The role of audit evidence in a strategic audit
Introduction
The purpose of this study is to examine the conditions under which audit risk (the probability of audit failure) increases with information in a setting similar to Newman and Noel’s (1989) game. In the literature on strategic auditing, it has been reported that audit risk may increase when the auditor obtains more information. Patterson (1993) extends Newman and Noel’s (1989) game to an audit model in which an auditee decides whether to commit fraud and an auditor chooses sample size, obtains audit evidence, and decides whether to accept the manager’s assertion. She discovers an intriguing phenomenon whereby audit risk increases with sample size and reports a numerical example for such cases. It is difficult, however, to obtain explicit conditions for the cases, owing to the complexity of the model.
The current study analyzes the effects of audit evidence introduced to the benchmark game without audit evidence. In the benchmark game, the auditor utilizes only the information embedded in the payoff structure. The analysis focuses on how audit risk and its components change when audit evidence is introduced.
The current study finds that inherent risk and audit risk increase with audit evidence if the auditee has a sufficiently strong incentive for committing fraud. Detection risk always increases with audit evidence. If the auditor has a sufficiently strong incentive for avoiding false rejection, audit risk increases when the auditor obtains audit evidence.
Section snippets
The game
There are two risk neutral players in the game: a manager and an auditor. First, the manager decides whether to commit fraud. The manager’s action space is binary, so that his strategy is fully described by the probability of committing fraud, . Second, the auditor observes audit evidence . If the manager commits fraud, audit evidence x follows a distribution with a probability density function and a cumulative distribution function for the range . If the manager does
The equilibrium
The purpose of this study is to examine the role of audit evidence in a standard acceptance sampling model. This task is conducted by comparing some equilibrium values in the game described above with the corresponding values in the benchmark game without audit evidence.
In order to establish the benchmark results, the auditor is assumed to use only the information embedded in the payoff structure. In this case, the auditor’s strategy is fully specified by the probability () that she
Audit risk model
This section compares audit risk and its components in two games. Audit risk is defined as the probability that the auditor unknowingly fails to detect material misstatements in financial reporting. Generally Accepted Auditing Standards typically decompose audit risk into three componentswhere denote audit risk, inherent risk, control risk, and detection risk, respectively. Inherent risk is the probability that the manager’s assertion is subject to material
The auditor’s information collection decision
In the previous section, strategic aspects of the auditor’s information collection are abstracted from the analysis.2 One way to analyze the auditor’s information collection decision is to examine whether audit risk may increase when the auditor decides whether to observe audit evidence at the beginning of the game. The manager is assumed to know the auditor’s information collection decision when the
Concluding remarks
This study compares a simple acceptance sampling game with a simultaneous game of not having audit evidence. The analysis identifies the tendencies in exogenous parameters toward increases in audit risk and its components. When audit evidence is introduced, inherent risk and audit risk increase if the manager has a sufficiently strong incentive for committing fraud. The introduction of audit evidence always increases detection risk. Further, if the auditor has a sufficiently strong incentive to
Acknowledgements
I thank Lawrence Gordon and Martin Loeb (the editors), and two anonymous referees for many helpful suggestions. I also thank Atsushi Shiiba and the seminar participants of the 2005 CCRG Conference at Irvine, the 2006 JAA Annual Meeting at Tokyo, the 2007 Asian Pacific Conference on International Accounting Issues at Kuala Lumpur, and the 2008 AAAA Conference at Dubai for helpful comments and discussions. I gratefully acknowledge financial support from the Takahashi Industrial and Economic
References (5)
- et al.
Sampling information in strategic audit settings
Auditing: A Journal of Practice & Theory
(1989) Testing Statistical Hypotheses
(1986)