مثلث شواهدحسابرسی در ارزیابی ریسک تقلب
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|73||2012||13 صفحه PDF||سفارش دهید||10180 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Accounting, Organizations and Society, Volume 37, Issue 1, January 2012, Pages 41–53
Drawing on the triangulation framework of audit evidence (Bell et al., 2005 and Peecher et al., 2007), we experimentally test for the conditions, if any, under which financial-statement auditors alter their fraud-risk assessments based on whether external evidence provides positive or negative news about underlying business performance. We focus on the condition in which two kinds of management-controlled audit evidence – evidence from the financial statements and evidence from internal data depicting performance of a key business process – is contradicted by external evidence suggesting that a key business objective has not been attained. According to the triangulation framework, such contradictory external evidence should heighten auditors’ skepticism about the veracity of management-controlled evidence and increase their assessment of fraud risk. The experimental findings indicate that auditors’ assessments of fraud risk significantly depend on whether or not external evidence disconfirms the attainment of a key business objective, but only when conflicting messages are provided by the two kinds of internal evidence. Importantly, auditors did not rely on external evidence when, in isolation, the two kinds of management-controlled internal evidence both suggested low fraud risk. Auditors’ failure to use external evidence as a means of ensuring the veracity of management-controlled internal evidence is more consistent with a credulous than with a skeptical mindset.
Recent regulation both in the USA (SAS 99; AICPA, 2002; PCAOB, 2007) and internationally (ISA 240) has placed increased responsibility on auditors for the detection of financial statement fraud. The Public Company Accounting Oversight Board (PCAOB) has reminded auditors of the importance of being diligently focused on their responsibilities to detect fraud (PCAOB, 2007 and PCAOB, 2008). However, fraud can be difficult to detect as “some members of management may even seek to conceal outright fraud by strategically altering information they expect the auditor will obtain as evidence” (Bell, Peecher, & Solomon, 2005, p. 19). This changed regulatory environment, as well as society’s demand for greater protection, implies increasing minimum evidence requirements and increased responsibilities for auditors in relation to fraud detection (Peecher, Schwartz, & Solomon, 2007). This focuses interest on how auditors respond to different types of evidence when making fraud related judgments. With the aim of meeting society’s expectations with respect to financial statement fraud, new evidence frameworks have been developed. In Bell et al. (2005) the concept of evidentiary triangulation2 is positioned as a conceptually normative way for auditors to acquire and evaluate complementary sources of evidence and to rely on that evidence in updating their risk assessments. As part of triangulation, the auditor can obtain evidence from the management-controlled financial statement process, management-controlled evidence depicting performance in key business processes (e.g., internal controls, production and marketing reports) or external evidence on whether a key business objective has been attained (e.g., information from customers or other external parties). Use of this external evidence is of particular interest because it is not easily manipulated by management compared with other sources of evidence that are subject to different degrees of management influence (Peecher et al., 2007). Understanding the use of evidentiary triangulation (hereafter referred to as ‘triangulation’) by auditors is important given the view that “triangulation enables audit quality improvement” particularly in situations where the auditor is concerned about intentional misstatement (Bell et al., 2005, p. 29). Specifically, Peecher et al. (2007) note that while the triangulation framework provides considerable promise for improving auditor fraud detection capabilities, there is a need for research that demonstrates more precisely the conditions under which external evidence, providing information about underlying business performance, can better detect material misstatements that stem from management fraud (Peecher et al., 2007). It is this question that we address. In the context of an accounting fraud, we test hypotheses for auditors’ use of external evidence depicting performance of a key business process. Management has implemented an accounting fraud involving overstated client revenue3 (and profitability) using one of two types of strategies to conceal the fraud (‘concealment strategies’). The two concealment strategies produce financial results that have different levels of compatibility with the client’s strategic business objectives and results of operations during the period. In all treatments, the controller provided the same fraudulent explanation for the higher-than-expected revenue number. In order to assess fraud risks at the planning stage of an audit, senior auditors were given the unaudited financial statement numbers (under the two different concealment strategies), business process performance evidence, and external evidence on levels of achieved customer satisfaction for increased sales. Given both consistent and inconsistent fraud risk implications for profiles of the financial statement and internal business process performance evidence, we examine the impact on fraud risk assessments of external evidence on the performance of a key business objective. There are three major contributions of this research. First, in an environment where there is increased emphasis on fraud detection, there is a need to rethink the types of evidence used (Hammersley, 2011, Hoffman and Zimbelman, 2009 and Peecher et al., 2007). Here we examine fraud risk assessments of auditors when they simultaneously use different sources of evidence that are subject to different degrees of management influence. While some forms of evidence can be manipulated by management, other evidence is generally more difficult to manipulate as it comes from outside the organization. Second, as suggested by Peecher et al. (2007), there is a need for research that addresses the conditions under which auditors are more versus less likely to engage in triangulation. Importantly, we find that external evidence, related to key business objectives, impacts fraud assessments when the implications of two types of management-controlled evidence are inconsistent. However, given the ability of management to manipulate this evidence, external evidence related to business objectives should also be useful to detect fraud in situations where the two types of evidence controlled by management both consistently suggest a low likelihood of fraud. This was not the case in our study even though it is this very situation where external evidence should be of most benefit in detecting fraud. Third, we report results on the ability of auditors to use evidence on the performance of the client’s business model to assess the risk of a (seeded) accounting fraud. We manipulate the financial statements such that there is either relatively high or low compatibility of the asserted financial statement numbers with the design and performance of the client’s business model. While a lack of compatibility does not necessarily indicate a misstatement, it should result in auditors refining their misstatement and non-misstatement expectations (Peecher et al., 2007). We find that auditors can use evidence on the performance of the client’s business model, and its compatibility with the financial statements, to interpret appropriately the fraud risk implications.
نتیجه گیری انگلیسی
We developed a situation where an accounting fraud involving overstating of sales revenue was implemented by client management. Senior auditors evaluated potential causes of the higher-than-expected revenue, including the probability of a material seeded fraud, given a fraudulent explanation from management. Following the model of belief-based risk assessments developed by Bell et al. (2005), using two different client concealment strategies, we varied the level of compatibility between the content of the financial statements and the results of the client’s business model. The auditors correctly detected the different levels of fraud risk implied by the two concealment strategies and the resulting levels of compatibility with the business model. Specifically, for the higher compatibility to the business model strategy, the participants assessed the probability of the seeded fraud (p5) to be significantly smaller than for the lower compatibility to the business model strategy. These results are consistent with the Bell et al. normative model. In this model, the SSA auditor needs to acquire an understanding of how well management is executing its business model allowing the auditor to develop expectations that can later be compared with management’s asserted financial statement amounts (Bell et al., 2005 and Peecher et al., 2007). Our results show that participants used information on the client’s business model in making fraud risk assessments. The key issues addressed in our study relate to the use of external evidence, providing information about underlying business performance, on auditors’ fraud risk assessments. Given the potential importance of this EBS evidence for fraud identification (Bell et al., 2005 and Peecher et al., 2007), we examined the extent to which, and under what conditions, EBS evidence is used by auditors. Specifically, we tested for a marginal impact of EBS evidence when the MBR and MII evidence jointly imply inconsistent or consistent risk implications. First, when the MBR and MII evidence was inconsistent, providing conflicting signals as to the likelihood of fraud, unfavorable EBS (higher risk) evidence compared to favorable EBS evidence (lower risk) resulted in a significant increase in the likelihood of fraud. This is consistent with the normative model which suggests that if the other two forms of evidence provide conflicting signals, audit quality will improve by considering additional EBS evidence (Bell et al., 2005). A different situation is when the implications of the MII and MBR evidence are consistent, i.e. both suggest higher or lower fraud risk implications. We found that when they both suggest higher risk, consistent with the normative model, the use of EBS evidence was limited because both previous signals indicate higher fraud risk, even though these signals can be more easily manipulated by management. However, when both the MBR and MII suggest lower risk, the normative model suggests that EBS evidence would be particularly useful because EBS evidence should help the auditor “to spot MBR that are too good to be true” (Bell et al., 2005, p. 29). However, we found that EBS evidence did not impact fraud risk assessments when both MBR and MII suggest lower fraud risk. This finding is incongruous with the Bell et al. (2005, p. 35) model suggestion that “initial evidence from either MII or MBR that seemingly supports assertions should be presumptively doubted until it is corroborated via triangulation”. This result is of concern because this is the circumstance where EBS evidence could be of the most benefit (Bell et al., 2005). More generally, if management has committed a fraud and has been strategic enough to alter the evidence that is generally under their control, audit quality would benefit by making use of external evidence outside management’s control in this circumstance. Auditors failure to use external evidence as a means of ensuring the veracity of management-controlled, internal evidence is more consistent with a credulous than with a skeptical mindset. The question arises as to why EBS evidence was not found to affect fraud probability assessments in the situation where management-controlled evidence suggests lower risks. We suggest this result is most likely due to auditors’ overconfidence in the ability of traditional evidence, such as MBR and MII evidence, to detect fraud and/or the auditors undervaluing EBS type evidence which they have traditionally been less likely to collect. Some recent research has suggested that auditors rely much more on the type of evidence they have traditionally used rather than EBS evidence. For example, Harding and Trotman (2011) find that auditors given certain indications of fraud and asked to select additional pieces of evidence, rarely selected EBS evidence. Thus, while Bell et al. (2005) warn that “obtaining more evidence of a particular kind may not compensate for evidence that otherwise is of poor quality” (p. 26), it appears that auditors still tend to rely on traditional sources of evidence in these circumstances. In fact, Hammersley, Johnstone, and Kadous (2011) find, from in-depth analysis of specific audit program modifications, that while seniors modify the program where appropriate, they modify procedures in ineffective ways, with the bulk of the modifications being “indiscriminate increases in sample sizes”. This suggests the need for practitioners to consider the need for training that illustrates the benefit of triangulation and the use of evidence in situations where other forms of evidence that are under management’s control suggest that fraud is not present. Research that examines the effectiveness of alternative training methods and content in this area would be beneficial. This is an initial study to experimentally examine triangulation. As a result, there are some potential limitations resulting from difficult choices related to the content of independent variables. First, we chose only one type of each of MBR, MII and EBS evidence, each having two levels. There are many other examples of the three types of evidence and they vary in their susceptibility to management manipulation. Future research can further refine the impact of different forms and levels of MBR, MII and EBS evidence including variations in the diagnosticity of each type of evidence. For example, if MBR or MII is so diagnostic of fraud, unfavorable EBS evidence may not impact the fraud risk assessment but may impact confidence. In fact, our study does not measure auditors’ prior beliefs which is a possible explanation for auditors not using EBS evidence when it would have been most helpful. However, regardless of one’s prior beliefs, auditors still need to gather positive evidence to support an opinion of no error. Second, the different sources of evidence have different levels of independence from management. There are other ways of varying the extent of that independence and these could be addressed in future research. For example, discussion by Budescu et al. (2011) is useful in distinguishing between different forms of MII evidence and the likelihood of it being manipulated by managers. Third, the impact of EBS evidence may vary with other factors of the audit environment, including risk, management profile, review processes, etc. The circumstances where EBS evidence is/is not relied upon requires further research. Fourth, our study is about the evaluation of evidence. Future research may focus on the search for and selection of EBS evidence and how that varies with different combinations of MBR and MII evidence. Fifth, while there are benefits of using a rich set of research materials, the complexity results in the need for participants to digest a large and complex pattern of experimental stimuli. This could work against finding evidence in support of our predicted behavior. Finally, our finding that EBS evidence did not impact fraud risk assessments, when the other forms of evidence did not signal fraud, deserves further attention. Assessment of methods for improving professional skepticism in this situation would be useful.