in the market for audit

services The premiums are consistent with positive returns to Big N investment in brand name reputation.

2.2.4 Industry Specialization

In addition to the effects of the Big Eight brand names on audit quality, Craswell et al (1995) reveal that, within Big Eight product differentiation, industry specialist Big Eight auditors earn a price premium over non-specialist Big Eight auditors They argue that for industries with specialized contracts and accounting technology, industry specialist auditors may have more specialized industry knowledge that constitutes a significant component of auditor technical expertise, and therefore will supply higher quality audits compared to non-specialist auditors Their findings imply that industry specialization increases a Big Eight auditor’s reputation within that industry Industry specialist auditors need to make additional investments in expertise beyond general expertise, and thus supply higher quality audits, and require a return that results in a price premium

DeAngelo (1981) defines audit output as the independent verification of management-prepared financial data and the production of a stated opinion with an associated quality dimension “A true and fair view” should be presented on audited financial statements, and audit quality and auditor independence is the key

determinant in presenting this view Investors assess both the information content in auditor reports and the attached audit opinions (Chen et al 2000) Previous studies have documented the relevance of audit outputs in an audit engagement (Haw et al 2003; Carcello and Palmrose 1994; Gaeremynck and Willekens 2003; Chen et al 2001; DeFond et al 2000) For example Haw et al (2003) show that the audit opinion is an important determinant of the timeliness of annual eamings announcements, and that firms with modified audit opinions delay the announcement

of both positive and negative eamings surprises Chen et al (2000) reveal that audit modifications are associated with a negative market reaction in China, which implies that, from the perspective of Chinese investors, audit outputs have valuable informational content In addition, Carcello and Palmrose (1994) and Gaeremynck and Willekens (2003) document the role of modified reports that are issued before bankruptcy in protecting auditors from subsequent litigation if bankruptcy finally occurs

Recall that an important characteristic that is correlated with an auditor’s willingness to report financial statement irregularities is auditor independence (DeAngelo 1981) An independent auditor is less willing to acquiesce to client pressure to issue clean reports when it is inappropriate to do so Therefore, a greater degree of auditor independence is consistent with increased audit quality (DeFond et

al 2000), and recent research studies have used an auditor’s propensity to issue modified reports (Carcello and Palmrose 1994; DeFond et al 2000; Yang et al 2001)

as a measure of auditor independence In conclusion, the assessment of audit output may help to better understand audit quality and auditor independence

Other measures of audit quality and auditor independence (dependence) are the ratio of non-audit fees to total fees received from a client (Abbott et al 2003; Ashbaugh et al 2003; Frankel et al 2002) and the ratio of client-specific audit fees to total audit fees received by the auditor2 (Craswell et al 2002) The argument is that the greater the dependence on clients for fees (including non-audit fees), the less willing the auditor will be to disclose a breach for fear of losing the future revenue from that client Contrary to the view of regulators, previous research on the alleged association between non-audit service fees and impaired auditor independence has been inconclusive For example, Craswell et al (2002), DeFond et al (2002), and Geiger and Rama (2003) investigate the impact of non-audit fees on the propensity of auditors to issue modified audit reports, and their results show no significant association between the non-audit service fees that are received by auditors and impaired auditor independence Ashbaugh et al (2003) draw similar conclusions, and

2 B efore the availability o f audit fee data, researchers used the ratio o f a single c lien t's total assets to the total assets o f all o f the clients as a proxy for auditor independence (D eFond 1992; R eynolds and Francis 2001).

find little evidence to support the claim that auditors violate their independence as a result of receiving high non-audit fees from clients In contrast Firth (1997) and Frankel et al (2002) document that the joint provision of non-audit services by auditors may impair auditor independence.

2.3 PREVIOUS RESEARCH ON AUDIT QUALITY

The recent studies that examine audit quality in developed economies can be classified into two main streams: the demand for audits of a certain quality and the supply of audits of a certain quality As regards the demand for auditing services of a certain quality, Johnson and Lys (1990) show that the determinants of a change of auditor are related to changes in the operations and activities of client firms over time Firms that have a poor operating performance (relative to the previous periods) are more likely to change their auditor to one that has relatively few specialized resources and therefore charges a lower fee In other words, auditor-client realignments can generally be attributed to cross-temporal changes in client characteristics and differences in the cost structures of audit firms Other factors that affect auditor choice include potential agency conflicts (Francis and Wilson 1988;

DeFond 1992; Fan and Wong 20053) and firm-specific risk factors (Clarkson and Simunic 1994; Copley and Douthett Jr 2002) In particular, Francis and Wilson (1988) and DeFond (1992) provide empirical evidence that firms with higher agency costs have greater demand for high-quality auditing services, because high-quality auditing services serve as a monitoring mechanism for the resolution of costly agency problems (Jensen and Meckling 1976; Watts and Zimmerman 1986) Clarkson and Simunic (1994) and Copley and Douthett Jr (2002) show that firms with a higher firm-specific risk are more likely to employ high-quality auditors.

As audit quality is costly to evaluate, clients compensate auditors who enable them to avoid these costs by specializing in a relatively uniform quality level (DeAngelo 1981) On the other hand, the existence of clients with differential agency costs may result in a demand for auditing services of different levels of quality This implies that auditors have an incentive to specialize in different (albeit uniform) quality levels to capture higher fees By and large, high-quality auditors are associated with more credible eamings (Teoh and Wong 1993), are more independent

in the sense that they issue more modified reports if it is appropriate to do so (Morris

3 Fan and W ong (2005) exam ine w hether external auditors serve as m onitoring and bonding

m echanism s to alleviate agency costs in em erging m arkets (excluding C hina) T h ey find that firms

w ith concentrated ow nership structures (that is, higher agency conflicts betw een controlling ow ners and m inority shareholders) are m ore likely to em ploy high-quality auditors (B ig 5 accounting firms), and that these high-quality auditors generally charge higher audit fees and are m ore likely to issue

m odified audit reports T heir study provides evidence for the dem and for qu ality audits that is sim ilar

to that w hich is reported by Francis and W ilson (1988) and DeFond (1992).

and Strawser 1999), allow less aggressive eamings management (Davidson and Neu 1993; DeFond and Jiambalvo 1993; Becker et al 1998; Kim et al 2003; Francis et al 1999; Bartov et al 2001), and charge higher fees in connection with product differentiation (Francis and Simon 1987; Francis 1984; Craswell et al 1995; Ferguson and Stokes 2002).

Teoh and Wong (1993) find that the eamings response coefficients (ERC) of the clients of the Big 8 accounting firms, which are a proxy for high-quality auditors, are statistically higher than the coefficients for the clients of non-Big 8 accounting firms This implies that high-quality auditors are associated with more credible perceived reported eamings Morris and Strawser (1999) argue that Big 6 auditors are more likely to issue modified opinions to reflect going-concem uncertainties than non-Big

6 auditors, because they are more independent and have more to lose from future

economic rents and potential litigation

Because of the presence of agency problems, managers may be tempted to manipulate eamings figures to maximize their own interests, rather than the firm’s interests As documented in the previous section, auditing plays an important role in making the interests of owners and managers converge Davidson and Neu (1993) argue that management has an incentive to minimize the difference between forecasted and actual income for income smoothing purposes They find that

high-quality auditors (of a larger size) are less likely to allow less opportunistic management behavior, and are associated with larger forecast errors Similarly, DeFond and Jiambalvo (1993) show that Big 8 auditors have greater auditor-client disagreements over income-increasing accounting methods In addition, previous studies generally demonstrate that higher quality auditors are associated with lower discretionary accruals than low-quality auditors, where discretionary accruals are a proxy for the opportunistic behavior of the management of the client firms Becker et

al (1998) and Francis et al (1999) show that Big 6 auditors are associated with lower levels of discretionary accruals, which is consistent with the notion that high-quality auditors allow less flexibility in accounting treatments by the management of client firms Kim et al (2003) refine the findings by studying whether high-quality auditors are more effective than low-quality auditors in deterring eamings management cases when clients have an income-increasing incentive to manage eamings Their results show that Big 6 (non-Big 6) auditors are more effective than non-Big 6 (Big 6) auditors in constraining aggressive eamings management when managers have the incentive to prefer income-increasing (income-decreasing) accrual choices Bartov et al (2001) use the issuance of modified audit opinions as a proxy for audit quality, and investigate the ability of auditors to constrain eamings managements Overall, these findings suggest that a

high-quality auditor is more likely to constrain the aggressive and potentially opportunistic reporting of accruals, which in turn, lends greater credibility to financial statements.

High-quality auditors also charge higher fees than low-quality auditors Using Australian data, Francis (1984) reports that Big 8 auditors have significantly higher audit prices than non-Big 8 auditors, regardless of client size Francis and Simon (1987) also find that a Big 8 price premium exists for U.S publicly traded firms relative to national and regional non-Big 8 firms Craswell et al (1995) document that Big 8 industry specialists charge a 34 percent premium over Big 8 non-industry specialists, and that, in general, Big 8 auditors earn a 30 percent premium over non-Big 8 auditors However, Ferguson and Stokes (2002) examine the same issue in the post-merger years of accounting firms, and indicate that the findings of Craswell

et al (1995) no longer hold after the Big8 / 6 audit firm mergers

Recent corporate scandals, including those of Worldcom and Enron, have re-ignited the concern of regulators that non-audit fees may compromise auditor independence Chaney and Philipich (2002) find that the clients of Arthur Andersen suffered an economic loss because of the shredded reputation of the auditor Questionable auditor independence behooves researchers to study whether non-audit service fees compromise auditor independence While Frankel et al (2002) report a

positive association between non-audit service fees and impaired auditor independence (as measured by an auditor’s propensity to issue modified audit opinions and allow discretionary accruals), DeFond et al (2002), Geiger and Rama (2003), and Ashbaugh et al (2003) find no association The results are consistent with the finding of Reynolds and Francis (2001) that the protection of reputation has

a more important effect than the dependence on economic fees on the reporting behavior of auditors Higher quality auditors are more likely to report independently because of the threat of loss of reputation

The rapid emergence of the accounting profession in China was motivated by the economic reforms of the early 1980s Before the economic reforms, auditing played a very limited role in China, and the profession had a low status because it served as a government tool to ensure that the correct tax revenues were collected for the central government (Tang et al 1999; Xiao et al 2000; Lin and Chan 2000; Li and He 2000; Chen et al 2001; Yang et al 2001) Driven by the transformation of the planned economy into a market economy, the economic reforms resulted in the corporatization of state-owned enterprises and fast-growing foreign investment levels that created a high demand for external audits (DeFond et al 2000) Since then, the accounting profession in China has experienced rapid development that is driven by

an increasing demand for independent and high-quality audits Despite the rapid growth in the economy, China is undeigoing a series of accounting and economic reforms, such as the recent entrance into the World Trade Organization (WTO) Alongside the expansion of the stock markets, the accounting profession has undergone rapid development, such as the establishment of the Chinese Institute of Certified Public Accountants (CICPA) in 1988, the disaffiliation program, in which CPA firms were severed from their sponsoring government agencies, in early 1997, and the gradual harmonization of domestic and international accounting and auditing standards (Lin and Chan 2000; Yang et al 2001).

Table 2-1 provides a description of the Chinese accounting profession in terms

of the firms that audit listed companies As is shown in Panel A, there were seventy-two CPA firms in 2004 that had licenses to audit the financial statements of listed companies4 Panel B reports the opinions that were issued by the auditors of listed companies from 1993 to 2004, and shows that the numbers and percentages of

audit modifications have gradually increased in recent years Ceteris paribus, high

audit quality is associated with a high level of auditor independence in the issuing of modified audit opinions (Morris and Strawser 1999; DeFond et al 2000) The increase in modified audit opinions provides some evidence for greater auditor

independence in China in recent years.

[Insert Table 2-1 About Here]

Recent studies that focus on audit quality and auditor independence in China provide empirical evidence that supports these observations For example, DeFond et

al (2 0 0 0) investigate the impact of the adoption of new auditing standards on auditor independence Their results reveal a nine-fold increase in modified audit opinions (MAOs), which are a proxy for auditor independence, following the adoption of the new standards in 1996 However, the increase is followed by a decline in the market share of large auditors, as clients with qualified opinions tend to shop for unqualified opinions from smaller auditors This flight from quality audits is a result of the lack

of incentive to demand independent audits Yang et al (2001) examine whether the disaffiliation between Chinese CPA firms and their sponsoring government agencies improves auditor independence They find that the number and percentage of MAOs issued have increased dramatically since the implementation of the program in 1997 Chen et al (2001) also examine auditor independence, and link it with eamings management by investigating whether auditors are able to detect discretionary eamings manipulation by management Their results show that managers engage in significant eamings management to meet regulatory target profitability levels, and that their opportunistic accounting choices are associated with an increased

frequency of MAOs.

Previous studies examine the impact of external changes (environmental or

institutional changes) on audit quality and auditor independence in China, and their results generally report an improvement in audit quality in the past decade (DeFond

et al 2000; Yang et al 2001; Chen et al 2001) However, despite these improvements in audit quality, there are still a substantial number of reports of fraudulent financial statements (Wu 2002; Xuan 2002; Li 2002)

Previous studies also show some evidence of quality differentiation, the perceived importance of audit quality, and the information content of auditor independence (using the issuance of modified audit opinions as a proxy for independence) in China DeFond et al (2000) and Qiu (2004) report a significant association between larger auditor size (top 1 0 auditor in terms of the total assets of clients) and audit quality, in which audit quality is measured by an auditor’s propensity to issue modified audit reports (DeFond et al 2000) or an auditor’s ability

to constrain the aggressive eamings management practices of clients (Qiu 2004) Gul

et al (2003) examine the market reaction to the changes in eamings of firms that are listed on the Shanghai stock exchange Their findings show the market reaction to an increase in eamings to be stronger for firms that are audited by larger auditors (top

10 auditors), which demonstrates the perceived importance of audit quality in China

Investors may also respond to the opinions that are issued by auditors Chen et al

opinions, which reveals that modified audit opinions have an information content in the opinion of investors5 Chen et al (2001) also examine the role of independent auditors in the detection of eamings management in China Similar to findings in the U.S (Bartov et al 2001), they find that Chinese firms that engage in eamings management are more likely to receive modified audit opinions

By and large, the literature in China reveals the importance of audit quality and auditor independence during the development phase of the economy A comprehensive study of audit quality and the factors that are associated with audit quality is needed to gain a better understanding of the institutional development in this fast-emerging economy The regulatory environment is changing to accommodate the needs of business and the development of the accounting profession Two examples of such changes are the increased scrutiny of the accounting profession by regulators and the disaffiliation of state-owned CPA firms into either partnership firms or limited liability firms The aim of introducing certain changes to the profession is to increase the quality of audits and auditor independence However, whether the changes have been effective remains an

5 H aw et al (2003) docum ent sim ilar results for the inform ation content o f m odified reports They show that firm s that receive m odified opinions are more likely to d elay the tim ing o f th eir annual eam ings announcem ent because o f client-auditor disputes over the accounting treatm ent.

under-explored empirical issue The unique institutional environment in China therefore allows me to examine some unresolved issues in auditing practice.

by the China Securities Regulatory Commission against auditors for fraudulent financial reporting committed by listed companies in China I find that auditors are

more likely to be sanctioned by regulators for failing to detect and report material

misstatement fraud than for failing to detect and report disclosure fraud Further

analysis of these material misstatements indicates that auditors are more likely to be

sanctioned for failing to detect and report revenue-related fraud than for failing to detect and report asset-related fraud In sum, my research results suggest that regulators believe that auditors are responsible for the detection and reporting of fraud that is egregious, transaction-based, within audit scope, and related to accounting eamings The results contribute to the knowledge on the responsibility of auditors for the detection of fraud as perceived by regulators