Does Non-Audit Service (NAS) impair the
Auditor independence? Evidence from
Chinese auditing market
Yu Bai 10602798 August 10, 2014
MSc Accountancy & Control, Accountancy Amsterdam Business School
Faculty Economics and Business University of Amsterdam
Supervisor: Dr. Georgios Georgakopoulos Second Reader: Dr. Rui Vieira
Table of Contents
Abstract ... 3
1. Introduction ... 4
2. Literature Review and Hypothesis Development ... 5
2.1. Auditor independence ... 5
2.2. Non-audit services (NAS) ... 7
2.3. Audit market in China ... 9
2.4. Hypothesis development ... 11 3. Research Methodology ... 13 3.1. Data selection ... 13 3.2. Research design ... 13 4. Results ... 15 4.1. Descriptive statistics ... 15
4.2. Descriptive statistics by opinion type ... 16
4.3. Multivariate test results ... 17
4.4. Robustness checks ... 18
5. Conclusion ... 18
Abstract
This paper presents the results of an empirical research carried out in China’s context to add some evidence of China’s audit reality to the discussion of whether the independence of audit service is impaired by the non-audit service or not. China has its unique law and market environment which differentiate it from other western countries. The question is investigated by analyzing the audit reports for a sample of financial distressed firms listed on Shenzhen and Shanghai Stock Exchange over the period of 2002 to 2013. According to my research, there is no significant impact of non-audit service on auditor independence in China’s context.
1. Introduction
The high profile accounting scandals taking place recent years not only draw significant public attention to the reliability of financial statements, but also raise questions on auditors’ independence when they give opinions on companies’ reports. Among those scandals, Enron’s collapse in the year 2003 apparently was the biggest strike. Arthur Andersen, which was the external auditor of the company at the moment, could not claim to be innocent in this case. Not to mention the problem existing in regulation, the construct of service Arthur Andersen provided to Enron was of great concern --- in the meantime being the external auditor of Enron, Arthur Andersen also acted as the legal advisor, business advisor, and internal auditor of the company. In other words, Arthur Andersen performed both audit and non-audit service (NAS) to the same company. Tracing back to early 80’s, experts have been arguing that the mechanism will impair auditors’ independence when auditing a company. Watts and Zimmerman mentioned in their paper (Watts and Zimmerman, 1981) that in the United States independence was impaired particularly when auditors supply non-auditing services (NAS). Simunic (1984), Palmrose (1986) and Wines (1994) also found that under conditions of higher audit and NAS fees, auditors were not perfectly independent.
It has been suggested, locally and internationally, that the way to enhance auditor independence is to prohibit the incumbent auditor from providing non-audit services. As a consequence, standard setters successively issue related regulations to constrain the provision of non-audit service of auditing firms on the same company they provide audit service. Legislation in US and UK require the disclosure of the level of NAS as an implication to assess the independence. To take the step further, Italian regulations prohibit the provision of NAS to audit clients. And severe restrictions have also been implemented in France, Germany and the Netherlands on this problem (Sharma and Sidhu, 2001).
However, opinions do diverge on this issue over years of discussion and research. While some argue that rendering of non-audit services impair the auditor’s independence in both fact and appearance, Barkess and Simnett (1994) and Craswell (1999) show that auditors do not compromise their independence. “Based on publicly available information for Australian listed companies for several years, the evidence suggests that auditors’ decisions to qualify their opinions are not affected by the provision of non-audit services” (Craswell, 1999). And some researchers in favor of the joint provision of audit and non-audit services point out that the NAS also helps auditors get access to more information of the firms they are auditing. An important implication of the model of Simunic (1984) is that the potential for increased
efficiency and lower client fees resulting from knowledge spillovers associated with the joint provision of audit and other services.
Though being discussed for a long time, there is no exact conclusion of whether the independence of audit service is impaired by the non-audit service or not. This paper is not aiming to give a conclusion to this question, but to add to the discussion some evidence of China’s audit reality. According to my research, there is little discussion on this topic in Chinese auditing market. I believe the study of this topic is important for the future auditing regulation in China, where there are still remarkable loopholes in related laws because of the immature audit market. Also I hope it will contribute to the discussion among standard setters on whether NAS provided by the same auditing firm should be prohibited or not. Following DeFond et al. (2002) and Robinson (2008), tests are undertaken to identify any association between the provision of non-audit services and auditors’ reporting opinions.
This paper is structured as follows. The next section contains a discussion of previous research, an examination of the nature of auditor independence and non-audit services (NAS), and an introduction of Chinese audit market. Data characteristics and research design are then discussed followed by empirical results, conclusions.
2. Literature Review and Hypothesis Development
2.1. Auditor independence
Independence is considered as an important attribute of external auditors. Yet the phrase “auditor independence” traditionally has had no precise meaning. As critiqued in Antle (1984), the AICPA and SEC rules on auditor independence are lengthy and subject to constant reinterpretation, and both bodies have abandoned attempts to provide a concise definition. As a consequence, there are many definitions of auditor independence according to different academic literature. Although stating itself that “independence is not susceptible of precise definition” (Code of Ethics, 1062), AICPA (1979, p. 26) offers a general guide of auditor independence which is “the ability to act with integrity and objectivity”. That is, whether an auditor can withstand the pressure from the management, and still gives the opinion that does not favor the company. To make the concept more visible and measurable, DeAngelo (1981) regards the level of auditor independence as the probability of auditors reporting errors conditional upon an error being discovered.
There are two kind of independence: independence in fact and independence in appearance. Auditors should not only be independent in fact but also be independent in appearance. According to Corless and Parker (1987), independence in fact refers to
the “mental attitudes” of the auditors. It is based on the auditors’ actions in a situation or real independence. And independence in appearance, on the other hand, is what is believed by society including investors, government officials and the general public that an auditor is independent (Titard, 1971). It is what mentioned in Craswell (1999) as perception of independence as well. Independence in appearance is based entirely on perceptions. Actually it has become a surrogate for independence in fact, because it can be observed while the latter may not be. Though might be imperfect, the public’s perceptions are its only practical measure of auditor independence.
Auditors have incentives to maintain their independence, even in the absence of governmental regulations. In perspective of the exogenous demand, Jensen and Meckling (1976) conclude that managers tend to hire independent auditors in order to reduce agency costs. On the other hand, a large body of theoretical and empirical research suggests that auditors have market-based institutional incentives to act independently. In particular, loss of reputation and the litigation costs are likely to provide strong incentives for auditors to maintain their independence. Watts and Zimmerman (1983) find evidence that 84% of New York Stock Exchange (NYSE) companies voluntarily engaged independent auditors several years before the Securities Acts that mandated external auditing in 1926. And a more recent example to illustrate the reputation concerns is the client losses of Arthur Andersen’s in the months following the Enron collapse.
Although there are market-based incentives for auditors to remain independent, there are also forces that potentially threaten auditor independence. In general, any situation which increases the probability that an auditor will not truthfully report the results of his audit investigation can be viewed as a threat to independence (Simunic, 1984). Having a financial interest in the client, having a family relationship with employees, managers or the owners of the client, auditing work that was originally completed by the auditors are some of the situations where auditor independence might be hurt. In the scandal of Enron, the diminished independence of auditor, Arthur Andersen, was often linked to the provision of non-audit services. Enron’s financial statement for 2001 indicated that Arthur Andersen earned more fees for non-audit services than for audit fees. The reason that the magnitude of the fees they generate relative to audit services are getting so much public attention is because regulators believe that non-audit service fees make auditors financially dependent on their clients, and therefore less willing to stand up to management pressure for fear of losing their business. Academically, DeAngelo (1981), Simunic (1984), Beck et al. (1988) also express the concern that the provision of non-audit services creates economic bonds that weaken an auditor’s independence. And the consulting nature of many non-audit services puts auditors in managerial roles, potentially threatening their objectivity about the transactions they audit.
2.2. Non-audit services (NAS)
Non-audit services are any professional services provided by a qualified public accountant during the period of an audit engagement which are not connected to an audit or review of an institution's financial statements. Non-audit services (NAS) are also sometimes referred as management advisory services (MAS). So the two terms will be used interchangeably in this paper. An increasing number of clients are purchasing other services from their auditor. The sample in Barkess and Simnett (1994) shows that 85% of companies purchased non-audit service and a steady average increase in the percentage of clients purchasing NAS from their auditor each year is identified.
The addition of NAS as an auditor's practice stimulated considerable debate on its social, personal, and economic effects on auditor independence. As Watts and Zimmerman (1981) found out, the charges on auditors’ not being independent taking place around that time are the consequence of assuming an expanded role for the auditor. The Metcalf Committee Staff Study, which is the U.S. Congressional committees and their staff reports, suggested that supplying both services can create a conflict of interests, particularly when a CPA firm recruits a client’s executives and is interested in assuring their success, or when it installs a management information system and subsequently audits the reliability and accuracy of its own work. In these situations, a CPA firm acting as both auditor and consultant may be motivated not to report consulting deficiencies observed during the audit, thereby avoiding erosion of its consulting “brand name”. In terms of legislation, the original provision of the Corporate Laws Amendment Bill in clause 29 specified certain non-audit services that would not have been open to the current auditor of a public interest company. It reads as follows: ‘An auditor appointed to a public interest company may not for the duration of the appointment, perform any book-keeping or accounting (as distinct from auditing) services, and to the extent that these would be subject to its own auditing, internal audit or tax advisory services, for that company.’” In addition, in January 1999, the Independence Standards Board in the US adopted Independence Standard No. 1, Independence Discussions with Audit Committees. The standard recommends that auditors communicate in writing with the audit committee matters likely to influence audit independence. Specifically, the standard requires auditors to disclose the level of audit fees, and nature and level of NAS fees derived from the client. In the UK, legislation requires companies to disclose the level of NAS fees so that interested parties can assess independence implications. And regulators in Europe and Australia have also voiced their concern that the provision of NAS to audit clients poses threats to auditor independence.
the effects of NAS on auditor independence. However, as stated in Sharma and Sidhu (2001), while regulatory concerns prima facie appear appropriate, the lack of consensus in the literature does not provide it strong empirical support.
Wines (1994)’s finding suggests that auditors are less likely to qualify a given company’s financial statements when higher levels of non-audit services fees are derived. This means that there is potential for auditor independence impairment when higher levels of non-audit services are provided to audit clients. However, contrary to that, the results of Barkess and Simnett (1984) and Craswell (1999) show that NAS does not impair auditor independence. Also, the view taken by the Cohen Commission, an independent commission formed by AICPA, on the potential threats to truthful reporting by auditors engaged in NAS stated that it failed to find “any significant relationship between the provision of management services (i.e. NAS) and substandard audits.”
Furthermore, the Cohen Commission emphasized that there may be efficiencies associated the joint supply of the two services. For example, when NAS can be purchased from the auditor, a client’s search costs for a credible consultant, and other transaction costs, should be reduced. Agree with that, Simunic (1984) points out there may be a knowledge externality, known as well as knowledge spillover. Palmrose (1986) also mentions the effect of knowledge spillover of providing NAS. Knowledge spillovers refer to information generated while performing management consulting services that can produce economic rents by reducing auditing costs. The production of auditing generates knowledge useful in NAS or the production of NAS decreases the marginal cost of auditing, since each function requires knowledge about a company’s operations, its industry, etc. Francis (2004) suggests that the one area of non-audit service in which knowledge spillover would seem plausible is the “joint preparation of tax returns and the audit of tax-related accounts in the financial statements.” Empirically, consistent with knowledge spillover, Kinney et al. (2004) provide some evidence that financial reporting quality (measured as firms’ propensity to restate previously filed financial statements) is higher for firms that purchase larger amounts of tax services from the incumbent auditor.
However, the views and studies discussed above are all based on the research on developed countries. They might not have the same explanation power for developing countries, especially for China which is still under the transition from a planned economy to a market economy. The next subsection is devoted to introduce some background of Chinese audit market.
2.3. Audit market in China
China is currently on the course to become one of the world’s largest economies. However, the accounting profession in China is much younger and inexperienced compared to most western countries. The Chinese Institute of Certified Public Accountants (CICPA) was founded in November 1988. This is the first professional accounting body founded in China since the establishment of the PRC in 1949.
Over the last decade, China has experienced tremendous social and economic changes. China has made progress in creating a viable capital market in an economy that until very recently was governed entirely by non-market forces. The development of socialist market economy, privatization, and large inflows of foreign investment demand the innovation of a Chinese accounting system. And the government also recognized the need for an external auditing profession after the establishment of stock enterprises and the opening of stock exchanges in Shanghai and Shenzhen in 1990 and 1991, respectively. In Chan and Lin (2000), they point out that primary function of auditing in China has begun to shift away from the traditional tax compliance assessment towards the credibility lending to financial statements.
However, due to lack of capital, the new accounting firms affiliated themselves with existing institutions and three types of auditing firms emerged: government-affiliated audit firms, university-affiliated audit firms, and audit firms that are joint-ventured with an international CPA firm. The government-affiliated firms are by far the dominant group with currently over 75% of the audit market in terms of number of clients. According to Hao (1999), the large proportion of government-affiliated auditors may be partially due to the government’s desire to maintain control of the economy. DeFond et al. (2002) conclude that NAS provided by the auditor do not impair auditor independence based on the belief that the benefit for auditors to maintain independent outweigh the cost to compromise independence. However, the situation might be more complex in China. As stated above, the main incentives for audit firms in western developed countries are reputation loss and litigation cost. The market mechanism and the regulation play important role in it. As for China, the market is not as developed as that in western countries. Aside from government regulations that require listed companies to be audited, there are few institutional features in the Chinese capital markets that provide incentives for managers to demand independent auditors. DeFond et al. (2000) attribute the most significant impediment to the demand for auditor independence in China to the controlling share of government-related entities in virtually all listed companies. Government entities effectively own more than 50% of the stock of 85% of the listed companies. The government has the final say about the operation of the firms. And the firms have ways to get financial support from government, so the financial source
from investors is not that important relative to western countries. They do not have an incentive to maintain a highly appreciated reputation in market. On the other hand, audit committees are nonexistent in China and there is no requirement that independent outside members sit on boards of directors (DeFond et al., 2000). There are virtually no cases of shareholder litigation against Chinese auditors. Without the threat of costly shareholder litigation, auditors have less incentive to avoid audit failures, and shareholders have less incentive to discover them. So it is reasonable to conclude that the influence of provision of NAS on auditor independence is different from that in western countries.
Looking on the bright side, although state-owned enterprises are still the dominant part of the Chinese economy, as a result of the economic reforms, collective and private enterprises and foreign investment enterprises coexist and compete with state-owned enterprises. There are over 300,000 private enterprises and 320,000 foreign investment enterprises in operation in China by the end of October 1998 (Beijing Review, 1998). Estimates in National Bureau of Statistics (2000) indicate that about 73 percent of industrial output was generated outside the state sector in 1999. The movement in China towards private ownership that is almost totally divorced from management requires an independent audit on the financial reports made by management. If investors are to have confidence in the veracity of the financial representations of management, it is necessary to prove an independent opinion on the truthfulness and fairness of the reports.
In addition, the largest government-affiliated auditors are also expected to face stronger incentives to behave independently in response to the new standards. The following quote well explaining the independence demand Chinese central government is facing is from DeFond et al. (2000):
“The central government wishes to increase auditor independence and is grooming the largest Chinese audit firms to someday compete with the Big 5 (in current case, Big 4). The high visibility of the largest government-affiliated auditors means that audit failure among this group would be particularly embarrassing to the central government and hence we expect their behavior to be more closely monitored. In addition, in the event of an audit failure, the central government is able to exert costly career-related penalties on the government agency bureaucrats that are associated with the largest government-affiliated auditors.”
Furthermore, the large amount of growth of the foreign direct investment in China since 1993 had driven the demand of more independent and higher quality audit work. Therefore, more leading international accounting firms have moved into the country by opening representative offices (Beijing Review, 1999). While direct competition from international auditing firms is currently prohibited in China, the
government intends to gradually open up the domestic stock markets to international investors and allow competition between large Chinese auditors and the Big 4, which have high proportion of clients with foreign owners. The presence of foreign shareholders provides an incentive for the international accounting firms to act independently in order to protect their reputation in international capital markets. This also adds positive elements to Chinese audit market.
And as stated in Chan and Lin (2000),while some significant differences exist, the new Chinese auditing standards are, in a number of important aspects, similar to the professional standards promulgated by the International Federation of Accountants. China recently adopted new auditing standards in an attempt to increase credibility in its capital markets. Although there is no prohibition on the provision of NAS in China, in order to binding the behavior of certified accountants and therefore increase the audit quality, in 2001 China Securities Regulatory Commission (CSRC) enacted codes requiring the detailed disclosure of remuneration paid to the auditors for public listed companies, including both audit and non-audit fee. In DeFond et al. (2000)’s research, subsequent to the adoption of the new standards, frequency of modified pinions increases nine-fold. This point of time is also chosen in my paper to justify the time frame selection of data.
2.4. Hypothesis development
In previous research, there are different ways researchers study the relationship between provision of non-audit service and auditor independence. As discussed in Craswell (1999), existing research on the provision of NAS by auditors focus on the following parts: the economics of the joint supply of NAS and auditing; third party perceptions of the effect of NAS on auditor independence; and direct tests of an association between NAS and auditors’ internal control evaluations and reporting decisions. And DeFond et al. (2002) mentions the work done by Chung and Kallapur (2001), Francis and Ke (2002), Frankel et al. (2002), and Reynolds et al. (2002) adopting factors commonly associated with earnings management, such as discretionary accruals and managers’ propensity to meet earnings targets, as indicators of auditor independence to empirically investigate whether non-audit services threaten auditor independence. However, the results of these literatures are not compelling because of potential internal validity threats. The research findings in the studies of the economics of the joint supply of NAS and auditing were limited by the absence of publicly available information; and it is considered indirect and inconclusive for studies on the third party perceptions. And the results from the accrual earnings management research yield mixed support for the contention that non-audit services impair auditor independence.
A number of other studies (Sharma and Sidhu, 2001; DeFond et al., 2002; Geiger and Rama, 2003) utilize the going-concern decision as indicator of auditor independence. Following them, this paper is going to directly test the impact of NAS on auditor independence by investigating the auditor’s willingness to issue a going concern audit opinion (also known as modified opinions). According to DeFond et al. (2000), an independent auditor is more resilient to client pressure to issue a clean audit report when a modified report is appropriate. This approach permits a more objective investigation of the effects of NAS on auditor independence. It is more direct and less noisy.
The audit report communicates the auditor’s findings to market participants and plays a crucial role in warning financial statement users of impending going concern problems. Statement of Auditing Standard (SAS) No.59 (AICPA, 1988) requires auditors to issue going concern modified audit opinions when substantial doubt exists regarding the client’s ability to continue as a going concern for one year beyond the financial statement date. Issuing a going concern opinion, however, means that the auditor must be able to objectively evaluate firm performance and withstand any client pressure to issue a clean opinion. This suggests that, ceteris paribus, the auditor’s propensity to issue a going concern opinion is positively correlated with the auditor’s level of independence. As introduced in previous section, DeAngelo (1981) regards the level of auditor independence as the probability of auditors reporting errors conditional upon an error being discovered. Therefore, when a firm is going to file for bankruptcy, and its auditor fails to issue going concern opinions before the bankruptcy, it can be regarded as an impair of auditor independence (Blacconiere and DeFond, 1997; Weil, 2001). As for the level of NAS provision, Simunic (1984), Palmrose (1986) and Wines (1994) found that under conditions of high NAS fees, auditors were not perfectly independent. And regulators in the US, UK, Europe and Australia have voiced their concern that the provision of NAS to audit clients poses threat to auditor independence parallel to exponential increases in the proportion of NAS fees to total fee revenue of accounting firms. So I use the NAS fees as a proxy for the provision of NAS. If regulators’ concerns are justified, NAS fees will be inversely related to the probability of auditors’ issuing going concern audit reports (DeFond et al., 2002).
Thus, following this argument and the methodology of DeFond et al. (2002), I test the following hypothesis in China’s context:
Hypothesis: Ceteris paribus, non-audit service fees are inversely related to auditors’ propensity to issue going concern audit opinions.
3. Research Methodology
3.1. Data selection
As in prior research, I limit the analysis to a sample of financially distressed firms because the going concern problem is a more salient decision among this group. Similar to Reynolds and Francis (2000), I define financially distressed firms as firms that report negative cash flow increment during the current fiscal year. I get access to sufficient data in CSMAR database of all 88,563 companies (excluding financial institutions) listed on the Shanghai and Shenzhen stock exchanges during the period 2002-2013. The GSMAR is the biggest database devoted in economic and finance data in China. In 2001, China issued new auditing standards. And the choice of 2012 as the starting year can effectively avoid the difference cause to different companies by the implication of the new code. By filtering out the companies without negative cash flow during the 12 years, I obtain a sample of 10,294 firm-year observations, including 569 going concern opinions. The large number of observations (10,294 in total) and the fact that the information used has been taken from published financial reports that have been externally verified should significantly increases the confidence that can be placed in the results.
3.2. Research design
Follow the going concern model used in DeFond et al. (2002) testing for an association between the provision of non-audit services and auditors’ reporting opinions, this paper is going to run regression of auditors’ reporting decisions to a financially distressed client against measures of non-audit service fees, the log of non-audit service fees and the ratio of non-audit service fees to total fees (fee ratio), and control for factors that have been shown to influence the reporting decisions. Some additional explanatory variables are also used based on Robinson (2008) and adjusted according to China’s practical situation.
The model is as follow:
OPINION = β0 + β1(log(ASSETS)) + β2(LEV) + β3(CLEV) + β4(INVESTMENTS) + β5(BIG 4) + β6(OP CASH FLOW) + β7(log(NON-AUDIT FEE)) + β8(FEERATIO) + ε
OPINION
= an indicator variable equal to 1 for firms received a going concern audit opinions prior to the bankruptcy filing, and 0 otherwise
log(ASSETS) = natural logarithm of total assets at the end of the year
measured in millions of dollars
LEV = total liabilities over total assets at the end of the fiscal year
CLEV = change in LEV during the year
INVESTMENTS = the sum of cash and short- and long-term investment
securities scaled by total assets at year-end
BIG 4 = an indicator variable equal to 1 when the auditor is a member
of the Big 4, and 0 otherwise
OP CASH FLOW = operating cash flows divided by total assets at fiscal year end
log(NON-AUDIT FEE)
= the natural logarithm of the sum total of all non- audit fees paid to the incumbent auditor
FEERATIO = the ratio of non-audit fees to total fees paid to the incumbent
auditor
The last two variables log(NON-AUDIT FEE) and FEERATIO are independent variables in the model. The inclusion of FEERATIO is to avoid the situation where the non-audit fee is the main source of income from the company for the audit firm. If non-audit service impair auditor independence, negative coefficients on log(NON-AUDIT FEE) and FEERATIO are expected.
Other variables are all control variables. Log(ASSETS), the natural logarithm of total assets at the end of the fiscal year prior to filing, is the proxy measure for firm size. I include the variable because large firms have more negotiating power in the event of financial difficulties and hence they are more likely to avoid being given a going concern opinion. Mutchler et al. (1997) find that debt covenant violations are positively associated with the probability of issuing a going concern opinion. Because firms with high and increasing leverage are more likely to be closer to debt covenant violations. Therefore LEV is included to capture proximity to covenant violation because firms close to violation are likely to have high leverage (Beneish and Press, 1993), and CLEV is included because increases in leverage are likely to move firms closer to covenant violation (Reynolds and Francis, 2000). INVESTMENTS is the sum of the firm's cash and investment securities (long and short term), scaled by total assets, and is a liquidity measure that captures the ability to quickly raise cash. It is included because firms with large cash and investment securities have more resources to avoid financial distress and hence avoid being given a going concern opinion. I also include BIG 4, an indicator variable equal to 1 if the auditor is a member of the Big 4,
and 0 otherwise. This variable is included because prior research argues that Big 4 auditors are more likely to issue going concern audit opinions (Mutchler et al., 1997), consistent with potentially higher litigation costs influencing auditor reporting decisions. OP CASHFLOW (operating cash flows divided by total assets) is included because operating cash flows are often associated with the going concern ability.
4. Results
4.1. Descriptive statistics
I limit the sample to distressed firms (defined as firms with negative cash flow in the current year) and then winsorize all continuous variables at the 99th percentile of their absolute values. Table 1 presents descriptive statistics on the sample for the variables used in the going concern model. I scaled NON-AUDIT FEE by thousand and ASSETS by million. And all the currency units present in the table is Chinese Yuan (CNY). The first two rows present the measures of fee ratio of non-audit fees in total fees received by audit firms and non-audit fees used in the analysis. Row 1 indicates that the mean of FEERATIO are 3%. The ratios are far less than the mean fee ratios reported in DeFond et al. (2002) and Robinson (2008) which is 49% and 37%, respectively. Other than the difference with western countries in the average level of provision of NAS, what is interesting in the sample is that the maximum non-audit fee ratio amounts to 40.16% with the extreme zero provision of NAS in the sample, which illustrates the relatively uneven situation in Chinese audit market. Row 2 reports that the mean value of non-audit fee is ￥34,429. Agree with row 1, the sample firms report relatively low non-audit fees compared with those reported in DeFond et al. (2002)after taking the currency rate into consideration. Row 3 in table 1 shows that on average 58% of the sample receives going concern opinions, which is much larger than the 8% reported in DeFond et al. (2002). However, it is consistent with DeFond et al. (2000) which concludes that the frequency of modified opinions increases nine-fold subsequent to the adoption of the new standards. Row 4 indicates that mean and median firm sizes, measured in total assets, are ￥6464 million and ￥1804 million respectively, indicating a skewed distribution and justifying the decision to log assets in the logit analysis. Row 5 and 6 indicate that mean and median values of leverage are 0.51 and 0.48, respectively, and that the median change in leverage is relatively small. Row 7 reports that mean and median values for investments are 18.3% and 14.2%, respectively, nearly half of what is reported in DeFond et al. (2002). Row 8 reports that only 6.2% of the sample has a Big 4 auditor.
Table 1
Descriptive Statistics for 10,294 Financially Distressed Firm-years (Including 569 Firm-years with Going Concern Opinions) for Fiscal Years 2002-2013.
Variables Mean Median SD Max. Min.
1. FeeRatio 0.030 0 0.088 0.402 0
2. Non-Audit Fee (￥thousands) 34.429 0 122.776 800 0
3. Opinion 0.580 0 0.234 1 0 4. Assets (￥millions) 6464.356 1804.585 19289.54 158522.5 118.933 5. LEV 0.509 0.478 0.368 3.019 0.048 6. CLEV 0.026 0.016 0.192 1.053 -0.788 7. Investments 0.183 0.142 0.144 0.665 0.002 8. Big 4 0.062 0 0.241 1 0 9. Op Cash Flow 0.633 0.514 0.492 2.763 0
Compared to the 91% sample firms being audited by Big 5 (Big 5 by then) auditor in DeFond et al. (2002)’s paper. DeFond et al. (2000) found that along with the increase in the frequency of Chinese auditors issuing modified opinions, there is a decline in audit market share among large auditors which have the greatest propensity to issue modified reports. This is justified in this set of descriptive data. According to Tang (1999), at present, there are 22 representative offices, 11 joint ventures and seven member firms of international CPA firms. The Big 4 takes up about 15 percent of the total market share in China. Finally, row 9 reports that mean and median operating cash flows divided by total assets are 63.3% and 51.4%.
4.2. Descriptive statistics by opinion type
Table 2 classifies the variables from Table 1 by opinion type, along with p-values from t-tests of differences across the two groups. Comparing the fee variables in the first two rows represents univariate tests of the hypotheses. Row 1 indicates that the mean values for FEERATIO is 2.1% for the going concern sample, compared with 3.03% for the clean opinion sample, with the differences significant at p=1% (two-tailed). Row 2 shows that the mean for NON-AUDIT FEE is 16,668 for the going concern sample, and 35,533 for the clean opinion sample. The difference between the means is significant at p<1% (two-tailed). Row 3 indicates that the assets for clean opinion sample are about four times than that of companies with going concern opinion. This is consistent with the reason for choosing assets as a control variable that large firms are more likely to avoid being given a going concern opinion. Consistent with expectation, row 4 finds that the going concern firms have higher leverage (LEV). Row 5 shows that increase in leverage in both groups is not
significant. Row 6 shows that INVESTMENT is lower at p<1% (two-tailed) among the going concern group, suggesting that firms with less liquidity are more likely to receive going concern opinions. Row 7 indicates that the relative frequency of Big 4 auditors is lower for going concern sample, significant at p<1%. This can be explained by the reputation concern of Big 4 audit firms. Big auditors firms resign from clients if the risk is high after evaluation of the clients. As expected, the last row shows that the mean value of OP CASH FLOW is significantly lower among the sample with going concern opinions at p<1% (two-tailed).
In summary, the descriptive statistics presented in tables 1 and 2 are consistent with the distressed nature of the total sample and with the going concern sample being even more distressed. They also roughly reflect the reality in Chinese audit market.
Table 2
Comparison of Going Concern and Clean Opinion Sample for 10,294 Financially Distressed Firm-years (Including 569 Firm-years with Going Concern Opinions) for Fiscal Years 2002-2013.
Variables Mean
GC Sample No GC Sample p-value
1. FeeRatio 0.021 0.030 0.010
2. Non-Audit Fee (￥thousands) 16.668 35.533 0
3. Assets (￥millions) 1656.38 6761.21 0 4. LEV 0.933 0.478 0 5. CLEV 0.016 0.027 0.564 6. Investments 0.139 0.186 0 7. Big 4 0.023 0.068 0 8. Op Cash Flow 0.478 0.640 0
4.3. Multivariate test results
Table 3 presents the results of estimating the logit model. Model 1 presents a baseline case of the going concern model without including the fee variable, and model 2 and 3 sequentially introduce variable log(Non-audit Fee) and Fee Ratio.
The results indicate that significance is found in the predicted direction at p<5% (two-tailed) for all cases for the coefficients log(ASSETS) and LEV.
Table 3
Going Concern Opinion Models, with Non-audit Fee Included as Independent Variable, for 10,294
Financially Distressed Firm-years (Including 569 Firm-years with Going Concern Opinions) for Fiscal Years 2002-2013.
Model 1 Model 2 Model 3
Predicted Sign Coefficient (p-value) Coefficient (p-value) Coefficient (p-value) Intercept \ 0.275 (0) 0.201 (0.004) 0.207 (0.003) log(ASSETS) - -0.367 (0) -0.027 (0.002) -0.032 (0.001) LEV + 0.183 (0) 0.129 (0.001) 0.127 (0.001) CLEV + 0.015 (0.293) 0.200 (0.001) 0.199 (0.001) INVESTMENTS - -0.430 (0.028) -0.090 (0.183) -0.091 (0.178) BIG 4 + 0.024 (0.034) 0.139 (0.616) 0.005 (0.875) OP CASH FLOW - -0.036 (0) -0.029 (0.120) -0.031 (0.097) log(NON-AUDIT FEE) - \ 0.004 (0.608) 0.194 (0.249) FEERATIO - \ \ -0.154 (0.301)
The results in models 2 and 3 indicate that the estimated coefficients on both fee variables are insignificant in each case. The lack of significance on log(NON-AUDIT FEE) in models 2 and 3 and on FEERATIO in model 3 indicates that support for the hypothesis is not found. The lack of significance on the fee variables in the regression presents concise evidence that the hypothesis that non-audit service fees are inversely related to auditors’ propensity to issue going concern audit opinions is rejected.
4.4. Robustness checks
As in prior research I limit the sample to distressed firms (defined as having negative cash flow in this paper) that receive going concern opinions. To test the robustness of the results to these sample restrictions, I replicate the analysis in table 3, using a sample of all available firms (i.e. not limited to distressed firms) that receive going concern opinions. The results from the regression report that all of the coefficients on all of the coefficients on the fee variables remain insignificant at conventional levels. Thus, the findings are not sensitive to the sample-selection criteria.
5. Conclusion
The Enron scandal and the role its auditor, Arthur Andersen, played in the scandal brought the effect the provision of non-audit services have to auditor independence
into public attention again. Regulators have taken actions to curtail auditor-provided non-audit services. These actions are based on the premises that non-audit service fees impair auditor independence by making the auditor economically dependent on the client and that the consulting nature of non-audit services reduces the auditor’s objectivity. There are already studies empirically investigate the validity of the these concerns by gathering systematic evidence on the association between non-audit and audit fees paid to incumbent auditors and auditor independence measured as the propensity of auditors to issue going concern opinions. However, the study of Chinese audit market on this topic is rare. Therefore, I carry out the study aiming to add to the discussion some evidence of China’s audit reality.
I derive the conclusions from a research design that uses the auditor’s propensity to issue a going concern opinion to proxy for auditor independence. According to the results presented above, contrary to regulators’ concerns and consistent with most studies on this topic, I find no evidence of a significant association between the auditor’s propensity to issue a going concern opinion and provision of non-audit service in Chinese audit environment. This finding is robust to replacing the financially distressed sample with full sample.
A potential limitation of the research design is that the tests lack the application of bankruptcy prediction models to predict auditors’ assessments of companies’ going-concern status. Mutchler (1984), and Peel (1989) find that the bankruptcy prediction models are better at predicting going-concern qualifications than auditors. However, due to the restriction of the level of academic skill, I fail to include the bankruptcy prediction models in my research. The results are supposed to be more accurate and with more explanation power with the model adopted.
References
Allen, F.; Qian, J.; and Qian, M. (2005) Law, finance, and economic growth in China. Journal of Financial Economics, 77: 57-116
American Institute of Certified Public Accountants (AICPA, 1979) Scope of Services by CPA Firms, Public Oversight Board (AICPA, 1979)
American Institute of Certified Public Accountants (AICPA, 1988) Analytical Procedures. Public Oversight Board (AICPA, 1988)
Antle, R. (1984). Auditor independence. Journal of Accounting Research, 22 (Spring): 1-20
Ashbaugh, H.; Lafond, R.; and Mayhew, B. W. (2003) Do Non-Audit Services Compromise Auditor Independence? Further Evidence. Accounting Review, 78: 611–639.
Barkess, L. and Simnett, R. (1994) The provision of other services by auditors: Independence and pricing issues. Accounting and Business Research, 24: 99– 108.
Beck, P.J., Frecka, T.J., and Solomon, I., (1988) A model of the market for MAS and audit services: Knowledge spillovers and auditor-auditee bonding. Journal of Accounting Literature, 7: 50-64
Beneish, D. and Press, E. (1993) Costs of technical default. The Accounting Review, 68: 233-257
Beijing Review. (1998) Foreign Investment Achievement Exhibited. 41(50): 5
Beijing Review. (1999) FDI Rises in 1998. 42(6): 5
Blacconiere, W. and DeFond, M. (1997) An investigation of independent audit opinions and subsequent independent auditor litigation of publicly traded failed savings and loans. Journal of Accounting and Public Policy, 16: 415-454
Chan H. K. and Lin K. Z. (2000) Auditing standards in China --- A comparative analysis with relevant International Standards and Guidelines. The International Journal of Accounting, 35, No. 4: 559-577
Chung, H. and Kallapur, S. (2001) Client importance, non-audit services, and abnormal accruals. Working paper, Purdue University.
Corless, J.C. and Parker, L.M. (1987) The impact of MAS on auditor independence: An experiment. Accounting Horizons, 1: 25-29
Craswell. A.T. (1999) Does the provision of non-audit services impair auditor independence? International Journal of Auditing Int. J. Audit, 3: 29-40
DeAngelo L.E., (1981) Independence, “low bailing”, and regulation. Journal of Accounting and Economics, 3: 113-127
DeFond, M.L., Wong, T.J., Li, S.H. (2000) The impact of improved auditor independence on auditor market concentration in China. Journal of Accounting and Economics, 28: 269-305
Defond, M. L.; Raghunandan, K.; and Subramanyam, K. R.. (2002) Do non-audit service fees impair auditor independence? Evidence from going-concern audit opinion. Journal of Accounting Research, 40: 1247–1274.
Francis. (2004) What do we know about audit quality? The British Accounting Review, 36: 345-368
Francis, J., and Ke, B. (2002) Do non-audit services compromise auditor independence? Working paper, University of Missouri.
Francis, J., and Ke, B. (2006) Disclosure of fees paid to auditors and the market valuation of earnings surprises. Review of Accounting Studies, 11: 495–523.
Frankel, R. M.; Johnson, M. F.; and Nelson, K. K. (2002) The relation between auditors’ fees for non-audit services and earnings management. The Accounting Review, 77: 71–106.
Geiger, M.A. and Rama, D. V. (2003) Audit fees, nonaudit fees, and auditor reporting on stressed companies. Auditing : A Journal of Practice and Theory, 22, No. 2: 53-69
Gul, F.A. (1991) Size of audit fees and perceptions of auditors' ability to resist management pressure in audit conflict situations. Abacus (September): 162-172.
Hao, Z.P. (1999) Regulation and organization of accountants in China. Accounting, Auditing and Accountability Journal, 12: 286-302
Hylas, R.E. and Ashton, R.H. (1982) Audit detection of financial statement errors. The Accounting Review, 57: 751–765.
Jensen, M., and Meckling, W. (1976) Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3: 305-360
Kinney, W., Palmrose, Z-V., and Scholz, S. (2004) Auditor independence, nonaudit services, and restatements: Was the U.S. Government right? Journal of Accounting Research, 42: 561-588
Kreutzfeldt, R.W. and Wallace, W.A. (1986) Error characteristics in audit populations: Their profile and relationship to environmental factors. Auditing: A Journal of Practice & Theory, 6: 44-74.
Krishnan, J.; Sami, H.; and Zhang, Y. (2005) Does the provision of nonaudit services affect investor perceptions of auditor independence? Auditing: A Journal of Practice & Theory, 24: 111-135.
Levitt, A. (1998) The “Numbers Game.” Remarks of SEC Chairman A. Levitt at the New York University Center for Law and Business, New York, NY, September 28.
Mutchler, J. (1984) Auditors’ perceptions of the going-concern opinion decision. Journal of Accounting Research (Autumn): 668-682
Mutchler, J., Hopwood, W. and McKeown, J., (1997) The influence of contrary information and mitigating factors in audit opinion decisions on bankrupt companies. Journal of Accounting Research: 295-310
National Bureau of Statistics, PRC. (2000) China Statistical Abstract. Beijing: China Statistics Press. P. 105
Palmrose, Z. (1986) The effect of non-audit services on the pricing of audit services: Further evidence. Journal of Accounting Research, 24: 405-411.
Pany, K. and Reckers, P.M.J. (1988) Auditor performance of MAS: A study of its effects on decisions and perceptions. Accounting Horizons (June): 31-38.
Peel, M.J. (1989) The going-concern qualification debate: Some UK evidence. The British Accounting Review, 21: 329-350
Reynolds, K.; Deis, D.; and Francis, J. (2002) Professional service fees and auditor objectivity. Working paper, Louisiana State University.
Reynolds, K., and Francis, J. (2000) Does size matter? The influence of large clients on office-level auditor reporting decisions. Journal of Accounting and Economics, 30: 375-400
Ramage, J.G., Krieger, A.M. and Spero, L.L. (1979) An empirical study of error characteristics in audit populations. Journal of Accounting Research, 17(Supplement): 72-102.
Robinson, D. (2008) Auditor independence and auditor-provided tax service: Evidence from going-concern audit opinions prior to bankruptcy filings. Auditing: A Journal of Practice & Theory (November): 31-54.
Schulte, A.A., Jr. (1965), Compatibility of management consulting and auditing. The Accounting Review (July): 587-593.
Securities and Exchange Commission (SEC). Final rule: Revision of the commission’s auditor independence requirements. 17 CFR Parts 210 and 240. Washington, DC: SEC, 2000.
Sharma, D.S. and Sidhu, J. (2001) Professionalism vs Commercialism: The association between non-audit services (NAS) and audit independence. Journal of Business Finance & Accounting, 28(5) & (6), June/July.
Simunic, D. (1984) Auditing, consulting, and auditor independence. Journal of Accounting Research, 22: 679-702.
Tang, Y.W. (1999) Issues in the development of the accounting profession in China. China Accounting and Finance Review, 1: 21-36.
Titard, P. L., (1971) Independence and NAS: Opinions of financial statement users. The Journal of Accountancy (July): 47-52
Watts, R. and Zimmerman, J., (1981) The markets for independence and independent auditors. Unpublished manuscript (University of Rochester, Rochester, NY).
Watts, R. and Zimmerman, J., (1983) Agency problems, auditing, and the theory of the firm: Some evidence. Journal of Law and Economics, 26: 613-633
Weil, J. (2001) “Going concerns”: Did accountants fail to flag problems at Don-Com casualties? The Wall Street Journal (February 9, 2001): C1
Wines, G. (1994) Auditor independence, audit qualifications and the provision of non-audit services: A note. Accounting and Finance, 34: 75–86.
