Audit Firm Reputation and Client Stock Price Reactions: Evidence from the Enron Experience

Abstract

The financial scandal surrounding the collapse of Enron Corporation caused an erosion in the reputation of its auditor, Andersen. We study the effect of this reputational loss on the stock price of its clients by examining the abnormal returns for a portfolio of 662 Andersen's clients over three windows. For two windows, both following damaging disclosures about Andersen we find, consistent with Chaney and Philipich (2002), that Andersen's clients suffered significant value losses. Further, we find the magnitude of these returns are associated with variables proxying for firms' demand for assurance, insurance and specialized services offered by the auditor. A third window follows the day that Enron suffered its greatest percentage loss in market value, but we find Andersen clients did not suffer value losses during this window. Combined, the results suggest the losses suffered by Andersen's other clients were linked specifically to disclosures relating to Andersen rather than to negative information surrounding Enron. We also investigate spillover effects by examining the abnormal returns of clients of the other Big 5 auditors. We find that, with the exception of Deloitte & Touche, clients of the other Big 5 auditors experienced share price declines during one or both windows following negative disclosures about Andersen. However, the magnitude of these returns seem less related to assurance and insurance factors than those of Andersen clients. Finally, we examine whether some of the losses suffered by Andersen clients were subsequently recovered when the clients changed auditors, by comparing the market reaction to announcements of auditor change by Andersen clients and non Andersen clients over a four month period. If investors view the switch by the Andersen clients as restoring the assurance and insurance value of the audit, we would expect a positive market reaction to these announcements. Our results are consistent with this hypothesis: firms switching away from Andersen experienced positive cumulative abnormal returns during a three-day window following the date of announcement of auditor change. In contrast the abnormal return over the same three-day window for non Andersen switchers was insignificant. In addition, the magnitude of the response is positively associated with the extent of losses suffered by these clients previously during windows of negative disclosures about Andersen.