This study argues that the incremental audit productions associated with issuing a qualified opinion are difficult for public accounting firms to recoup through audit fees alone. The research is based on a Nigerian data, where the effect on audit fees, however, occurs only on a lagged opinion basis, while the effect on NAS fees occurs only on a concurrent opinion basis. The studies that report an increase in independence perceptions due to NAS find this only when separate personnel perform NAS. Knapp (1991) used these two aspects (misstatement detection probability and disclosure probability) as dependent variables in a study of factors (e.g., audit firm size) used by audit committees to judge audit quality and found that these two aspects were affected by his independent variables differently. Although auditor independence was not a part of this study, our results suggest that these two aspects should be considered separately. In the following hypothesis development, the ability of the auditor is assumed to be a constant, so that it is the knowledge aspect of competency that will be addressed. ANOVA planned comparisons yielded similar results for this and the other dependent variables. Equality of variance tests for each comparison indicated that pooled sample t-tests were appropriate for all analyses except one where the Satterthwaite statistic was used.
Today, our National Assembly, the security & exchange commission (SEC), the press, and the public are scrutinizing the accounting profession. The recently legislated money market act (2002) establishes new regulations governing the financial reporting of public companies. One concern has been the rapidly growing trend of audit firms providing extensive non-audit services (NAS) to the firms they audit. The ability of the auditor to remain independent while receiving fees for these services has been questioned by the SEC. Both the larger financial stake (i.e., increased total fees received) and the conflicting roles of the auditor (i.e., the auditor acts as a client advocate when performing non-audit services, in opposition to the public advocacy role he or she plays as an auditor) might affect the auditor’s objectivity and independence from client influence. Although the new money market Act addresses this issue by prohibiting nine non-audit services, other non-audit services (e.g., tax services) are still permitted, so that the independence issue will continue to be of concern.
Also in response to this concern, the SEC (2000) recently issued new independence regulations. In reference to NAS, the SEC now requires all public companies to disclose, in their proxy statement, all fees paid to the external auditor, disaggregated into 3 categories: audit fees, information technology service fees, and fees for other non-audit service. These new disclosures reveal an average NAS to audit fee ratio well over 200 percent (Weil and Tannenbaum 2001; Frankel et al. 2002). Members of the public accounting community have argued that these NAS fees do not affect independence (Weil and Tannenbaum 2001), while prior academic research investigating the effect that various NAS have on perceptions of auditor independence has reported mixed results. Of ultimate concern is how the provision of non-audit services affects the credibility and perceived reliability of the information contained in a public company’s financial statements. This study uses documented average NAS fee levels, in a form publicly available, to experimentally test the effect of NAS provision on investors’ perceptions and decisions. Graduate business students, proxying for investors, are presented with financial statements and auditor information for a large company. The fee levels and types of audit firm service are varied between subjects. The study also addresses the hypothesized desegregations of audit quality into competence (knowledge) and objectivity (independence) components (DeAngelo 1981). Participants judged these two aspects of audit quality separately to determine how each affects the overall perception of audit quality. In addition, I investigate how investor judgments of financial statement reliability and investment attractiveness are affected by the presence of non-audit services.
Results indicate that the provision of NAS reduces investors’ judgments of auditor independence, audit quality, and the attractiveness of the firm as an investment, but not their auditor knowledge or financial statement reliability judgments. Results also suggest that it is the nature of the NAS and not the magnitude of the total fee that affects the participants’ judgments. Interestingly, when one of the NAS services (information technology) was separately identified investor concerns about independence were reduced.
BACKGROUND AND LITERATURE REVIEW
Generally Accepted Auditing Standards (GAAS), first written in 1947, state that an auditor must maintain independence in mental attitude in all matters relating to the assignment. Independent auditors furnish critical assurance that the financial statements have been examined by “an objective, impartial and skilled professional” (SEC 2000, p. 2). This assurance adds credibility to financial statements, lowers information risk, and facilitates capital formation, adding value to the entire capital markets system (Elliot and Jacobson 1998; Carmichael 1999; Kinney 1999). This independence can be maintained through external constraints (i.e., legislation and regulation) or through the profession itself, which will maintain independence to preserve its market value (Kinney 1999). Overall, the business press appears skeptical of auditors’ ability to withstand client pressure, emphasizing the effect of economic interests. An Investment News article (November 20, 2000) calls consulting services “the tail that wags the dog” in the accounting industry.
The SEC issued a new set of rules governing auditor independence on November 21, 2000, and included an auditor fee disclosure requirement. Since February 5, 2001, companies must disclose in their proxy statements, the total fees billed for services rendered by the principal accountant (the external audit firm) disaggregated into three categories: audit services, information technology services, and all other services (AICPA 2000; SEC 2000). Study of these disclosures reveals that nearly all firms purchase non-audit services from their audit firm (Abbott et al. 2001). Frankel et al (2001) show that NAS fees average over 246 percent of audit fees for a cross-section of publicly traded companies (Frankel et al. 2001), while Weil and Tannenbaum find a NAS/audit fee ratio of almost 300 percent for a sample of Standard & Poor‘s 500 firms. Although the Big 5 auditing firms continue to maintain that their independence is not impaired by these services, recent problems at Enron, WorldCom, and other public firms have brought accounting and auditing concerns to the attention of Congress and the public. Recently passed legislation prohibits nine NAS, including financial information services and internal auditing services (Money Market Act 2002).
The SEC is not only concerned about independence-in-fact, also about the perceptions or appearance of independence. They believe perceptions of the auditor’s independence (objectivity) affect investor confidence in financial statements, and affect financial statement users’ decisions based on those statements (SEC 2000). Although Wallman (1997) and an AICPA white paper (1997) suggest that the provision of NAS does not impair apparent independence, independence perception responses from survey participants do not always bear this out (Jenkins and Lowe 1999; Farmer et al.1987; Pany and Reckers 1984; Schulte 1965).
Earlier research used NAS fee levels ranging from three to 90 percent of audit fees, when realistic levels were thought to be 25-40 percent or less and 90 percent was considered unusually high. This study utilizes a NAS to audit fee ratio (293 percent) that is realistic for a large (S&P 500) firm (Weil and Tannenbaum 2001; Frankel et al. 2002).
Also, in all the previously mentioned experimental studies, the specific non-audit service was identified for the experimental participants. With the exception of information technology services, the SEC disclosure rule includes all types of service under the same heading, “other fees.” Although firms have the option of listing the “other” services and even the portion of the other fees paid for each in a footnote, this is not required and the accounting industry has objected to disclosing this information (SEC 2000). The NAS information in this study follows the SEC mandated format.
In addition, many prior experimental studies specified whether the NAS was performed by the external auditors themselves or by separate personnel within the external audit firm (e.g., Pany and Reckers 1984; McKinley et al. 1985; Lowe et al. 1999; Swanger and Chewning 2001). This study makes no such separation, limiting NAS information to the fee information currently available to the investor in the real world. In summary, this study focuses on how realistic amounts and disclosures about NAS affect investor perceptions of auditor independence.
Another important difference between this study and prior studies is that I investigate independence as one of the two separate aspects of audit quality. Audit quality is considered to consist of a competency (knowledge) component and an objectivity (independence) component (Elliot and Jacobsen 1998; Carmichael 1999). DeAngelo (1981) defines audit quality as the combined probability of (1) detecting a material misstatement in financial statements (a competency issue), and (2) disclosing that misstatement (an independence issue). Previous experimental literature on the effect of NAS on financial statement users’ independence perceptions did not address the knowledge issue in their experimental instruments.
By providing NAS to a client, the auditor potentially has more extensive knowledge about the company, e.g., the auditor knows the extent and effectiveness of the information technology used within a company if he or she had a hand in designing the system (Jenkins and Krawczyk 2001). Bonner and Lewis (1990) found that task-specific knowledge and experience affected auditor performance more than general business knowledge or years of experience, suggesting that the auditor may gain useful knowledge by performing NAS. Prior studies have hypothesized that “knowledge spillover” can occur when NAS is provided (Beck et al. 1988a and b). The SEC (2000) considered the knowledge spillover effect as well. If investors believe this knowledge spillover exists, the auditor who performs NAS might be perceived to be more knowledgeable and competent regarding their client, and more likely to detect an error in the system or the output (financial information). Thus, the first hypothesis, stated in alternative form, posits a positive relation between the level of NAS provision and perceptions of an auditor’s knowledge about the client.
Hypothesis one: Investors’ evaluations of auditor knowledge about a client firm will be higher in the presence of NAS than when NAS is not provided.
The SEC considers that “an auditor is not independent if a reasonable investor, with knowledge of all relevant facts and circumstances, would conclude that the auditor is not capable of exercising objective and impartial judgment” (SEC 2000, p.3). The fees received by auditors from their audit clients, especially the large amounts revealed in the new proxy disclosures, seem to create an economic bond between client and auditor that could tend to bias an auditor in favor of a client. Also, the client advocacy role that the auditor assumes by providing NAS might reduce the auditor’s objectivity during the audit. Hypothesis two, stated in alternative form, posits a negative relation between the provision of NAS and investors’ perception of auditor independence.
Hypothesis two: Investors’ evaluation of auditor independence will be lower in the presence of NAS than when NAS is not provided.
Hypotheses one and two posit that the provision of NAS will affect perceptions of the two aspects of audit quality (i.e., auditor knowledge and auditor independence) in opposite directions. Since the extent and relative weighting of these two aspects, and even the functional form of how these two aspects combine is unknown, the effect of NAS on audit quality is investigated as a research question.
Research Question 1: How will the provision of NAS affect investors’ evaluation of the overall quality of the audit?
Reliability of the Financial Statements
A major purpose of an external audit is to attest to the reliability of a client’s financial statements, adding credibility to the statements. Whether or not NAS affects investors’ perceptions of financial statement reliability is an empirical issue of considerable interest to the SEC. I investigate this issue as noted in research question two.
Research Question 2: How will the provision of NAS affect investors’ evaluation of the reliability of the financial statements?
The above hypotheses and questions assess the effect of NAS on financial statement users’ perceptions. The third research question considers how these perceptions (judgments) may affect investment decisions. Frankel et al.’s (2001) finding of a relation between the ratio of NAS fees to total auditor fees and stock prices suggest that there will be an effect, while Ashbaugh et al. (2002) find no market reaction to audit fee disclosures. The extensive attention given to the issue of auditor independence, especially in recent years, and the accompanying increase in NAS that may threaten independence, lead to the prediction that financial statement users’ decisions based on the audited financial statements could be affected by the provision of NAS
Research Question 3: How will the provision of NAS affect investors’ judgments regarding the attractiveness of a firm as an investment?
Type of fee (audit or non-audit service)
The renewed concern about auditor independence is due to the increasing magnitude of audit firms’ NAS fees, compared to the audit fee itself. There are two issues of concern. One is that non-audit services create a client advocacy role for the auditor as the services are provided for the benefit of the client. This role is in conflict with the public advocacy role the auditor is supposed to hold as an external auditor. In addition, some of the non-audit services imply management functions are provided which could also impair the public advocacy role of the auditor. The other issue is the magnitude of the fees themselves. When NAS fees are more than double the audit fee, the total fees are over three times as large as the audit-related fee. Perhaps any effect of non-audit service provision on investor perceptions (of auditor independence, audit quality, etc.) is a reaction only to the magnitude of the fees, and not to the nature of the fees. I investigate the effect of fee magnitude versus the client advocacy role through research question four.
Research Question 4: How will an increase in the audit fee only (without the presence of non-audit services) affect investor’s judgments of (a) auditor independence, (b) knowledge, (c) audit quality, (d) financial statement reliability and (e) investment attractiveness?
Type of non-audit service (information technology provision or other)
Different non-audit services provide different threats to auditor independence. Internal audit services (sometimes considered to be management activities) are considered independence threatening, while other services, such as tax preparation, are of lesser concern (SEC, 2000). The amounts paid for both these services are included under the same heading (“other fees”) in the required proxy disclosure. Investors viewing “other fees” may differ in their perceptions of how much the implied services impair independence. Interestingly, financial information system-related services, considered to be more threatening to independence than many other services, must be listed separately in the audit fee disclosure (SEC 2000). In addition, the recently passed Sarbanes-Oxley Act (2002) specifies the elimination of several NAS, including information technology services. I investigate whether identification of a potentially problematic NAS affects investors’ perceptions differently from a more general and vague category of NAS.
Research Question 5: How will auditor provision of information technology related services affect investor’s ratings of (a) auditor independence, (b) knowledge, (c) audit quality, (d) financial statement reliability and (e) investment attractiveness?
The experiment used a between-subjects design in which participants received one of four cases. All information was identical except that the services provided by the external auditor were different for each case. The participants were informed, through the audit fee disclosure segment of the client’s proxy statement, that the auditor received either (1) audit fees only (NAIRAIT), (2) both audit and non-audit fees, listed as “other fees” (NAIRA/NAS), (3) a high level of audit fees (High NAIRA), or (4) audit, information technology (IT) fees, and other fees (NAIRA/NAS/IT). for this condition, These two cells (AUDIT and NAIRA/NAS) were used to test the two hypotheses and the first three research questions.
The High NAIRA cell presented the case where no NAS were reported, but audit fees were set at the sum of audit and NAS fees used in the NAIRA/NAS cell. Finally, in the NAIRA/NAS/IT cell, NAS was disaggregated into fees for both information technology and “other.” Since fees for IT services were about 30 percent of the fees for NAS for the firms in Weil and Tannenbaum’s study that purchased IT services, IT services fees in the NAIRA/NAS/IT condition were set at 30% of the NAS from the NAIRA/NAS cell. The remaining NAS fee remained in the “other fees” category
The goal was to generate a sample firm that was typical of large firms that are considered for investment purposes. Since the mean S&P500 total assets were higher than the third quartile of S&P500 firms, a sample firm was chosen whose total assets were somewhat lower than the mean total assets of the S&P500.
To test the effects of NAS on auditor independence, auditor competency (defined as knowledge and ability, with ability held constant), audit quality, and financial statement reliability participants were asked to evaluate (1) the knowledge level of the auditor (defined as the auditor’s ability to discern a financial reporting problem, if one exists), (2) the independence of the auditor (defined as the auditor’s ability to be objective and unbiased, likely to disclose a financial reporting problem, if one is discerned), (3) the overall quality of the audit, and (4) the reliability of the financial statements. In addition, participants were asked to rate (5) the attractiveness of the firm as an investment to investigate how their perceptions affect their decisions.
REGRESSION COEFFICIENTS OF INDEPENDENT VARIABLES
ON AUDIT QUALITY JUDGMENTS
QUAL = α0 + β1 KNOW + β2 INDEP + ε (1)
Coefficient T-Statistic P-Value
Intercept 0.61039 1.63 0.11
KNOW 0.49512 9.20 <.01
IND 0.44538 9.30 <.01
QUAL = participants’ judgments of audit quality, on a 0-10 scale with 10 the highest,
KNOW = participants’ judgments of auditor knowledge, on a 0-10 scale with 10 the highest,
INDEP = participants’ judgments of auditor independence, on a 0-10 scale with 10 the highest,
ANALYSIS OF RESULTS
Manipulation and other checks.
Data from the post-experimental questionnaire indicate that the case materials successfully manipulated the total audit fee levels. Participants were asked to recall the total fees paid to the auditor and how much of these fees were for non-audit services. Responses were made on a 0 to 10 million-dollar scale, with numerical indications at 2.5, 5, and 7.5 million dollar intervals. There were no statistically significant differences between the total fee estimates for the NAIRA/NAS (mean 9.18, standard deviation 1.97), High NAIRA (mean 8.80, standard deviation 2.39), and AUD/NAS/IT (mean 8.16, standard deviation 3.00) conditions. The mean estimate of 2.72 million (standard deviation 0.5) of total fees paid to the auditor for the audit only (AUDIT) condition differs statistically significantly (p 0.40), but were statistically significantly different from both the NAIRA/NAS (mean 6.72, standard deviation 2.29) and NAIRA/NAS/IT (mean 5.27, standard deviation 1.94) cells, (p <.01 for both comparisons). However, there was a marginally statistically significant (p = .10) difference between the NAIRA/NAS and NAIRA/NAS/IT mean estimates of non-audit services.
Participants were also asked to rate how understandable and realistic the materials were on a scale of 0-10. The overall mean ratings of 6.45 for understandability and 6.91 for realism were not statistically
Hypothesis one: Perceived auditor knowledge
Hypothesis one states that investors’ evaluation of auditor knowledge about a client firm will be higher in the presence of NAS than when NAS is not provided. Mean knowledge judgments are shown in Table Two, along with mean judgments for the other dependent variables. An analysis of variance (ANOVA) on the knowledge judgments shows that the fee manipulations did not have a statistically significant effect on knowledge perceptions (F = 0.11, p = .95). More specifically, related to the first hypothesis, the mean knowledge judgment of 6.15 for the AUD/NAS cell was not statistically significantly smaller (t = 0.30, p = .62, one-tailed) than the AUDIT mean of 6.33. This suggests that the provision of NAS is not perceived to increase (or decrease) an auditor’s knowledge about the client firm.
(Insert Table Two here)
Hypothesis two: Perceived auditor independence
Hypothesis two posits that investors’ evaluation of auditor independence will be lower in the presence of NAS than when NAS is not provided. An ANOVA indicates a statistically significant effect of the fee manipulations on independence judgments (F = 5.05, p < .01). More specifically, the mean auditor independence evaluation of 4.06 for the NAIRA/NAS cell was statistically significantly lower (t = 3.72, p < .01) than the mean independence evaluation of 6.33 for the AUDIT cell. This suggests that investors do perceive a decline in the independence of the auditor when an audit firm provides non-audit services to its audit client.
Research question one: Perceived Audit Quality
The first research question asks how the provision of NAS affects investors’ evaluation of audit quality. An ANOVA of perceived audit quality judgments indicated a marginally significant effect of the fee manipulation (F = 2.36, p = .08). Focusing specifically on how the presence of NAS fees in addition to the audit fee affects audit quality perceptions, the mean audit quality evaluation of 5.25 for the NAIRA/NAS cell is statistically significantly lower (t = 2.20, p = .03) than the mean audit quality evaluation for the AUDIT cell of 6.42. This suggests that the provision of NAS affects audit quality perceptions negatively.
To further explore the manner in which knowledge and independence are combined the following regression was run,
QUAL = α0 + β1 KNOW + β2 INDEP + ε
QUAL = participants’ judgments of audit quality,
KNOW = participants’ judgments of auditor knowledge, and
INDEP = participants’ judgments of auditor independence.
As reported, regression results from equation (1) indicate that both knowledge (β1 = 0.495, t = 9.20, p < .01) and independence (β1 = 0.445, t = 9.30, p .30). This suggests that it is the type of fee and not the magnitude of the fee that affects investors’ perceptions.
Research question five: Type of Non-Audit Service
The final fee manipulation included the desegregation of the NAS fee into information technology (IT) fees, and “other” fees to see if the provision of a NAS service where an auditor might be auditing his own (firm’s) work affects perceptions differently than NAS listed under a generic “other” designation. Results indicate that there is a difference. When comparing the AUDIT cell judgments to the judgments for the NAIRA/NAS/IT cell, the only statistically significant difference was for the independence ratings (t = 1.83, p = .04). Unlike the AUDIT to NAIRA/NAS comparisons, there were no statistically significant differences for audit quality or investment attractiveness. This suggests that the disclosure or presence of information technology services mitigates some of the perceived audit quality, financial statement reliability, and investment attractiveness rating impairments created by NAS in general.
DISCUSSION, LIMITATIONS, AND CONCLUSION
The proposition that NAS provision beneficially affects perceived audit quality due to the increased knowledge it provides to the audit firm about the client is not supported by this investigation. However, perceptions of auditor independence were statistically significantly lower for the cases where the audit firm performed NAS in addition to audit services, compared to cases where only audit services were provided. This decrease in independence perception occurred even when the audit fees themselves were high (i.e., equal to the combined audit and NAS fees paid), suggesting that it is the nature of NAS and not the fee amount that concerns investors. Audit quality and investment attractiveness perceptions were also lower when the client’s auditors performed NAS. Financially statement reliability perceptions were not significantly affected by the presence of NAS. A possible explanation is that there are other non-measured factors (e.g., the perceived integrity of the client firm) that affect decision-makers when evaluating the financial statements themselves.
Interestingly, when information technology (IT) was identified as part of the non-audit services, the decrease in perceived audit quality and investment attractiveness did not occur. Also, the decrease in independence perceptions was smaller for NAS including IT than when all NAS was listed under an “other fees” heading. A possible explanation for this unexpected finding is that knowing the specific non-audit service performed mitigated the uncertainty of the “other” designation, which might have been perceived to include suspicious services (such as SPE advisement). This finding might interest both legislators and firms, when deciding on the extent of auditor fee information to require or provide voluntarily.
Regression results support the belief that perceptions of audit quality are affected by both auditor knowledge (competency) and auditor independence (objectivity) perceptions (DeAngelo 1981; Eliott and Jacobsen 1998; Carmichael 1999). Results also suggest that these two factors are approximately equally weighted, and the weighting is not affected by the presence of NAS.
There are several limitations to this study. The actual participants in this investigation may be atypical of investors in general. For example, student participants may have more accounting and auditing knowledge than average individuals. On the other hand, participants in this study do not have as much training as professional investors, another large component of the investing community. However, non-professional financial statement users represent the class of investors that the SEC seeks to protect. The investment scenario used may be unrealistic. For time purposes, the amount of information was limited compared to the amount publicly available. Also the audit fee disclosure was purposely made readily accessible, while real-life investors would have to intentionally search for the information in the firm’s proxy statement. This was necessary, however, to evaluate the value of this information to financial statement users. This study indicates that, when explicitly available, this information affects investors’ investment perceptions and judgments.
In summary, investors do not believe the accounting industry’s contention that NAS does not affect independence. This study’s results indicate that audit fee disclosures recently required by the SEC provide relevant information to individual investors. Congress (Stout 2002), the Securities and Exchange Commission (Glater 2002), and the Financial Accounting Standards Board (Norris 2002) continue to scrutinize the issue of auditor independence, and new accounting legislation and controls have been passed (money market 2002). This study provides insight into the effect that the curtailment of NAS might have, by documenting that investors are concerned about the issue. This is especially relevant during this time of increased scrutiny and regulation of the accounting and auditing profession.
Abbott, L., S. Parker, G. Peters, and D. V. Rama. 2001. Audit, non-audit and information technology fees: Some empirical evidence. Working paper. University of Memphis.
Asbaugh, H., R. LaFonda, and B.W. Mayhew. 2002. Do non-audit services compromise auditor independence? Further evidence. Working paper, University of Wisconsin-Madison.
Beck, P.J., T.J. Fricka, and I. Solomon. 1988a. A model of the market for MAS and audit services:knowledge spillovers and auditor-auditee bonding. Journal of Accounting Literature : pp. 50-64.
Bonner, S. and B. Lewis. 1990. Determinants of Auditor Expertise. Journal of Accounting Research (supplement): 1-21.
Carmichael, D. R. 1999. In search of concepts of auditor independence. The CPA Journal 69 (May): 38-43.
DeAngelo, L. 1981. Auditor size and audit quality. Journal of Accounting and Economics 3 (December): 183-199.
Elliott, R. K. and P. D. Jacobson. 1998. Audit independence concepts. The CPA Journal, 68 (May): 30-37.
Frankel, R. M., M. F. Johnson and K. K. Nelson. 2002. The relation between auditors’ fees for non-audit fees and earnings quality. Working paper, Stanford University.
Jenkins, J. G. and K. Krawczyk. 2000. The relationship between nonaudit services and perceived auditor independence. Working paper, North Carolina State University.
Jenkins J. G. and D. J. Lowe. 1999. Auditors as advocates for their clients: Perceptions of the auditor-client relationship. The Journal of Applied Business Research 15 (Spring): 73-78.
McKinley, S., K. Pany and P. M. J. Reckers. 1985. An examination of the influence of CPA firm type, size and MAS provision on Loan Officer Decisions and Perceptions. Journal of Accounting Research 23 (Autumn): 887-896.
Norris, F. 2002. Promises of a nimbler accounting board. New York Times (April 25).
Palmrose, Z-V. 1999. Empirical research on auditor litigation: Considerations and Data. Sarasota, FL: AAA.
Pany, K. and P. M. J. Reckers. 1984. Non-audit services and auditor independence- A continuing problem. Auditing: A Journal of Practice & Theory 3 (Spring): 89-97.
Stout, D. 2002. House approves accounting oversight inspired by Enron. Wall Street Journal. April 24.
Swanger, S. L. and E. G. Chewning, Jr. 2001. The effect of internal audit outsourcing on financial analysts’ perceptions of external auditor independence. Auditing: A Journal of Practice & Theory. 20 (September). 115-129.
Wallman, S. M. H. 1996. The future of accounting, part III: Reliability and auditor independence. Accounting Horizons. 10 (December). 76-97.
Weil, J., and J. Tannebaum. 2001. Big companies pay audit firms more for additional services. Wall Street Journal (April 10):C1.
Wines, G. 1994. Auditor Independence, Audit Qualifications and the Provision of Non-Audit Services: A Note. Accounting and Finance. 34 (May): 75-87.