The pricing of statutory audit services in Greece

Abstract

In this study, we investigate the determinants of corporate audit fees following the liberalization of the statutory audits market in Greece. Though the Greek statutory audits market is liberalized, it is relatively less matured, less competitive and, non- or less-litigious compared to those in the U.K. and U.S. Using data on fees paid to external auditors by 145 companies listed on the Athens Stock Exchange as of December 2000, we note that our results are similar to those of many prior studies conducted on matured, competitive, litigious audit environments. Specifically, we find that the size of an auditee, the number of hours spent on a particular audit engagement, the size of an audit firm and the financial condition of the auditee have positive and significant influence on the amount of fees charged by auditors. In addition, we find that auditor change has a significant negative effect on audit fees, especially in the large auditee segment of the market. Further, as our additional analysis shows, we document that audit fees and audit hours are endogenously related. We perform several sensitivity tests, the results of which suggest that our primary results are robust.

Keywords:

• Audit fees
• Pricing audit services
• Statutory audit market
• Athens Stock Exchange
• Greece

Acknowledgements

We acknowledge the assistance of Stergios Athianos and Nikos Grantas in data collection. We thank George Alifantis of Ernst & Young, Greece and Abraham Loutridis of KPMG, Greece for responding to our inquires. We are also indebted to the Institute of Certified Auditors of Greece for providing us with the proprietary data on audit fees and audit hours. Further, we appreciate the insightful comments by anonymous reviewers on earlier drafts of the paper.

Notes

3 A historical survey is unnecessary here, as quite a number of review articles have recently appeared in the literature, notably, a comprehensive review and synthesis by Walker and Johnson (1996). It is, however, worth noting that, as at date, only six European countries have been studied in the extant literature, namely, Finland, France, Ireland, Norway, the Netherlands, and U.K. (see GonthierBesacier & Schatt, 2007; Niemi, 200-5). The U.K. alone has been studied eight times (see Cobbin, 2002:5).

4 SELE was established in 1979 by partners of the then Big-8 international audit firms that were operating in Greece. Branches of these audit firms were established in Greece during the 1960s with the inflow of foreign investments to the country.

5 Understandably, the competitiveness of the Greek audit market is incomparable to those of the U.K. and U.S. but the degree of its competitiveness was enhanced significantly following the liberalization of the market in 1992.

6 Several prior studies have examined similar variables, though slightly differently measured. The variables have been found to be robust across different samples, different periods, and different countries. However, our study is the first to test the effect of audit hours on audit fees in a non-Anglo-Saxon environment and, the first to test the inter-relationship between the two variables.

7 Though the Greek audit environment is non-or less-litigious, we investigate this determinant because prior studies on auditor economic incentives and type of audit report (see, e.g., Vanstraelen, 2002) find that the exposure to third-party lawsuits also drives the likelihood of auditors issuing qualified audit opinions in other continental European non- or less-litigious audit environments, as it had been in litigious environments such as the U.K. and U.S. In addition, the expected costs of losing a client and loss of reputation may also be of concern for auditors operating in Greece when a client's financial condition worsens. Auditors in such circumstances would have the incentive to expand substantive audit tests, which in turn, would require levying of higher audit fees.

8 The Zmijewski's (1984) index is derived with the following formula: −4.336 − 4.513 (return on assets) + 5.679 (gearing) + 0.004 (liquidity). It is an estimated risk index of the financial condition of a company, which indicates the company's susceptibility to failure. Higher values of the index suggest higher propensity of a company failing. The reverse is true as well.

9 The Hellenic Auditing Standards and Professional Ethics require auditors to include in their audit reports inter alia any observations or remark on material matters taken into consideration by the auditors to support the audit conclusions (Body of Certified Auditors–Accountants, 1999:180). These remarks are similar to “subject to” qualifications that were in use in the past in both the U.K. and U.S.

10 The reasons for the exclusion of the financial companies are fourfold. First, we follow the practice in prior research that had applied the established audit fee model to industrial companies (e.g., Copley and Douthett, 2002; DeFond, Francis, & Wong, 2000; Francis, 1984; Simunic, 1980). Second, financial companies have some unique characteristics—risk and complexity—that impact audit pricing and set them completely apart from non-financial companies. For example, Fields, Fraser, and Wilkins, (2004) suggest that audit fees are significantly higher for banks because of having more transactions and higher degree of credit risk. Third, most of the variables usually investigated in prior studies are meaningless for financial companies and do not easily lend themselves to measurement due to the nature of their operations. For example, financial ratios such as gearing, current/quick ratio, inventory as a percentage of total assets and, receivables as a percentage of total assets cannot be computed for financial institutions. Fourth, empirical evidence suggests that the relationship between bank regulatory bodies and audit firms is important due to high levels of litigation risk in the industry (see Palmrose, 1988). This is particularly true in Greece, where the banking sector is always under public scrutiny because of the scandal that involved the Bank of Crete in 1988.

11 There are many reasons other than fee reduction why a company may change its external auditor (see, e.g., Eichenseher & Shields, 1983; Simon & Francis, 1988). For example, a company may change to a Big-5 audit firm if it believes that such a change would add credibility to its financial statements in the investment community. The change could also be due to a company's policy to rotate auditors intermittently.

12 None of the Chow test statistic for the other variables in Eq. (2) is statistically significant except for the AUD_HRS variable (detailed results not tabulated).

13 It is important to note that the Chow test failed to find any statistical difference in the regression parameter for AUD_TYPE between the large and small auditee segments (see footnote 10 [χ²-statistic = 0.31, two-sided p-value = 0.5766]).

14 The model estimated with the two additional variables, Eq. (3) is specified as follows:(3) $\text{AUD}\_\text{FEE}=b_0+b_1\text{CO}\_\text{SIZE}+b_2\text{FIT}\_\text{AUDHRS}+b_3\text{AUD}\_\text{TYPE}+b_4\text{FIN}\_\text{CON}+b_5\text{PROFT}+b_6\text{AUD}\_\text{CHG}+b_7\text{AUD}\_\text{OPN}+b_8\text{ISSUE}+b_9\text{INVENT}+b_{10}\text{FRGN}\_\text{SUB}+ϵ\text{.}$(3)