Peeling away financial reporting issues one layer at a time

A Perspective on “Professional Skepticism” — Part 2 of 2

(Part 1 is here.)

I have found that a useful way to home in on the drivers of audit quality is to regard an audit as two distinct engagements: (1) verifying facts, and (2) assessing whether management’s estimates “appear” reasonable. For one thing, it is near certain that the inexorable change in ‘product mix’ over the decades since the enactment of the federal securities laws is why the profession forfeited the privilege to self-regulate.  For another, the PCAOB’s continual emphasis on PS is their way of letting auditors know that even though their relationship with a client is long, their memories should be short.  Yet one wonders if guidance from the PCAOB on PS can have much of an effect on audit quality when:

  • Management’s knowledge and capabilities exceed those of auditors;
  • Their intent to take future action often has to be a consideration when applying an accounting standard; and
  • Their estimates are almost always biased to favor their personal interest in the financial statements.

These realities make it difficult to understand — even for the most straightforward of audits —  why investors would place much trust in the auditor’s three-paragraph, boilerplate pass/fail report.   The circumstances in which that occurs just doesn’t pass the sniff test.  To drive that point home to my audience of veteran auditors, I ventured an analogy:

Accounting Professor X permits students to grade their own exams. In determining one’s grade, students may even take into account the intention to learn the material better during the coming months while studying for the CPA examination. Professor X understands that she must rein in unreasonably high grades, but that’s not as easy as it sounds. All of the students are giving themselves the ‘benefit of the doubt,’ so to speak.

Professor X’s compensation and tenure is determined in part by students’ ratings.  S/he certainly can’t, and does not wish, to confront every student and remove the bias from every grade.

Future employers (i.e., investors) of Professor X’s students – who will rely on grades to identify her best students – are clearly not well-served by the rules of engagement for the class. What if the entire university system permitted students to grade their own exams?

To some, the analogy to education might not suit, but the conference planners were eager for me to pose it, and my audience responded very empathetically.  The resulting grade inflation* and the rubber-stamp audit report are not very dissimilar.  When evaluators (auditors or professors) are not truly independent of those being evaluated, the evaluators will have perverse economic incentives that will produce certain dysfunction.  In auditing, a concept such as PS is necessary to offset structural incentives to implicitly collaborate with one’s client — yet, one is left to wonder whether it can be very effective, or whether there is a better way.

Moving on from the example of Professor X, I assumed that my audience was familiar with the specific pronouncements of the PCAOB pertaining to PS.  I didn’t want to take up their time with a mechanical recitation, and tried to get them to think a little bit outside of that box.  To set that up,  I stated that a checklist strategy could be adopted to comply with those specifics, but that might not always be effective, or what the PCAOB actually hopes to see when conducting its inspections of firms.  The remainder of my presentation were my thoughts as to how auditors could address the letter of PCAOB guidance without resorting to burdensome checklists, and in doing so, more effectively comply with the spirit of the rules.

The first idea I proposed was that PS and independence are similar concepts.  Both ultimately boil down to one’s (unobservable) state of mind.  Consequently, guidance has been produced that provides for at least the appearance of independence.  Similar guidance does not exist for PS.  So, one question leadership might ask itself is what they might do to convince PCAOB inspectors that a real culture of PS exists at their firm?  These were some of the ideas I brainstormed with the group:

  • An audit program is highly contingent.  The actual audit should rarely, if ever, proceed exactly according to plan.  Accordingly, the original plan/budget should contain slack to handle unforeseen developments.  Documentation of all midstream course corrections should be accumulated in one place within the work papers for easy reference.
  • The audit tests that are the riskiest should be performed earliest.  Doing so could mitigate the “halo” effect and similar psychological biases that are inherent in personal relationships.  It also increases the likelihood that costs of changes to these tests fit within original cost estimates.
  • PS should be made explicit in staff evaluations, and a culture of tolerating PS, even in the face of cost/time pressures, should be continually emphasized and positively reinforced.
  • The less the audit team collectively knows about the best way to make a particular estimate, the less value is added by PS; hence, the more it should plan on employing an independent specialist.
  • Directly test for management bias.  To do this, select a sufficient number (I recommend at least 10) of management’s estimates, and compare them to the auditor’s independent estimates.  For example, given 10 independent tests of management estimates, the probability of more than 8 less conservative estimates by management occurring at random is only 5%.

* * * * * *

Finally, I revealed to my audience of auditors my core belief that the entire system of historic-cost based accounting and audits of management’s estimates does not adequately serve the needs of modern investors.  Accounting and auditing have not kept pace with economic change.  New sources of management bias just add poison to the well.

We could throw the whole thing out and start all over, but that is unlikely to be politically acceptable.  It is also possible to advocate for incremental improvements so as to gradually wean financial statements from its dependence on management’s estimates.

I suggest that we start by reforming the production and auditing of Level II and III fair value estimates.  What if management’s estimates of these fair value were replaced with independent appraisals?  Management bias would be eliminated, and the auditor would do what it does best: verify that the factual information provided by management to the appraiser is accurate.  The auditor could also be responsible for: (1) verifying that the appraiser satisfies specific independence standards (established by the SEC); (2) that the appraiser performed it’s work in accordance with GAAP, and in accordance with their engagement letter; and(3)  that the appraiser’s calaculations are free from material error.

* * * * * *

Overall, I think my presentation was well received by auditors who were genuinely interested in doing quality work that served the public interest.


* Retired accounting professor Jagdish Gangolly has pointed out on the AECM listserv that academic institutions in other countries recognize more effectively deal with the problems of independence and grade inflation:  the professor designs the exam, but independent graders score them on a no-name basis.

No Comments

Leave a Comment