We provide results from an experiment designed to assess whether mandatory rotation and/or retention of auditors increases auditors' independence by reducing their willingness to issue reports biased in favor of management. Auditors' reporting is compared across the following four regimes: one that does not require either rotation or retention, a second that requires retention only, a third that requires rotation only, and a fourth that requires both rotation and retention. We find that the rotation requirements in the third and fourth regimes decreased auditor‐subjects' willingness to issue biased reports, relative to the two regimes in which rotation was not imposed. In these other two regimes, however, many manager‐subjects voluntarily retained the same auditor‐subjects over multi‐periods. While the long running interactions between manager‐ and auditor‐subjects resulted in more favorable reports by auditor‐subjects (a lower level of independence), the established relationships also induced manager‐subjects to make higher investments.
In this paper, we examine the relative efficiency of audit production by one of the then Big 6 public accounting firms for a sample of 247 geographically dispersed audits of U.S. companies performed in 1989. To test the relative efficiency of audit production, we use both stochastic frontier estimation (SFE) and data envelopment analysis (DEA). A feature of our research is that we also test whether any apparent inefficiencies in production, identified using SFE and DEA, are correlated with audit pricing. That is, do apparent inefficiencies cause the public accounting firm to reduce its unit price (billing rate) per hour of labor utilized on an engagement? With respect to results, we do not find any evidence of relative (within‐sample) inefficiencies in the use of partner, manager, senior, or staff labor hours using SFE. This suggests that the SFE model may not be sufficiently powerful to detect inefficiencies, even with our reasonably large sample size. However, we do find apparent inefficiencies using the DEA model. Audits range from about 74 percent to 100 percent relative efficiency in production, while the average audit is produced at about an 88 percent efficiency level, relative to the most efficient audits in the sample. Moreover, the inefficiencies identified using DEA are correlated with the firm's realization rate. That is, average billing rates per hour fall as the amount of inefficiency increases. Our results suggest that there are moderate inefficiencies in the production of many of the subject public accounting firm's audits, and that such inefficiencies are economically costly to the firm.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.