هزینه های حسابرسی و غیرحسابرسی و درک بازار سرمایه از استقلال حسابرس

This study provides evidence on whether auditor independence-in-appearance, proxied by earnings response coefficients, is related to the non-audit fee ratio (non-audit to total fees from a client) or client importance (total fees from a client as a percentage of the total revenues of the audit firm). The results from large samples over the period 2001–2006 show, contrary to popular belief and the findings of some prior studies, that there is no evidence of a relation between perceived auditor independence and the non-audit fee ratio. However, perceived auditor independence is negatively associated with client importance, consistent with the economic theory of auditing. Our paper adds to the literature by examining the relative importance of non-audit fee ratios and client importance as determinants of independence-in-appearance.

Regulators have long been concerned about the potential impact of audit and non-audit fees on auditor independence (SEC, 1979). The Securities and Exchange Commission (SEC) has repeatedly asserted that auditors must be independent in fact and in appearance (SEC, 2000). Independence-in-fact is defined by SEC as the auditor’s mental state lacking any bias, while independence-in-appearance is a public perception that the auditor is objective and unaffected by a financial interest in the client (SEC, 2001). Recent papers examining the relationship between audit and non-audit fees and auditor independence tend to provide evidence on independence-in-fact (e.g., DeFond et al., 2002, Frankel et al., 2002, Ashbaugh et al., 2003, Chung and Kallapur, 2003, Larcker and Richardson, 2004, Kinney et al., 2004, Srinidhi and Gul, 2007 and Lim and Tan, 2008). The few studies that provide evidence on whether audit and non-audit fees influence independence-in-appearance (Higgs and Skantz, 2006, Krishnan et al., 2005 and Francis and Ke, 2006) examine audit fees, non-audit fees, or non-audit fee ratio (non-audit to total fees from a client), but not client importance (fees from a given client to the total revenues of the audit firm). Economic theory suggests that auditors’ incentives to compromise their independence depend on client importance and not on non-audit fee ratio (DeAngelo, 1981, Ashbaugh et al., 2003 and Chung and Kallapur, 2003). Public attention has nevertheless focused on the non-audit fee ratio as causing perceptions of auditor independence impairment.1 In this study, we examine the relationship between audit and non-audit fees and capital market perceptions of auditor independence. Because prior studies do not address whether perceived auditor independence is a function of client importance or non-audit fee ratio, we investigate the relative importance of these two variables in influencing capital market perceptions of independence. The two fee-based metrics are not different measures of the same construct; rather, they are two separate concepts. For a given client, non-audit fee ratio measures the importance of consulting services relative to the total services provided by the external auditor. In contrast, client importance measures the significance of a client relative to the auditor’s portfolio of clients. Our study is important for conceptual and practical reasons. Conceptually, the results would show whether investors perceive independence as a function of non-audit fee ratio although academic studies conclude that client importance, and not non-audit fee ratio, impairs independence. Practically, regulators and audit committees might be misled into preventing non-audit services if the perception of auditor independence impairment is a function of client importance instead of non-audit fee ratio. Following prior literature (Teoh and Wong, 1993, Hackenbrack and Hogan, 2002, Ghosh and Moon, 2005 and Francis and Ke, 2006) we use earnings response coefficients (ERCs) as a measure of investors’ perceptions of audit quality.2 Following much of the prior literature (see Kothari, 2001), we compute ERCs from regressions of annual returns on annual earnings (levels and changes) and from regressions of returns on earnings surprises around quarterly earnings announcements. Using a large sample of firms with audit and non-audit fee data between 2001 and 2006, and after controlling for the other determinants of ERCs, we find that ERCs are negatively associated with client importance, but there is no evidence of an association between ERCs and non-audit fee ratio. Thus, our results indicate that investors perceive client importance, and not non-audit fee ratio, as compromising auditors’ independence. Further, when we decompose client importance into two components, audit fees and non-audit fees from a given client as percentages of the total revenues of the audit firm, we find that only the audit fee component is significantly negatively related to ERCs. Thus, the evidence suggests that investors are concerned about perceived auditor independence when client importance increases because of audit fees, but not because of non-audit fees. Because size might be correlated with non-audit fee ratio and client importance, we partition the sample by size into three groups and estimate our regression model separately for each size group. We find that ERCs are insignificantly associated with non-audit fee ratio in every size group. On the other hand, the significant negative association between ERCs and client importance holds for middle-sized firms only. One explanation for our results is that auditors’ concern for a loss of reputation if independence is impaired in engagements with large clients and less economic dependence in engagements with small clients are perceived as not impairing auditor independence. Another econometric-based explanation for the insignificant results is that there is little variation in non-audit fee ratio or client importance for firms in the large and small size groups compared to firms in the middle size group. Our results are robust to several sensitivity checks. First, the results hold after controlling for fixed firm effects, which control for omitted variables that are firm-specific (Himmelberg et al., 1999). Second, our results hold for periods before and after the Sarbanes-Oxley Act (SOX, 2002). Given that non-audit fee ratios were higher before the passage of SOX, an insignificant association between non-audit fee ratio and ERC suggests that non-audit fees are not perceived as impairing independence even during a period when non-audit fees were higher and the public was more sensitized to it. Non-audit fees are also less of a concern for the post-SOX period because of the creation of the Public Company Accounting Oversight Board (PCAOB) and other SOX provisions aimed at enhancing auditor independence. Third, while our primary regression results rely on pooled data, we find similar results using yearly regressions. Finally, our results are insensitive to the treatment of outliers. We contribute to the literature by examining auditor independence-in-appearance and whether non-audit fee ratio or client importance affects auditor independence. While the other related studies also use ERCs as a measure of independence-in-appearance, they do not compare the relative importance of the non-audit fee ratio and client importance as determinants of ERCs. In contrast to the findings of prior studies, our results question the long-held belief that higher non-audit to total fee ratios are perceived by investors to compromise auditor independence. Moreover, our results are more generalizable because we include a comprehensive sample spanning a period of six years and examine sub-periods with relatively high and low levels of non-audit fee ratio. We organize the rest of the paper as follows. Section 2 establishes the links between audit and non-audit fees and perceptions of auditor independence. Section 3 discusses the research design, and Section 4 describes the sample selection procedure. Section 5 reports the results and Section 6 concludes the paper.

In this paper we provide evidence on the determinants of auditor independence-in-appearance using earnings response coefficients (ERCs) as a proxy for investor perceptions of earnings, and therefore audit, quality. An examination of auditor independence-in-appearance is important because regulators and the AICPA have emphasized that auditors should be independent not only in fact but also in appearance. Although economic theory suggests that auditor incentives to compromise independence depend on client importance, SEC rules and public surveys indicate that perceived auditor independence is a function of non-audit fee ratio. Accordingly, our paper examines the relative importance of client importance and the non-audit fee ratio as determinants of perceived auditor independence. Based on a large sample of firms over the years 2001–2006, we find that the non-audit fee ratio (ratio of non-audit to total fees from a given client) is insignificantly associated with ERCs. In contrast, client importance (ratio of fees from a given client to the total revenues of the audit firm) is significantly negatively associated with ERCs. Our results suggest that a negative investor perception exists toward high levels of client importance, not toward non-audit fee ratio. Our results are robust to the inclusion of fixed firm effects to control for correlated omitted variables and to periods before and after the Sarbanes-Oxley Act. We further estimate our regressions separately for three size groups to explore whether the relation between perceived auditor independence and auditor related fees differs for different size firms. We find that ERCs do not vary with non-audit fee ratio in any size group. However, the relation between ERC and client importance holds only for middle-sized firms. The ERCs for the large and small size firms are both are statistically insignificant. We conjecture that because auditors are likely to be especially concerned about the loss of reputation from a perception that independence is impaired for large clients, and because auditors are less economically dependent on small clients, investors are less concerned about loss of independence for engagements with large and small clients. Our study contributes to the literature on auditor independence by examining whether the non-audit fee ratio or client importance affects auditor independence-in-appearance. Although prior studies provide some evidence on independence-in-appearance (Higgs and Skantz, 2006, Krishnan et al., 2005 and Francis and Ke, 2006), they do not compare the relative importance of non-audit fee ratio and client importance. Moreover, these studies focus on the period around the initial fee disclosures in 2001. By examining the relative importance of client importance and non-audit fee ratio as determinants of independence-in-appearance with a large sample covering the years 2001 to 2006, our paper provides additional market-based empirical evidence on the influence of audit and non-audit fees on perceived auditor independence.