Group think and the troubles at Tesco

The latest fiasco at Tesco could prove an embarrassment for more than just the retailer. There appears to have been an over-recording of profit of some £250m, and some are asking questions about the company's auditors.

Of course, the full story has yet to emerge, and Tesco's auditors did flag issues in their most recent report. Further, no-one is suggesting that this is remotely like the scandal at Enron, which led to the effective dissolution of that firm's auditors Arthur Andersen, then one of the five largest audit and accountancy partnerships in the world.

But getting this sort of thing right is very tricky. Financial professionals are only human, after all, and everyone makes mistakes from time to time. Regrettably, the instinct of many people is to call for tighter regulations to “prevent this happening again". It is a litany familiar from the long succession of inquiries into child abuse scandals, and it has hardly been a successful one.

Clamping down on the ability of professionals to make judgements, and trying to cram everything into a tick box, is a recipe for failure.

This whole approach is one of the most damaging legacies of the Gordon Brown era. We also see the problem in China at the moment: the rigorous clampdown on corruption is leading to a virtual paralysis of the government machine, with no-one wanting to risk making a decision of any kind.

One of the best ways of avoiding the sorts of problems which have arisen at Tesco is to try and ensure that there is diversity within management.

Diversity in this context means that opinions which differ from those of the majority are encouraged rather than frowned upon. The phenomenon of groupthink can be very dangerous indeed, as the financial crisis showed. Financial institutions, regulators and accountancy firms all relied, for example, on models of risk which depended upon the assumption that very large changes to asset prices were almost never seen. Even though it had been established beyond doubt scientifically that this was not true, groupthink prevented more realistic assumptions being built into the models.

A key practical question, of course, is how to detect when diversity is disappearing within an organisation.

An intriguing recent paper by David Tuckett, director of UCL's Centre for the Study of Decision Making Uncertainty, uses the power of computer technology to shed some light on the problem. He analyses, or rather gets algorithms to analyse, very large-scale text databases to detect when groupthink is emerging and different perspectives are being squashed. One of them is the internal email database of Enron, and the other is Reuters's' news feeds, containing millions of articles concerning Fannie Mae. In both cases, clear early warning signals could have been identified.

I work with Tuckett on other topics at the Centre, so maybe I see the results through rose-tinted glasses. But innovative ways are needed of trying to avoid the persistent accounting problems which dog our corporate sector. Regulation alone will not work.