Do audit opinions reflect earnings persistence?

Abstract

Purpose

The purpose of this paper is to examine the association between audit opinion and earnings persistence of listed companies in Thailand from 2004 to 2008.

Design/methodology/approach

We use archival data and hand collected data in regression analysis. Content analysis was used to perform decomposition analysis of audit modifications.

Findings

Firms receiving modified opinions have lower earnings persistence than firms receiving unqualified opinions, and the degree of earnings persistence varies among types of modifications. We find that firms with a qualified opinion or a disclaimer have lower earnings persistence than firms receiving an unqualified opinion with an emphasis of matter (UEM). However, we find no difference in earnings persistence between firms receiving a qualification and a disclaimer. Content analysis reveals that there is information in certain types of modified opinions with respect to earnings quality. Firms receiving a scope limitation qualification and a going concern disclaimer have lower earnings persistence than firms receiving an UEM due to going concern issues.

Research limitations/implications

Audit modifications reflect different degrees of problematic issues in clients’ firms, resulting in different impacts on earnings persistence. Thus, policymakers and regulators should emphasize the importance of using auditors’ reports. Strengthened enforcement by regulators makes individual auditors more aware of reputation risk and more likely to express appropriate audit opinions.

Originality/value

We examine a broader set of modified audit opinions than those used in prior research. Our study offers the opportunity to examine the association between earnings persistence and different types of modified opinions, especially a disclaimer, which has been rarely found in prior research.

Citation

Vichitsarawong, T. and Pornupatham, S. (2015), "Do audit opinions reflect earnings persistence?", Managerial Auditing Journal, Vol. 30 No. 3, pp. 244-276. https://doi.org/10.1108/MAJ-12-2013-0973