Human nature is ... human

By Michael Coote

Friday 6th December 2002

That old cynic Sigmund Freud used to write of overvaluation of the love object. In his view, love necessarily entailed false estimation of the object's intrinsic merit. Only as an illusion or error of judgment could he explain the phenomenon.

Whatever the truth of his theory, it is common enough for outsiders to wonder what on earth a pair of apparently unprepossessing lovers could possibly see in each other. A lot seems to depend on whether one is inside or outside the charmed circle.

The Economist has come up with some evidence that Freud was on the right track, albeit in an area seemingly distant from love, namely in auditing. Romance and auditing may seem poles apart to those who are not cogniscenti but the psychologies of these twin sublime spheres have similarities in either case. Auditors may be more exciting than the great unwashed believe.

The magazine reports on a paper published in this month's Harvard Business Review entitled Why Good Accountants Do Bad Audits. An experimental psychologist, Max Bazerman, and his co-authors, Don Moore and George Loewenstein, conducted a controlled experiment with what is known as "self-serving bias" on a group of 139 auditor guinea pigs.

The experimental subjects were each provided with five ambiguous auditing vignettes to adjudge. Half the auditors were told to assume they were hired by the company that needed the audit. The other half were required to suppose they were working for a company that did business with the company under audit.

Significant bias arose, confirming he who pays the piper calls the tune. Those auditors who imagined that they were working for the company under audit were on average 30% more likely to confirm across all five vignettes that the figures conformed with generally accepted accounting practices (GAAP) than those auditors who pretended that they were employed by a supplier company. Note this was an experiment and the auditors were not on the payroll of the fictional hirers they were asked to imagine as requiring their professional opinions.

These results were borne out by comparable research. The study's authors said, "These experiments show that even the suggestion of a hypothetical relationship with a client distorts an auditor's judgment. Imagine the degree of distortion that must exist in a longstanding relationship involving millions of dollars in ongoing revenues."

The conclusion is drawn that auditors should be more like tax collectors than partners or advisers.

The study is saddening in light of the fate of the Sarbanes-Oxley Act that was supposed to clean up US auditing standards. Accountants lobbied relentlessly to strangle the act, which was nearly asphyxiated until the WorldCom scandal breathed new life into it again.

One key victory of accountants was to get a proposed requirement that auditors be rotated every five years diluted into the meaningless substitution that only the "lead audit partner" be changed. Mr Bazerman and his co-authors have shown good reason for auditors to get a clean sweep at finite intervals.

Their research also backs up a proposal aired earlier in this column (NBR, Nov 22). It was reported that New York University professor of accounting Joshua Ronen had called for auditors to be externally financed by insurance companies and not the companies being audited.

In the New York Times back in March, Professor Ronen argued that conflict of interest in being paid by the audited company introduced a tendency toward corruption by auditors. To prevent this, he suggested companies pay an insurance premium to cover the cost of their audits. The insurer would then hire the auditors to assess these companies under a risk-averse brief.

The Bazerman study demonstrates that even if not dishonest, auditors try to see things in their paymaster's way. They go soft on the numbers if the audited firm is picking up the tab but go hard if an external party is doing the honours. It's called keeping the boss happy.

The same psychology is likely to have been at work in investment banks that had a conflict of interest over organising corporate finance and IPOs for which their analysts behind a porous Chinese wall were supplying investment recommendations to clients. Even if not a crook, an analyst is more likely to behave like the auditors in the experiment who made believe they worked for the audited company and try to find reasons to like what his firm wants to sell to the punters.

Investors are also given to succumb to the same thinking. If they have fallen in love with the idea of making an investment, they are predisposed to give greater weighting to the rewards over the risks. This is where truly independent advice shows its worth.

An adviser with no stake in the matter above and with sources of income received through scheduled client fees or declared third-party remuneration is better able, on average, to give a balanced picture of the investment under review than even the most well-meaning employee of a company that is supplying and promoting the investment.

As legal review of the investment advisory industry gathers pace, the psychology probed in the Bazerman paper and confirmed by common observation should play an important role in determining how best to proceed. It may not always be dishonesty or negligence that is behind bad investment advice but a simple ingrained human tendency to please the boss and not bite the hand that feeds.

Comments from our readers

No comments yet