Is Your Auditor Improving?

An academic paper claims the PCAOB inspections have helped to improve accounting firms' quality, but the public has barely noticed.

Sarah Johnson
CFO.com | US

October 12, 2007

As the Public Company Accounting Oversight Board approaches its four-year anniversary, has the auditor watchdog been effective in improving audit quality? The answer is yes, according to a working paper from two accounting academics who say the PCAOB plays a "remedial role" for auditors.

However, a firms' clients wouldn't be able to tell from the PCAOB's inspection reports that their auditors' work has improved, they say. To put together their paper, Clive Lennox and Jeffrey Pittman, an associate professor and visiting scholar, respectively, at the Hong Kong University of Science and Technology, examined 483 of the PCAOB inspection reports issued during a two-year period.

Lennox and Pittman checked this data against changes in auditors' market shares over time and audit firms' peer reviews. They concluded that negative reports do not cut into firms' market shares — companies do not base their auditor decisions on the results of the PCAOB inspections. Rather, either they trust that the auditor will improve in quality or they simply don't find the reports themselves informative.

In fact, the paper concludes that the reports aren't valued by the firms' clients. "It may be quite difficult for users to extract meaningful information about audit firm quality from the PCAOB's reports," according to the paper. "The fundamental problem is that the PCAOB discloses the deficiencies on each engagement sampled by the inspectors and fails to provide a balanced evaluation of the firm's overall level of quality."

Indeed, critics of the inspection reports coming out of the fledgling organization echo Lennox's and Pittman's concerns. Stymied by the Sarbanes-Oxley Act's requirement to simultaneously make the reports public and keep portions of them confidential, the PCAOB is limited as to what its evaluations can reveal: they are essentially a list of faults the board's inspectors have found with each firm.

The inspectors ping the firms for weaknesses and deficiencies — mostly having to do with lack of proof that certain tests were done — but do not include any findings about the firms' quality-control systems (unless the firm fails to remedy the problem within 12 months). Some of the faults found are major; for instance, 7.1 percent of inspections Lennox reviewed had audit deficiencies that led to a restatement of financial statements.

The lack of information in the reports strips them of context and perspective for investors and the auditors' clients, PCAOB observers told CFO.com earlier this year. "I think the process is well intended, and it is helpful and constructive, but right now it is not producing the kind of results that it should for people who are using the results and trying to understand what this means," says J. Michael Cook, a former CEO of Deloitte & Touche.

Some of the firms have also chimed in that they disagree with the PCAOB's presentation of its reports. In its response letters to the reports, BDO Seidman wrote: "The format of the report...does not lend itself to a portrayal of the overall high quality of our audit practice." BDO's letters also noted that the engagements the PCAOB reviewed involved hundreds of decisions open to many interpretations — which touches upon another criticism of the reports: for the larger firms, the PCAOB will not reveal how many audits it looked over for each firm. Because its auditors don't pick the company audit clients randomly, but rather look at the highest-risk clients, the PCAOB discourages readers of the report from drawing conclusions based on the number of audit deficiencies and issuers listed.

Grant Thornton has also noted its distaste with the reports. This spring, the second-tier auditing firm took an unusually sharp tone in its response letter to the PCAOB's review of its 2005 audits. "We strongly disagree with the use of overly broad comments such as 'failed to identify' and 'failed to perform,'" wrote Grant Thornton. "They do not adequately describe all of the relevant facts, nor do they acknowledge the extent of the procedures that were, in fact, performed by the engagement teams."