Client risk and audit fees

Loss exposures to auditors can be measured differently than those methods proposed in Simunic (1980). One example is Beatty (1993) in which the loss exposure relating to litigation risk is more directly measured on the IPO firms. More specifically, using the setting where IPO clients are more likely to exhibit financial distress than established clients, Beatty (1993) shows that auditors charge a conditionally higher fee for clients that were subsequently delisted, filed for bankruptcy, or were the subject of a shareholder lawsuit.

In Beatty (1993), 223 IPO firms (24%) were delisted because of financial distress, 90 sample firms filed bankruptcies, and 40 lawsuits were filed against the sample firms. Beatty also uses IPO first-day return to control for the implicit insurance hypothesis which suggests that initial return will be negatively related to auditor compensation.

Another example is Lyon and Maher (2005), where alleged client misconduct is used as the proxy for loss exposure to examine whether the alleged client misconduct is associated with higher audit fees in the United States. In particular, the alleged misconduct in Lyon and Maher is that the clients paid bribes to high-level foreign government officials in return for favors. In the period studied (1974), bribery was not illegal since the FCPA (Foreign Corrupt Practice Act) was not enacted until 1977.

Lyon and Maher (2005) argue that client misconduct could affect both of the components of business risk defined by the American Institute of Certified Public Accountants (AICPA). According to the AICPA (1992), business risk has two components: (1) the client’s business risk, which is associated with the client’s continued survival and wellbeing, and (2) the auditor’s business risk of being associated with a particular client irrespective of whether an audit failure is asserted. Allegations of client misconduct could lead to a decline in the well-being of the client for several reasons, including developing public relations campaigns and legal defenses, facing consumer boycotts, and facing new regulations. Alleged client misconduct could also damage the reputation of the auditor for being associated with the client.

Lyon and Maher (2005) further provide several reasons why bribery increases a client’s business risk for auditors. Paying bribes to high-level foreign government officials risks incurring the direct costs of shareholder litigation and regulator intervention. Other potential costs include management turnover as a consequence of negative publicity aboutbriber}', the costs of investigating bribery, the costs of building an internal control infrastructure to prevent bribes, the costs of hiring third parties to shield company executives from bribe payments, and the lost profits from forgoing business in bribery-prone countries to avoid bad publicity. Furthermore, performing an audit in an environment of questionable management integrity created by bribery activities heightens audit business risk and can even have spillover effects on perceived audit quality.

Using a sample of 82 companies of which 25 of them report the paying of bribes, Lyon and Maher (2005) find that the proxy for alleged misconduct significantly increases audit fees.

Audit approach and audit fees

The auditing profession has incentives to improve the efficiency and effectiveness of doing audits, especially in a dynamic environment where the reporting entities are getting larger and more complex and the users are demanding more reliable and forward-looking information. Business Risk Auditing (BRA) is an audit approach that requires auditors to pay significant attention to the fundamentals from which the account balances are derived. More specifically, BRA requires auditors to develop a deep and comprehensive understanding of auditees’ industries, business models, strategies, and processes.

As mentioned in Bell et al. (2008), whether BRA is indeed an improvement in the efficiency and effectiveness of audits is not clear. On one hand, BRA calls for the use of expanded evidentiary bases, more comprehensive risk assessments, deployment of professionals who possess the requisite knowledge and competencies to perform these more complex risk assessments, and the redirection of audit resources in accordance with such assessed risks. On the other hand, BRA is a departure from traditional audit practices that can lead to under-auditing because of the substantial reductions in audit effort (Berkowitz and Rampell 2002; Weil 2004).

Using data consisting of labor hours, fees, and key client and engagement characteristics for 165 U.S. audits conducted by a Big-4 firm during 2002 and 2003, Bell et al. (2008) examine the impact of BRA on labor mix, total labor usage, and audit fees. Their principal findings are as follows. Comparing 2002 actual labor mix and levels to 1989 benchmarks (pre-BRA benchmark) reveals that the mean share of partner and manager labor in the labor mix is about 40% higher (the labor mix is 40% richer), while mean levels of total labor use are about 10% lower than pre-BRA benchmarks. BRA audits use a much richer labor mix without substantially reducing total labor hours relative to pre-BRA benchmarks. In 2002, compared to an auditee with low assessed BRA, an auditee with moderate or higher assessed BRA is about four times more likely to be assigned both a richer labor mix and more total labor than expected under pre-BRA benchmarks.

Auditors use a richer labor mix and assign audit labor in line with their BRA assessments while only modestly reducing total hours. Meanwhile, BRA audit fees (in constant 1989 dollars) are about 75% of pre-BRA benchmarks.