€4.9 billion? They wouldn't notice

"French bank loses money” is a headline that inspires Schadenfreude in many Englishmen. But Société Générale's €4.9 billion (£3.7 billion) loss to fraud by a single trader give us more than a good laugh. There are some lessons to learn from it.

One is that trading in financial assets is not rocket science. Emotions matter. SocGen's rogue trader, Jérôme Kerviel, was not a criminal mastermind. What probably happened is that he first lost a small sum. Then, rather than own up and face a few embarrassing meetings and loss of a bonus, he tried the simple strategy known to every member of Gamblers Anonymous: double or quits. And when he lost on that, he hid his losses while trying to recoup them. But the losses just grew.

This is just an extreme manifestation of some errors that investors and traders commonly make. One is to fail to learn from experience. Often, if a share falls after we've bought it we infer not that we were wrong, but that we were merely unlucky. So we hold on to it in the expectation that we'll be proved right. We fail to heed W.C. Fields's advice: “If at first you don't succeed, give up. There's no point being a damn fool about it.”

Also, our ego intrudes. We hate facing up to losses not just because they make us poorer but because they force us to acknowledge that we weren't as clever as we thought. SocGen's troubles show how far men will go to avoid facing this truth.

These errors contribute to what Meir Statman, of Santa Clara University, California, called the “disposition effect”: investors are disposed to hold on to falling stocks in the hope they'll turn around and get even.

These biases can affect share prices. Because some investors fail to sell falling stocks, prices of them don't immediately fall as far as they should. And, for similar reasons, prices of rising shares don't immediately rise as far as they should. The result is that momentum investing - buying past winners and selling past losers - can produce good profits.

A second lesson of SocGen's fraud is that it's amazing what doesn't trouble stock markets. On the day SocGen announced its loss, and an intention to raise €5.5 billion from the stock market, France's CAC-40 index leapt almost 4 per cent while SocGen's price fell less than 5 per cent - buttons in these markets.

Why is the market so relaxed? One reason is that there is lots of money sloshing around the global economy, much of it in the hands of Chinese and Arab sovereign wealth funds. These were quick to plug the holes in the balance sheets of Merrill Lynch and Citigroup. Markets are hoping they will do the same for SocGen.

But perhaps stock markets shouldn't be relaxed. The third lesson of SocGen is that top bosses cannot know everything that goes on in their organisations. The division in which SocGen suffered its fraud - equity futures hedging - is, by the standards of modern banking, a simple business. But it still had enough dark holes for a trader to hide huge losses. When you consider the countless other businesses that banks have - many of which make equity futures look like the Teletubbies - how many other ways are there for individuals to hide losses?

This is one of the threats still hanging over stock markets. It's quite possible that even now banks haven't yet announced the full extent of the losses they have made from holding US mortgage-backed assets. This isn't because rogue traders may be fraudulently hiding losses. It's because honest traders have lots of ways of pricing complicated illiquid assets and can fudge on the optimistic side. Unless a boss is more expert than his traders on multivariate copulas - the mathematical methods used to price such assets - he'll not see through their fudges.

And the boss won't be more expert. Why buy a dog and bark yourself? The division of labour that makes companies efficient - in so far as they are - is also a division of knowledge. It's just impossible for a bank boss to continually know more than every employee does. Ignorance, therefore, isn't a failing of a particular individual but an ineliminable fact about any organisation.

This highlights a curious paradox about modern organisations. Their hierarchical structure is much the same as that of the first factories of the industrial revolution. But one of the conditions that made hierarchy work back then is no longer present. That condition is that bosses know more than workers. In the first factories, bosses knew everything about production processes - men such as Arkwright and Watt had invented them - while workers knew little; they were illiterates and children. It was therefore sensible for information to flow up the hierarchy and orders to flow down.

But in today's firms, knowledge is spread throughout the business. Bosses aren't, and can't be, the know-alls that early factory owners were. Yet organisations are structured as if they are.

Perhaps we exaggerate the extent of management expertise. What looks like bosses' competence is in fact the skill and goodwill of their employees. Without this, they are like one-legged ducks. They might look calm and assured on the surface, but underneath they are paddling frantically without much idea where they are heading.

Luckily, financial markets give us a simple solution. Thanks to the easy availability of funds that track the stock market, ordinary investors don't need to worry about particular companies. We can back the field, rather than particular horses.

But voters don't have such an easy answer. SocGen shows us that big organisations can't always be run effectively, even when managed by highly intelligent and diligent people. So why should politicians pretend the opposite can be true of the State?

Chris Dillow, a columnist for Investors Chronicle.