Wednesday 12:50

This reorganization process I am trying to do to the current chapter I am working on is turning out to be very difficult. It is hard because of the amount of material (~70 pages) and because my original organization was pretty bad. I think it was bad all along, but it is particularly bad in the emerging logic of everything that is preceeding it.

So I cut the entire chapter and dumped it into a different document. I am now going through and trying to move it back in rearranged bits and pieces. And really, I am taking one paragraph from one section, the next from another. It gets very confusing very quickly.

And then I am making the argument kind of in reverse order. Before, I started with the puzzle of the expectation gap. (For those who are interested: Auditors believe they are not responsible for finding fraud, despite demand for them to do so and its codification in law and professional standards. Their argument is that since they cannot be expected to find every single tiny fraud, they shouldn't be asked to try to find anything. Just imagine if airplane engineers took the same attitude towards, say, making sure airplanes don't fall out of the sky. They call this blatant shirking of their statutory responsibilities the "expectation gap", meaning that silly non-auditors are naive and have unrealistic and unfair expectations of what auditors are responsible for...But I digress..) I then showed that the strange competitive environment they live in explains the bizarre argument.

Now, I start with the bizarre behavior of auditors caught in the major fraud scandals, which does not conform with the behavior predicted in the model of earlier chapters. I ask, is my model wrong? Or is there an extension of the model that will explain the unexpected behavior, but one that is fundamentally consistent with the dynamics I describe? Of course, lucky me, the answer is the second. Evidence for this is:

1. Audit firm incentives do not reward good auditors, so the problem is not just rogue auditors.

2. The model does not look at which auditor is selected. There might be something interesting going on there that changes auditor behavior.

3. Lo and behold, competition for engagements changes auditor behavior, since management gets to hire the auditor (effectively if not officially).

Around here is where I am currently bogged down...

4. This competition process has developed lots of subtle ways of financing those bribes I was talking about. Only this is an opportunity to bribe the auditor not to audit from the start, rather than collude after the fact. Hence the lame auditors in all of the scandals...

5. Then there is that pesky penalty for audit failures. How to reduce that, so that auditors can compete more completely for those bribes? Well the penalty is a conglomeration of legal penalties, loss of reputation, and personal and professional shame. Loss of reputation hasn't seemed to matter much, and the industry stopped going to trial to keep some of their more egregious behaviors from coming to light. Auditors have been pretty effective in reducing legal penalties through political lobbying, which is probably a cheaper way of reducing overall risk than incremental improvements in audit intensity. Then, the profession invented the expectation gap to help with the shame aspect: if it wasn't your responsibility to find the fraud, why should you particularly care if you didn't find it? It also helps with the reputation and legal penalties...