Do Investors Underreact to Auditor Change Announcements?

Financial Reporting Release (FRR) No. 31 (SEC 1988) changed the disclosure requirements about auditor-client frictions (i.e., disagreements and reportable events) in auditor change announcements. The first goal of the study is to document factors associated with an auditor change sample disclosing auditor-client frictions compared with an auditor change sample not making such disclosures in auditor change announcements. We use 5,128 observations of auditor changes obtained from a sample period following enactment of FRR No. 31. In the fiscal periods in which the frictions are reported to have occurred, firms disclosing auditor-client frictions tend to have (1) higher levels of audit risk, (2) deteriorating financial condition, (3) higher levels of growth, (4) higher incidence of new going-concern audit opinion, (5) more auditor resignations, (6) greater incidence of Big 5/6 predecessor auditors, and (6) lower incidence of Big 5/6 successor auditors. We also find that auditor change firms with disclosures of auditor-client frictions tend to be (7) younger firms, (8) exhibit higher audit risk, and interestingly, occur in (9) larger firms with (10) increasing levels of institutional ownership. The second goal of the study is motivated by a growing body of empirical evidence from a variety of contexts that indicates market participants under-react to certain disclosures. Also, the SEC (1988) suggests that both types of auditor-client frictions may be potential early warning signs of trouble. We find that firms disclosing auditor-client frictions exhibit long-run drift in stock prices in the same direction as the initial price reaction. Stock prices indicate that investors under-react to disclosures about auditor-client frictions in auditor change announcements. The evidence of under-reaction to the auditor change sample disclosing auditor-client frictions is robust to alternative asset pricing models and tests suggested by the literature documenting appropriate research design choices for long-run stock price tests.