March 16, 2006

Corp Fin Reverses WKSI Automatic Shelf Registration Position

Last week, I blogged about a position announced by Corp Fin at “SEC Speaks” that they would object to a takedown from a WKSI’s automatic shelf filed between filing of the 10-K and the proxy statement unless a company filed either an amendment to its 10-K or its proxy statement to include Part III information – or unless it included the Part III information in a prospectus supplement to the automatic shel.

From a bunch of kind members, I have heard that the SEC Staff has now withdrawn that position and the Staff’s current position is that they will not object to such a filing or a takedown during that period – rather, the Staff will leave it up to companies to make their own decisions as to whether the registration statement and prospectus satisfy applicable requirements. This position is consistent with the Staff’s traditional position regarding shelf takedowns.

However, the Staff stated that they have not changed their position – as historically stated in Telephone Interpretation H-6 – regarding their unwilliness to declare a non-automatic registration statement effective during the period between filing of the 10-K and the proxy statement unless the current year’s Part III information is included in the filing or incorporated (but not forward incorporated).

A Parody

March Madness begins (here is some tourney history)! Check out this video of an Ipod/Microsoft parody. No, I didn’t get it from Office Pirates, Time Warner’s new site designed to look like it was developed by renegades (just like “independent films” are now produced by the big movie studios).

Should Auditors Be Airport Screeners?

From Gary Zeune – who did this interesting podcast with me a while back – as posted on

“To test terrorism readiness, British authorities conducted Threat Image Projection (TIP) and digitally inserted one of 250 images of guns, knives and other banned objects into luggage. Initially the screeners’ performance was mediocre. But with practice it improved dramatically. Then the images were changed, and the screeners’ performance dropped like a rock to no better than when the program started. But why? There are several reasons.

First, because we ‘see’ only what we expect to see. It’s the same reason you don’t see the bottle of beer on the shelf in your fridge, RIGHT IN FRONT OF YOU, because you’re hunting for, and expect to see, a CAN of beer. So when the images of the grips on the guns or the orientation of the knives were changed, the screeners detection skills were back at the starting gate. For accountants, this means that if they haven’t received significant training in what the ‘red flags’ of fraud look like, they likely won’t detect fraudulent financial statement items.

Second, is ‘distractions’ or ‘noise’. That is, the more items in the bag, ‘noise’ or distractions, the more likely the screeners failed to detect the weapon. One’s ability to pick out a single item declines when it’s part of a complex scene, loaded with similar items. That may be why screeners at the Newark, NJ airport missed a butcher knife in a cluttered handbag. Likewise, financial statement auditors look at hundred or even thousands of transactions, journal entries, and events. So finding the one or few fraudulent items is difficult at best, and near impossible at its worst.

Cognitive scientist J. David Smith at State University of New York, Buffalo, used origami-like items. The 88 participants studied not just the original shape but variations and orientations. They eventually the spotted up to 76% of the targets. Then Smith slightly changed the target and performance fell off a cliff. Why? Because the screeners were looking for the original shapes and orientation, not the variations. Just like screeners have trouble applying their knowledge to new situations, might auditors not recognize a fraud or misstatement because they’ve not seen it, or been trained to look for it, before?

The failure to recognize variations of learned items is specific-token strategy. Screeners recognize the Beretta 9mm but not the 32 caliber revolver or the Beretta packed upright in the bag instead of lying flat.

Just to be sure the nature of the origami items was causing the poor results; Smith had the participants look for actual guns, knives and scissors in over-packed suitcases. Again, their ability to spot the items started low then eventually increased to 90%. But as soon as new banned items were inserted in the luggage, the number of missed items soared 300%.

Statement on Auditing Standards Number 99 says, “The auditor has a responsibility to plan and perform the audit to obtain reasonable assurance about whether the financial statements are free of material misstatement, whether caused by error or fraud.” Thus, auditors have a positive affirmative duty to detect fraud. So, the question is whether auditors will detect methods of cooking the books if they haven’t been trained on those methods? Of course, there are about a bazillion ways to cook the books. In light of the specific-token phenomenon, is there any amount of training that will allow auditors to fulfill their obligation under SAS 99?

In the final analysis the question is, how do auditors train their brains to generalize their fraud detection skills to recognize fraud indicators that they haven’t been trained on? So far, I’m not aware of anyone has figured out how to do that.”