The Audit Review Process—Are Two Heads Better Than One?
By Stephen Asare
Visiting Professor, Boston College

Introduction

The audit review process is an essential element of account- ing firms’ quality control programs and is mandated by authoritative standards. Both the practice and academic worlds, albeit for slightly different reasons, devote significant resources to this process. The old saying that “two heads are better than one” motivates the practice world focus. Specifically, public accounting firms acknowledge the fallibility of human judgments and devote resources to the review process with the expectation that the process can detect and correct judgment and decision errors. In this vein, the review process is a control mechanism geared to enhancing the overall quality of the audit.

On the other hand, and predictably so, a curious academic world focuses on examining the conditions under which the review process improves audit quality. That is, rather than accept the notion that “two head are better than one,” academics seek to “audit” that notion with the expectation of identifying the conditions under which the notion is true, false or only partially true. Notice that because academicians seek to “audit” the review process, they gain valuable insights that can enhance the review process (in practice) even as a well-conducted audit of a client’s financial statements can add value to the client’s activities.

The purpose of this essay is to discuss some of the academic findings in a way that, I hope, will enhance additional dialogue between the practice and the academic worlds. As these worlds are interrelated, the importance of such an ongoing dialogue cannot be overemphasized. The following issues are discussed:

• Does the review process lead to an improvement in the overall quality of an audit (i.e., are two heads better than one)?
• How does familiarity with the preparer affect the review process?
• Does review effectiveness depend on whether the review is done in the presence or absence of the preparer?
• How does the rank of the reviewer affect the review process?

I stress that these issues represent a sampler of the academic literature on the audit review process. They are neither meant to be exhaustive or even representative of the many issues that have been (or are being) studied. The interested reader should see the articles written by Solomon (1987) and updated by Rich et al. (1997) for an excellent and comprehensive treatise.

Are Two Heads Better Than One?
Various academic studies have addressed this question. The typical study asks a reviewer to review simulated workpapers containing known or seeded errors. Review effectiveness is then evaluated with respect to whether the reviewers detect and correct these errors. The preponderance of the evidence from these studies suggest that reviewers detect a significant portion, but not all, of the errors. In a slight variation of these studies, reviewed judgments are compared to individual judgments. For instance, a manager may be asked to review a senior’s workpapers and the manager-reviewed judgments are then compared to an individual senior or manager judgments. Here again, the results suggest that reviewed judgments are more accurate than individual judgments.

Not only do these results suggest that “two heads are indeed better than one,” but also that “two heads are not perfect.” A key practice implication of these dual findings is that there are opportunities for additional risk reduction at successive levels of review. In an attempt to reduce cost, there have been attempts to re-engineer the review process away from the traditional, multiple-layered structure. Such attempts should proceed with caution.

Does Familiarity With the Preparer Affect the Review Process?
The adage “two heads are better than one” is as old as another one, “familiarity breeds contempt.” Accordingly, concerns have been raised as to whether effectiveness gains that, otherwise, will be derived from the review process can be negated when the reviewer is familiar with the preparer. This is a significant concern given that reviewers frequently, but not always, are familiar with the preparer. Notice that similar concerns have been raised in various recurring relationships, including the auditor client relationship, leading some to propose mandatory client rotation and concurring partner reviews. So does familiarity with the preparer impair review effectiveness?

In one study, reviewers were asked to review the workpapers of either a familiar or unfamiliar preparer (from another office) who had completed either a simple or a complex task. To further reflect practice, reviewers conducted their review under moderate time pressure. The workpapers were very comprehensive and contained the procedures performed, supporting evidence and the conclusions reached. Further, the workpapers contained known preparer errors which allowed review effectiveness to be evaluated.

Interestingly, the study found that being familiar with the preparer led to more effective reviews of the complex task. Specifically, the results indicated that reviewers were more confident of the work done by the familiar preparer, in spite of identical workpapers. This confidence allowed reviewers of the familiar preparers to focus more of the review time evaluating the preparers’ conclusions. In contrast, reviewers of unfamiliar preparers were less confident of the work done and spend too much of the review time reperforming the preparers’ audit work. The combined effect of this was that reviewers’ of the familiar preparers detected and corrected more of the conclusion errors. This is certainly good news for public accounting firms as audit team members, assigned to several engagements, become very familiar with each other. Perhaps the only bad news is that when the task was simple, overconfidence in the familiar preparer appeared to have slightly hurt review effectiveness.

Two clear practice implications can be drawn from these findings. Flexible time budgets are needed on occasions when a reviewer is asked to review the work of an unfamiliar preparer. Second, reviewers should guide against complacency when reviewing seemingly simple tasks completed by familiar preparers.

Reviewing in the Presence or Absence of the Preparer
A reviewer may review the workpapers in the presence or absence of the preparer. An interesting question that has attracted the attention of academics is the relative effectiveness of the two approaches. In one study, preparers were asked to generate plausible causes of an unexpected fluctuation in a client’s financial ratios. The preparers’ output was then reviewed by a reviewer in the preparer’s absence or presence. The results indicated that reviewing in the presence of the preparer allowed the “two heads” to communicate more effectively and increased the number of plausible causes identified. A key implication of this finding is that the review process could be enhanced if the reviewer maintains regular contacts and discussions with the preparer as the audit work is being done. In effect, and subject to logistical feasibility, a “real-time” review may be more beneficial than a “batch” review.

Does the Rank of the Reviewer Affect the Review Process?
Current review technology is structured such that seniors perform detailed, mechanical reviews and managers perform general, conceptual reviews. Such a work design is justifiable to the extent that seniors are better at detecting mechanical errors and managers are better at detecting conceptual errors. Again, not taking the justification for this work design for granted, academics have studied whether this division of labor matters. In one study, managers and seniors were asked to review simulated workpapers containing known mechanical and conceptual errors. The results confirmed that seniors are better at detecting mechanical errors while managers are better at detecting conceptual errors. Again, this result suggests that the multiple-layered review structure adds value which must be taken into account when redesigning the review technology.

The foregoing discussion highlights four main findings from the academic world. First, the review process enhances audit quality but is not perfect. Second, familiarity with the preparer does not negate the benefits of the review process. On the contrary, it enhances review effectiveness. Third, reviewing in the presence of the preparer may be more beneficial than reviewing in the absence of the preparer. Finally, the multiple-layered review structure, wherein managers review senior-reviewed judgments, is not redundant. As noted, and reflecting my own bias, this is just a sampler of the issues addressed by the academic world. I hope that these findings and their implications will persuade the audit practice world that the academic world can help solve the puzzle on how to enhance audit efficiency and effectiveness.

References

Asare, S., and McDaniel, L. 1996. The effects of familiarity with the preparer and task complexity on the effectiveness of the audit review process. The Accounting Review 71: 139–160.

Ismail, Z., and Trotman, K. 1995. The impact of the review process in hypothesis generation tasks. Accounting, Organizations and Society 20: 345–357.

Ramsay, R. 1994. Senior/manager differences in audit workpaper review performance. Journal of Accounting Research 32: 127–135.

Rich, J., Solomon, I., and Trotman, K. 1997. Multi-auditor judgment/decision making research: A decade later. Journal of Accounting Literature 16: 86–126.

Solomon, I. 1997. Multi-auditor judgment/decision making research. Journal of Accounting Literature 6: 1–25.

Next Item
Back to Front Page