This study first examines the associations between auditor’s non-audit market share on firm’s decision to purchase non-audit service(NAS) from their incumbent auditor, and then investigates whether auditor’s non-audit market share affect the relation between firm’s auditor-provided NAS purchase and audit quality. If an audit firm has a large non-audit market share, it can be regarded as it has enough experience and skill for (overall or a particular) NAS. On the other hand, this brings concerns that such an auditor might hardly perform its supposed duty of providing audit service. Prior research has examined whether the provision of non-audit service by auditors impair their independence on the basis of economic relation between audit firm and auditee. While previous studies focus on the auditor’s economic dependence or client importance from the joint provision of audit and other services, however, they do not provide consistent results. To test our hypotheses, we use the 8,484 firm-year observations of firms listed on Korean Stock Exchange and Kosdaq for the period 2005 through 2012. We employ the logistic regression to test first hypothesis that whether firms tend to consider incumbent auditor’s nonaudit market share to make decisions about NAS purchase. We then test the second hypothesis by using performance-matched discretionary accruals as a proxy of audit quality in ordinary least squares regressions following prior literature. The results are summarized as follows. First, firms are more likely to purchase NAS from their incumbent auditors if they have large non-audit market share. Second, in this case audit quality measured by performance- matched discretionary accruals is on average negatively associated with non-audit market share of the auditor. In particular, this result is found when the weight of audit service is low in an auditor’s portfolio share. Additionally, these results are robust when (1) we use other measurement for auditor’s nonaudit market share, (2) we consider the impact of total fee per audit hour on audit quality of firm purchased auditor-provided NAS, and (3) we distinguish the impact of purchase timing of NAS from auditor. This study contributes to the literature on firms’ NAS purchase decision and the audit quality when auditors provide non-audit service to the same auditee by examining the effect of the auditors’ non-audit market share. Although prior research provides some evidence on whether auditors are compromised by NAS in various countries, they do not investigate the effect of audit firms’ characteristic of revenue structure. We expect that our results would be helpful for policy makers and regulators in recent moves concerning the prohibition of auditors’ joint provision of audit and NAS. However, there needs to be a special attention to interpret our results, because we do not broadly consider audit firms’ human resources or education programs. More importantly, although specific partners or teams in an audit firm could have different abilities to make quality of particular services, we do not consider these factors due to the limitation of data availability.