The purpose of this study is to analyze the relation between auditor-related variables and earnings management. More specifically, it attempts to empirically analyze what effect auditor changes and auditor designation imposed by the Korean SEC have o...
The purpose of this study is to analyze the relation between auditor-related variables and earnings management. More specifically, it attempts to empirically analyze what effect auditor changes and auditor designation imposed by the Korean SEC have on discretionary accruals, proxies of earnings management in this study. Moreover, dichotomized auditor size(Big4 versus NonBig4) is used as an interaction variable to see if the effects on discretionary accruals of auditor changes and auditor designation vary on the auditor size. Auditor-related data and financial variables in this study are collected from Kis Value database, and auditor designation data are obtained from the Korean SEC. Three measures of discretionary accruals are measured in this study; discretionary accruals using modified Jones model, performance(ROA) controlled discretionary accruals, and performance(ROA) matched discretionary accruals. Both bivariate and multi-variate statistical analyses are performed for listed non-financial institutes Korean firms during 2002-2007, of which fiscal year end at December.
The main results of this study can be summarized as follows:
First, t-test and Wilcoxon rank-sum test generally show that firms changing auditors tend to have higher discretionary accruals than those maintaining the same auditor. These results hold for both firms changing auditors voluntarily or those changing auditors mandatorily by the six-year auditor rotation rule effective during 2006-2007. Regression analyses including several control variables also show some evidences that firms voluntarily changing auditors have higher discretionary accruals. These results suggest that firms changing auditors, particularly voluntarily, tend to select more lenient successor auditors who allow the aggressive accounting than predecessor auditors. Auditor size, however, is found to be irrelevant for discretionary accruals; there is insignificant difference in discretionary accruals between firms changing auditors from NonBig4 to Big4, those from Big4(NonBig4) to Big4(NonBig4), and those from Big4 to NonBig4.
Second, to see if earning management by firms is suppressed by auditor designation, this study performs some empirical analyses for both newly designated firms sample and all(newly and continuingly) designated firm sample. Statistical test results produce some evidences that firms whose auditors are newly or continuingly designated by regulatory bodies tend to have significantly lower discretionary accruals than non-designated firms. This result suggests that designated auditors be more conservative than free-engaged auditors. As to auditor size, similarly to auditor changes, there seems no significant difference in discretionary accruals among auditor-designated firms. These results cast a strong doubt on the argument that auditor quality may differ depending on the auditor size, at least in Korea.