This great blog entry just came out on auditor term limits:
I have to credit Tom Selling for considering many the issues involved – his analysis seems to be fair, but I’m not convinced. A few specific arguments:
Costs of changing auditors would not be so severe because auditors regularly turn over their staff anyway:
many executives complain that even long-tenured auditors impose significant switching costs on their clients, just because of the year-to-year turnover involving middle and lower staff. It could well be the case that auditor term limits might reduce that kind of turnover from changing assignments within the firm.
Would term limits reduce the number of audit failures? It’s apparently not so simple:
I would say that this is too narrow a specification of the problem.
He gives no answer, but, he replies, maybe term limits would affect executive compensation and prevent a mega-failure like Enron:
we are debating this issue today, and the PCAOB is considering it, because of Lehman, AIG, Worldcom, Enron and a number of others mega-failures that contributed to severe disruptions in the economy as a whole.
Perhaps a “fresh set of eyes” by the new auditor would catch problems that the old auditor missed? This isn’t explained, nor does the writer present any evidence from past auditor changes. However, the old auditor might be more careful when they know that a new auditor will be looking through their working papers:
the beneficial effects of auditor rotation begin with the re-freshening of old eyes – the eyes of the incumbent auditor.
Furthermore, the responder explains that accounting is so complex that financial statements might be useless anyway:
some would say that the much of the problems at Lehman, Citigroup and WAMU is that these firms are not just too big to fail without severe fallout, but they are so complex that a reliable audit may not even be possible. How much of this problem is simply due to the fact that the accounting standards themselves are too darn complicated, counter-intuitive and oversaturated with abstractions that an auditor is simply in no position to evaluate? Answer: lots. And, don’t kid yourself, IFRS is no better.
Term limits would not simplify accounting standards.
Without conclusively explaining why the (unknown) benefits from term limits would exceed the (unknown) costs, the blog goes on to discuss whether Congress would enact term limits anyway:
Let me see if I understand the logic here:
The current audit system doesn’t work, and we don’t know if term limits work, or how expensive they would be. Furthermore, financial statements are really complicated. So we have no choice but to switch to term limits.
Term limits are a nice idea. I have a few questions:
- How much does it cost to change auditors?
- Is the rate of audit failure lower for firms that have mandatory auditor rotation policies than for firms that tenure the same auditor?
- Is the rate of audit failure higher for firms engaging in a first-year audit, versus a repeat engagement, due to lack of expertise?
- How can you require mandatory rotation among only four big auditors?
- Suppose X Firm switches from Auditor ABC to Auditor DEF. And suppose staff, formerly working for Auditor ABC, on the X Firm audit, change jobs to work for Auditor DEF. Would these staff be prohibited from working on the X Firm audit at Auditor DEF? This may seem like an obscure question. However, in many smaller cities, there are only a few large audits. Moving the audits from one firm to another would also force some local staff to move with the clients.