suggest that these results
may be driven by differences in client characteristics. They use matching models to control
client characteristics and find that the audit quality of Big 4 auditors does not differ
significantly from that of non Big 4 auditors. Further, Francis and Wang (2008) report that the
decrease of the magnitude of accruals earnings management is restricted to the clients of Big
4 auditors in the U.S. (in which legal regime is the strongest in the world) and those of
non-Big 4 auditors are not affected by the legal regime changes. As a result, it is possible that
auditors do not play any monitoring for the decrease of REM because the REM is harder for
external auditors, regulators and other stakeholders to detect compared with AEM. One
exception is Kim et al. (2003), who provide empirical evidence that Big auditors are less