第五章財務報告質量

從更廣泛的財務報告使用者角度來看,按照先前的資料，由現任審計師對審計獨立性和審計質量提供非審計服務評估提供的財務質量報告這一方面，可以推測，收益管理指標，財務報表和審計滯后報告是可以衡量財務報告質量的。至于這三個報表如何判定和規定審計和非審計服務以及審計獨立性和審計質量之間的聯系將分別討論。

 

會計師

人們一般傾向于認為滿足基準收益和更多自由支配的代理會損害審計師的獨立性和降低財務報告的質量。然而,非審計服務費用和會計師的客戶之間關系是復雜的,我將一一分析細節。

 

Chapter 5 Financial Reporting Quality

 

From the perspective of wider financial reporting users, prior literatures apply financial reporting quality as one aspect for assessing the effects of non-audit services provided by incumbent auditors on the auditor independence and audit quality. It is presumed that indicators as earnings management (discretionary accruals) (Frankel et al., 2002; Kinney and Libby, 2002; Ashbaugh et al., 2003; Chung and Kallapur, 2003; Deis et al., 2004; Ferguson et al., 2004; Larker and Richardson, 2004; Antle et al., 2006; Huang et al., 2007; Cahan et al., 2008; David et al., 2008), financial statement restatements (Agrawal and Chadha, 2003; Raghunandan et al., 2003; Ferguson et al., 2004; Kinney et al., 2004; Schneider et al., 2006; Paterson and Valencia, 2011; Seetharaman et al., 2011), and audit report lags (Ashton et al., 1987, 1989; Ashton and Newton, 1989; Bamber et al., 1993; Knechel and Payne, 2001; Jensen and Payne, 2002; Barua et al., 2010; Knechel et al., 2012; Knechel and Sharma, 2012) are appropriate to measure the financial reporting quality. How these three proxies to evaluate the relationship between joint provisions of audit and non-audit services and auditor independence and audit quality will be discussed respectively.

 

5.1會計師Discretionary accruals

It is widely assumed that the propensity to meet earnings benchmarks and the higher discretionary are proxies which indicate the impaired auditor independence and the reduced quality of financial reporting. However, the results of the association between non-audit services fees and discretionary accruals of clients are mixed, which will be analysed in details.

 

5.1.1三個沖突的觀點Three conflict views 

While some literatures demonstrate the association between non-audit services fees and earnings management is positive (Frankel et al., 2002; Ferguson et al., 2004; Cahan et al., 2008), others find such association is negative (Kinney and Libby, 2002; Larker and Richardson, 2004; Antle et al., 2006). In addition, most literatures find that no evidently association exists between non-audit services fees and earnings management of clients (Ashbaugh et al., 2003; Chung and Kallapur, 2003; Deis et al., 2004; Antle et al., 2006; Huang et al., 2007). These conflict views will be analysed in detail and possible reasons for the conflict results will be discussed.#p#分頁標題#e#

 

Both Frankel et al. (2002) and Ferguson et al. (2004) support the assertion that non-audit services fees increase the earnings management. Specifically, Frankel et al. (2002) use a sample of U.S. companies reporting from February to June in the year 2001 and report that non-audit services fees are significantly positive related with the propensity to meet the analysts’ forecasts and the magnitude of discretionary accruals, both of which are proxies for earnings management and indicators for impaired auditor independence and low quality of financial reporting. This conclusion is consistent with economic bonding theory as providing non-audit services can motivate auditors to acquiesce to pressures from clients. Using a sample of UK companies during the year 1996 to 1998, Ferguson et al. (2004) apply three measures of earnings management and three measures of non-audit services purchases levels. As a result, their relationship is positive. In conclusion, a high level of non-audit services will result in corresponding high discretionary accruals, which indicate compromised auditor independence and low audit quality. In addition, by examining the association between discretionary accruals and non-audit services fees and time-period of non-audit services, Cahan et al. (2008) argue that a significantly positive relationship exists. 

 

However, Larker and Richardson (2004) find the association between non-audit services fees and the magnitudes of discretionary accruals is negative. In addition, they explain the positive relationship stated by Frankel et al. (2002) is because companies in the sample are small and management-controlled. Deis et al. (2004) also support Larker and Richardson (2004) because the association between discretionary accruals and non-audit services fees documented by Frankel et al. (2002) disappears when additional controls for high-growth firms are included in the model. Kinney and Libby (2002) also criticism the findings of Frankel et al. (2002) as limitations, such as problems in definitions of earnings management, flaws operationlisation of theoretical constructs, failure to control for other potential causes, and data or statistical problems, in the research might compromise the results. Moreover, by examining U.K. data of non-audit services fees during the period 1994 to 2000, Antle et al. (2006) intends to explore a better understanding of audit fees, non-audit fees, and abnormal accruals by combining them as endogenously determined in a system of three equations. As a result, the negative relationship between discretionary accruals and non-audit services support the knowledge spill-over or economies of scope (see Chapter 3).

 

Apart from above opposite statements, most studies suggest that no significant association between non-audit services and accounting accruals or earnings surprises (Ashbaugh et al., 2003; Chung and Kallapur, 2003; Deis et al., 2004; Huang et al., 2007). Although these studies reach the same conclusions, their processes to carry out the empirical research are different. For example, Ashbaugh et al. (2003) adopt an alternative estimate of discretionary accruals and find that no association between income-increasing accruals and client purchases. Chung and Kallapur (2003) use the ratio of non-audit services fees from a specific client to the audit firm’s whole U.S. revenues to assess the client importance and find that no statistically significant relationship between client importance and the estimate of discretionary accruals. More detailed experiment is conduct by Huang et al. (2007) as they adopt abnormal accruals and meeting earnings benchmarks to measure financial reporting quality and involve 3,269 companies which disclose non-audit fees in the year 2003 and 2004. They finally fail to find evidence to support the notion that clients involve a high level of tax fee ratio and other non-audit fee ratio have a corresponding high possibility to biased financial reporting.#p#分頁標題#e#

 

5.1.2 結果不一致的可能原因Possible reasons for inconsistent findings

These three different opinions towards the relationship between non-audit services fees and discretionary accruals might result from the indirect influences of auditors’ on the client’s accruals and earnings characteristics (DeFond et al., 2002) and empirical problems existed in discretionary measurements (Dechow et al., 1995; Guay et al., 1996; McNicholes, 2000; Collins and Hribar, 2002). Therefore, it is worthy other evidence to support the conclusion drawn by using only proxy of discretionary accruals and a more direct test that can clearly reflect auditors’ judgement is required in future research. Moreover, as stated above, sample selection is an important issue which might cause flaw conclusions, particular the sample is too small or other potential causes cannot be eliminated.

 

5.2 財務報表重述Financial statement restatements

A restatement is necessary when a company, either voluntarily or forced by auditors or regulators, amends public financial information that was previously reported (Government Accountability Office (GAO), 2002). It is assumed that such restatements are indications of lower financial reporting quality because the financial statements were not fairly presented in compliance with generally accepted accounting principles (GAAP). Francis and Michas (2013) even define “audit failure” as audit engagements in which a downward restatement of previously audited client earnings. They also argue that when there is subsequent correction of clients’ overstated earnings by a down statement, the audit quality is low. Therefore, financial reporting quality can be measured by restatements of previously issued financial statements. For eliminate restatements not caused by low financial reporting quality, restatements derived by mandated retrospective utilisation of GAAP or by changes in the accounting entity are excluded for research purposes (Seetharaman et al., 2011).

 

In order to explore whether clients’ enterprises which pay higher fees for non-audit services are more likely to restate their financial statements, Raghunandan et al. (2003) compare the fees paid by the 110 firms that restated their financial statements previously with 3,481 other firms in 2001 and conclude that no significant differences between these two groups for unexpected non-audit services fees.

 

Moreover, Kinney et al. (2004) abandon the single-period approach but investigate five types of non-audit services’ (financial information systems design and implementation, internal audit, audit-related, tax, and unspecified fees) influences on auditor independence. By examining the detailed audit firm fees for financial information systems design and implementation, internal audit, and other non-audit services provided by incumbent auditors during calendar years 1995 to 2000, Kinney et al. (2004) find that the relationship  is not evident between financial information systems design and implementation or internal audit services and financial statements restatements. However, there is a negative association between restatements and tax services while a positive relationship between financial statements restatements and unspecified fees. Therefore, it is suggested that the impacts of economic dependence on clients are more than compensated by benefits of financial reporting, or alternatively, that registrants with high quality are more likely to choose their audit firms to provide tax services. In other words, prohibition or restrictions of tax services from registrants’ audit firms might either impair the financial reporting quality or increase professional services costs to registrants without relevant benefits from strengthened independence of audit firm. Mishara et al. (2005) confirm that it is necessary to investigate different types of non-audit services fees because investors’ perceptions vary with the type of non-audit fees paid to the incumbent auditor.#p#分頁標題#e#

 

However, the results of Kinney et al. (2004) might not generalise in a post-SOX environment and the restatements are not categorised in accordance with types of NAS. Seetharaman et al. (2011) further their research by investigating the relationship between non-audit tax services (NATS) provided by incumbent auditors and financial reporting quality for listed companies in a post-SOX environment. On examination, they find that no significant association between auditor-provided NATSs and general financial statement restatements but the auditor-provided NATSs are significantly associated with tax-related financial statement restatements, which is defined as an error or a misapplication of GAAP in the tax field cause for the restatements. 

 

In addition, acknowledged the importance of non-audit services types to research the restatements, Schneider et al. (2006) suggest that the factor of recurring engagements from nonrecurring ones should also be considered in investigating the interaction with auditor independence, which would enhance the ability of investigation to systematically study the impacts of non-audit services on auditor independence empirically. Paterson and Valencia (2011) also extend the research of Kinney et al. (2004) by concluding that only the recurring tax services, not nonrecurring ones, are significantly negative related with restatements.

 

5.3 審計報告滯后Audit report lags

Audit report lag has been considered as an important proxy to assess the audit efficiency and production (Ashton et al., 1987, 1989; Ashton and Newton, 1989; Bamber et al., 1993; Knechel and Payne, 2001; Jensen and Payne, 2002; Barua et al., 2010; Knechel et al., 2012). Ashton and Newton (1989) point out that audit delay is the only publicly observable proxy to assess audit efficiency, which is supported by empirical evidence from Gaeremynck et al. (2010). According to Bamber et al. (1993), efficiency refers to the utilisation of fewer inputs to gain a given output, and audit report lag, which is the time required to complete the audit, is an appropriate proxy to evaluate audit inputs. 

 

Although tremendous researches focus on the evidence that non-audit services seems to impair the effectiveness of the audit (e.g., Frankel et al., 2002; Ferguson et al., 2004; Huang et al., 2007; Hermanson and Ye, 2009), others do not find significant relationship between non-audit services provided by incumbent auditors and impaired auditor independence (e.g., DeFond et al., 2002; Ashbaugh et al., 2003; Geiger and Rama, 2003; Raghunandan et al., 2003). In addition, some studies suggest that audit effectiveness is promoted by the knowledge spill-over provided by joint provisions of audit and non-audit services (e.g., Kinney et al., 2004; Robinson, 2008; Knechel et al., 2012). Apart from the attention on the audit effectiveness, another stream of literatures is related to audit efficiency (e.g., Simunic, 1984’ Palmrose, 1986; Abdel-khalik, 1990; Davis et al., 1993; O’Keefe et al., 1994; Knechel and Payne, 2001; Antel et al., 2006). However, an important limitation of above research is that both effectiveness and efficiency are isolated and few researches consider their potential complementary relationship. #p#分頁標題#e#

 

Knechel and Sharma (2012) attempt to resolve this limitation by investigate the association between non-audit services and both the audit effectiveness and efficiency using evidence from fees paid for non-audit services and lags in audit reports. According to their research, they argue that the announcement of SOX will influence the relationship between audit efficiency (indicator of audit report lags) and audit effectiveness (measured by discretionary accruals and financial restatements). In other words, discretionary accruals and financial restatements do not increase with shorter audit lags with a high level of non-audit fees. Moreover, Knechel et al. (2012) also examine the association between non-audit services fees and audit quality in New Zealand and results show that a high level of non-audit fees paid to the incumbent auditors with shorter audit lag will not impair audit quality. In addition, the negative association between non-audit fees and audit lad reveal the impact of knowledge spill-over. Therefore, it can be seen that shorter audit lag is an indicator of higher audit efficiency and non-audit services have positive impact on the audit efficiency because of the effects from knowledge spill-over.

 

 

Chapter 6 監管機構的法規的發展觀念Regulators’ Perceptions and Regulations Developments

 

There always a tight relationship between auditing academics and regulators because advanced researches can provide theoretical information to regulators for their decisions on amendments on auditing rules or regulations. Therefore, it is worthy investigating on the regulators’ perceptions on impacts of non-audit services in order to understanding their intention behind the updating of rules and regulations. In addition, regulations developments are promoted by special events, such as Enron and WorldCom, thus examining the development of regulations can help understand the attitudes of government and forecast further change in such regulations.

 

6.1 美國證券交易委員會U.S. Securities and Exchange Commission (2000)

Substantial growth in the provision of non-audit services to audit clients before the audit scandals (e.g., Enron, WorldCom) had resulted in great concern by regulators and other stakeholders on the influences on auditor independence and audit quality by non-audit services. Although accounting professions have long argued that no material conflict of interests arise from joint provision of audit and non-audit services, regulators hold the view that the resulting economic bonding between auditor and clients will definitely impair auditor objectivity and integrity because auditors are more likely to sacrifice their independence in exchange for retaining clients who can pay large amount of non-audit fees. As stated by the U.S. Securities and Exchange Commission (SEC) 2000 (9):

“Under any circumstances, it can be difficult for an auditor to make a judgment that works against the auditor client’s interest. Where that judgment may imperil a range of service engagements of the firm, of which the audit is a fairly small part, it may be unrealistic to expect that an auditor can ignore completely what the firm stands to lose by the auditor’s action.”#p#分頁標題#e#

Such concerns have triggered tremendous studies focusing on the association between non-audit services and auditor independence and audit quality, which are discussed and compared in prior chapters (see Chapter 3, 4, and 5). 

However, in response to proposal of U.S. Securities and Exchange Commission (SEC) in 2000, Barry Melancon, president of the AICPA, argued that:

“There will be a loss of synergies that exist when a firm provides a broad array of audit and non-audit services to its clients… In addition, the loss of non-audit services lines will reduce the scope of knowledge available at the accounting firms” (Melancon, 2000).

Similar statements are also from Steven William, who is the former commissioner of the U.S. Securities and Exchange Commission (SEC), remark that restrictions on joint provision of audit and non-audit services abandon the benefits to the auditors of understanding more about the audit client and its business environment (Wallman, 1996).

 

Until the burst of Enron bankruptcy, increasing concerns about auditor independence have promoted Congress to enact legislation that prohibited most non-audit services provided by incumbent auditors. Specifically, rules about disclosure of non-audit services fees were issued by U.S. Securities and Exchange Commission (SEC 2000) and publicly traded companies in the U.S. started to disclose the non-audit services fees they paid to their auditors on February 5, 2001. The first-time fee disclosures reveals the substantial purchases of non-audit services by companies, which trigger the scepticism about the auditor independence (Weil and Rapoport, 2003). Regulators are more confirm about the assumption of economic bonding will compromise auditor independence. However, DeFond et al. (2002) point out that this assumption ignores auditors’ expected costs of sacrificing their independence because loss of reputation and costs of litigation will provide stronger motivations for auditors to keep their independence.

 

6.2 2002年薩班斯-奧克斯利法案Sarbanes-Oxley Act of 2002 (SOX)

As the same period, the U.S. Congress issued the Sarbanes-Oxley Act of 2002 (SOX) in order to rebuild confidence in U.S. market for making up the damages of a series of accounting scandals. Particular the Section 201 of SOX, which bans most provisions of non-audit services provided by the same auditor who is also responsible for external audit. As regulated by Section 201 of SOX, following non-audit services are prohibited:

“(1) bookkeeping or other services related to the accounting records or financial statements of the audit client; (2) financial information system design and implementation; (3) appraisal or valuation services, fairness opinions, or contribution-in-kind reports; (4) actuarial services; (5) internal audit outsourcing services; (6) management functions or human resources; (7) broker or dealer, investment adviser, or investment banking services; (8) legal services and expert services unrelated to the audit; and (9) any other service that the Board determines, by regulation, is impermissible.”#p#分頁標題#e#

 

6.3 美國證券交易委員會U.S. Securities and Exchange Commission (2003)

As the initial mandated disclosure of non-audit services attracts more concern from a number of interest parties, who are surprising of the frequency and magnitudes of non-audit services (Weil, 2001). Unger (2001) cites a statement from Laura Unger, who was a Commissioner of Securities of Exchange Commission, that:

“We were all quite surprised to learn this… in fact, the numbers appear to demonstrate the problem may be larger than we originally thought”.

 

Further development of SEC (2000) occurs in 2003, as the SEC enacted revised fee disclosure rules. Firstly, definition of audit fees is broadened.  SEC (2000) defines audit fees as:

“The fee paid for the annual audit and for the review of the company’s financial statements included in the compnay’s Forms 10-Q or 10-QSB for the most recent fiscal year”.

However, SEC (2003) expands above definition to include following:

    “Services that generally only the independent accountant reasonably can provide, such as comfort filed with the Commission”.

Secondly, while SEC (2000) only specifies two types of non-audit fees, i.e., information systems fees and other fees, SEC (2003) introduces two new types of non-audit services fees (audit-related fees and tax fees) and eliminates information systems fees. Thirdly, previous disclosure only requires for one year while new rules require disclosure for both current year and the prior year. The reclassification of non-audit fees is effective for fiscal years ending after December 15, 2003. However, the SEC encouraged companies to adopt new rules early.

 

The changes in rules are mainly because the lobbying efforts by the major accounting profession (Weil and Rapoport, 2003), possibly stemming from concerns about perceived auditor independence (Abbott et al., 2011). Dopuch et al. (2003) argue that the media coverage following the initial fee disclosure and the collapse of Enron Corporation leads to the perception that auditor independence is adversely affected by the joint provision of non-audit services despite if actual independence is not compromised. Levinsohn (2001) indicate that some investors even require companies to dismiss their auditors who offer substantial non-audit services. Therefore, it seems plausible that companies can reverse their reputation loss by reclassification, by reducing the reported amount of non-audit services fees or by classifying non-audit fees as audit-related fees, thus they have incentives to adopt new rules early. According research of Asthana and Krishnan (2006), which is consistent with Mishra et al. (2005), it is suggested that companies and their auditors are more likely to expect audit-related fees will be viewed more favourably than other non-audit fees.#p#分頁標題#e#

 

6.4 對比有關規定Comparisons between relevant regulations

In the environment of rule-making deliberations, the SEC (2000, 2003) adopts a demand-side approach to regulating on purchases of non-audit services. In other words, Securities of Exchange Commission supports the market to decide the optimal level of non-audit services purchases by means of mandating disclosures of non-audit services. Therefore, under the condition of penalties derived by a relatively high level of non-audit services purchases, the registrants tend to have the motivations to lower their purchasing of non-audit services level (Abbott et al., 2011). On the other hand, a supply-side approach has been adopted by the Sarbanes-Oxley Act of 2002 (SOX) as it directly prohibits most non-audit services (Canada et al., 2008; U.S. House of Representatives, 2002). However, despite the differences between demand and supply approaches created by the basic starting point, the detailed regulations are almost same.

 

Chapter 7 結論Conclusion

 

The conclusion includes a summary of context, limitations of this literature review, and suggestions for further research, which will provided respectively.

 

7.1 總結Summary of context

This literature review investigates the complicated influences of joint provisions of audit and non-audit services from four stakeholders’ perspectives, i.e., auditors, clients, financial reporting users, and regulators. A variety of proxies are used by previous literatures to evaluate whether auditor independence or audit quality is impaired, such as auditors’ propensity to issue going-concern opinions, restatements of financial reporting. Despite the results are mixed and elusive for a couple of reasons, this research is still contributing to existing literatures as limited research focuses on overall literature reviews specifies on non-audit services. However, some limitations still exist, which will be discussed. In addition, suggestions are also provided on further research directions about the impact of non-audit services provided by incumbent auditors on auditor independence and audit quality.

 

7.2 文獻綜述局限Limitations of this literature review

There are several limitations about this literature review. Firstly, as stated above, although most papers are gathered in leading journals, working papers, which represent the latest research achievements, are not included sufficiently because of the limited accessibility. Therefore, this literature review is not up-to-date. Secondly, the study put more emphasis on research results thus ignores the comparisons and analysis of research methodology. As research methodology might have direct and vital effects on research outcomes, lack of sufficient analysis might cause confusions about the mixed and elusive conclusions. Thirdly, most data used in literatures are based in U.S. and U.K., which are developed countries with sound regulation environment. As the increasing globalisation of world economies, other countries, especially developing countries like China and India, have limited research documents. It is worth further investigation on such less regulated environment in order to offer guidance on their development of regulatory environment. To sum up, this literature review is following an arranged logic line (perspectives of four stakeholders), therefore, it is unavoidable to neglect other important aspects, such as applied methodologies of those prior literatures.#p#分頁標題#e#

 

7.3 進一步研究Further research

Although the tremendous amount of researches already carried out on the impacts of non-audit services provided by incumbent auditors to auditor independence and audit quality, there is still much room for future research. Such research could provide more explicit proxies and valid conclusions, which are discussed in terms of relating to changes in current regulations and adoptions of more appropriate proxies. Expanding to data of more countries will also strengthen the theoretical basis.

 

7.3.1 當前規定的影響Consideration of impact from current regulations

Most researches focus on the proxies that are used to evaluate whether auditor independence and audit quality are compromised and ignore the influences created from changes in regulations. For example, as mentioned by a few researches, the issuance of Sarbanes-Oxley Act of 2002 (SOX) might change investors’ perceptions towards auditor independence and audit quality as the mandated disclosure of non-audit fees surprises investors. Therefore, changes in their perceptions results the changes on investment decisions (Abbott et al., 2011). If these effects of amendments of regulations and rules are neglected, misleading and invalid conclusions might be drawn and inconsistent research results cannot be explained. So it is an appropriate attempt to incorporate the impact of regulations in further studies.

 

7.3.2 采用直接和可靠的代理Adoption of direct and reliable proxies

As mentioned above, auditors’ propensity of issuance of going-concern is an indirect and noisy indicator to assess auditor independence and audit quality when compared with proxies such as discretionary accruals and financial statements restatements. It is widely acknowledged that the adoption of proxies will have significant effects on research results because indirect proxies might include other potential elements. Therefore, further studies might adopt indirect proxies as supporting evidence and more direct and reliable proxies should be accepted in order to guarantee the quality of research.

 

7.3.3 對發展中國家更廣泛的研究Wider research about developing countries

As discussed above, most data used in leading journals are based on U.S. and U.K., in which regulatory environments are sound and research information are more accessible than other countries. However, it is worth further investigating whether similar conclusions can be drawn under other regulatory environment, particularly those countries which are in rapid development thus more audit problems might occur. In addition, with the development of globalisation, there is a trend to establish more complete and sound regulations among these developing countries. Therefore, researches conducted on this area will provide guidance and suggestions on the developments of their relevant regulations.#p#分頁標題#e#

 

In conclusion, this paper provides an overall framework for literature reviews on influences of non-audit services provided by incumbent auditors from the perspectives of auditors, clients, financial reporting users, and regulators. Although limitations of this paper include insufficient up-to-date papers, ignorance of comparisons between methodology applied, and limited coverage of data selections, this literature reviews still provide some suggestion for further research, such as involvement of more current regulations, adoption of more direct and reliable proxies, and more concern about developing countries.