How to Prevent Your Auditor from Telling the World You’re in Deep Doo-Doo
Jonathan Weil, the accounting and corporate finance columnist for Bloomberg News, has written an excellent column (read it!) on auditors failing to provide going concern qualifications in reports when they are clearly warranted. I want to comment briefly on the following points Jonathan made:
- Following the bursting of the dotcom bubble, a Bloomberg study found that auditors failed to provide a going concern qualification for 54% of the 673 largest bankruptcies between 1996 and 2002. It reminds me of earlier days when I was at the SEC during the S&L crisis. Walter Schuetze, the Chief Accountant was particularly irritated by the auditor’s failure to include going concern qualifications in their reports–even when the book value of net assets (largely financial, since he was looking at banks) was substantially above market capitalizations. Evidently, nothing changed.
- Auditing standards require the auditor to add a going concern qualification to its report when there is ‘substantial doubt’ about an entity’s ability to survive. I bring these weasel words, ‘substantial doubt’ to your attention, because of my recent post on the subject of the weasel words in AS 5 and FAS 5. I have no clue what ‘substantial doubt’ means, and the ambiguity of the term may help explain why so many opportunities for going concern qualifications go by the boards.
- The FASB recently voted unanimously to add a project to its agenda that would make a company’s management responsible for assessing the possibility of failure to continue as a going concern.
The Really Interesting Part
Of news to me in Weil’s article was a situation he described when companies could be in technical default on their loans if the auditor were to issue a going concern opinion. It appears to be commonplace, and the story told about American Home is particularly depressing:
“You think your job is tough? Think about the poor schlimazels from Deloitte & Touche LLP who blessed the books at American Home Mortgage Investment Corp., mere months before it went belly up….
Tucked inside American Home’s credit-facility agreement was a clause that said the … company would be in default with lenders if its auditor tagged it with the dreaded going-concern language.
For the accountants, if they thought for even a second about this, it must have felt like staring into a house of mirrors. Had they made what proved to be the right call, they probably would have inflicted a mortal wound on American Home. Then again, looking back, a self-fulfilling prophecy would have spared investors from the company’s April 30 public offering of 4 million shares at $23.75 each, the prospectus for which incorporated Deloitte’s audit opinion. American Home’s shares closed yesterday at 22 cents.
Weil makes it clear that loan covenants triggering technical default can have an undue influence on the auditor, who is already gun-shy about issuing going-concern opinions. The Financial Accounting Standards Board project to require chief executive officers to provide disclosures of going-concern risk has only just begun; and I ask myself whether CEO’s have incentives to do a better job identifying high going concern risk than auditors.
As an interim step, the SEC should require prominent disclosure of contractual terms that would trigger material adverse consequences depending on the type of audit opinion issued. It should also direct the PCAOB to clean up the weasel words that auditors hide behind in avoiding going concern qualifications. (The term “substantial doubt” was actually enshrined by Congress when it added Section 10A to the Securities Exchange Act of 1934, which is another reason that the SEC could provide guidance as to its intended meaning.)
Beyond this, Sarbanes-Oxley has reinforced the principle that it just ain’t kosher for management to unduly influence the auditor; and this principle should be followed to its logical conclusion. That is, this type of arrangement should be prohibited. I am quite sure that the lending community should have no problem finding other technical-default triggers that are just as efficient and reliable, if not more so.