This paper examines the effect of external audit on accounting transparency using privatefirms. We use discretionary accruals as the proxy of accounting transparency. We furtherinvestigate a differential effect of external audit depending on auditor type. Although the effectof external audit has been investigated in many previous papers, most articles has focused onlisted companies or initial external audit for private firms. The previous papers which examine the effect of external audit for private firms mainlyexamine the relation between external audit and interest rates on their debt and the relationbetween external audit and earnings management. The previous papers that examine therelation between external audit and interest rates on their debt find that private companieswith audits pay significantly lower interest rates on their debt than private companies with noaudit. The previous papers which examine the relation between external audit and earningsmanagement has focused on initial audit. The previous papers on initial external audit forprivate firms proposed mixed results. In this paper, we examine the effect of external audit forprivate firms by comparing discretionary accruals between firms with audited by externalauditors and firms with no audit. The Korean environment provides a useful setting in which to examine the effect of an externalaudit. In Korea, the no-audit base case is available for private firms. The private Koreancompanies with total assets of less than 10 billion Korean won are not required to have theirfinancial statements audited by independent auditors. As a result, we observe two distinctgroups of private firms: firms with no audit and firms with audited by auditors. Using this unique setting, we aim to provide systematic evidence on the effect of an external audit per sein earnings management. Using 55,703 firm-year observations (3,348 firm-year observations with no audit and 52,355firm-year observations with audit) from 1999 and 2007 in private firms with audited byexternal auditors and firms with no audit, we test the effect of external audit. We measure twoproxies for audit quality: discretionary accruals measures using Dechow et al. (1995) (modifiedJones’ Model, no-performance-matched discretionary accruals) and performance-matcheddiscretionary accruals in Kothari et al. (2005). The results of this paper are following. First, when we use the two proxies of audit quality,controlling for other factors, discretionary accruals of firms with no audit is higher than that offirms with audited by external auditors. These results are still hold even after appling Neweyand West (1987) test and Clustering test and this suggests that our results are robust. Thisfinding suggests that on average, firms that receive a external audit for private firms experiencelittle earnings managements. This implies that there is the effect of external audit for privatefirms. Second, when we test the effect of auditor type, controlling for other factors, regardlessof auditor type discretionary accruals of firms with no audit is higher than that of firms withaudited by external auditors. This implies that the main factor of audit effect for private firmsis not auditor type but audit per se. Auditor type consists of Big 4 auditor, non-Big 4 auditorand a group of practitioners. A group of practitioners is comprised of more than 3 CPAs for providingaudit services in accordance with the Act on External Audit of Stock Companies. A group ofpractitioners can audit KOSDAQ firms and private firms but can’t audit KOSPI firms. So, when weexamine the effect of external audit for private firms, considering auditor type is interesting. We contribute to the literature in several other ways. Until now, there aren’t researches testingaudit effect on non-listed companies. Previous papers examine the effect of initial externalaudit on the earnings management using private firms. But this paper examines the effect ofexternal audit for private firms. The results of this paper imply that the external audit forprivate firms is important for decreasing earnings management. Academics can also apply thediscussion in this paper for related researches. Korea reduces the scope of external audit byamending Enforcement Decree of the Act on External Audit of Stock Companies in 2009. Theresults of this paper imply that accounting transparency for private firms can be decreasedfrom this amending.