Recent Changes in Auditor Going Concern Modification Resolution

Abstract

This study considers the resolution of auditor going concern modifications in the period 1997 to 2010. Hypothesis presented posit that in the period after the passage of the Private Securities Litigation Reform Act (PSLRA) of 1995, auditors provided going concern modifications only in the presence of severe financial distress. In such cases, more going concern opinion modifications would be resolved by bankruptcy than successful resolutions in the base period observed by Nogler (1995). An additional hypothesis posits that after the events surrounding the bankruptcy of Enron and the demise of Arthur Anderson, auditors became significantly more cautious and provided going concern opinion modifications in the presence of less severe financial distress. This would be expected to be evidenced by a greater percentage of successful resolutions rather than bankruptcies than in either the base period or the PSLRA period. Results show that, while statistically the first hypothesis cannot be accepted, the direction of the change in percentages in this period is as proposed. The second hypothesis is accepted, indicating that more firms successfully resolved their going concern modifications that in the base period or the period after the PSLRA. This indicates that auditor decision functions viewed the relative cost of a type 2 error (no going concern modification, firm fails) versus a type 1 error (going concern modification, file continues in existence) increased in the post Enron period. A logistic regression model is presented that support the changing decision model for going concern modification resolution. Further we introduce the use of the Receiver Operating Characteristic Curve to evaluate the optimal cutoff for the logistic model. Based on this analysis, we conclude that auditors lean toward Type 2 Errors over Type 1 Errors.