In a great book, well worth reading “Economyths”, the author, David Orrell, writes on Daniel Kahneman and Amos Tversky.

“One psychological quirk of human beings is that we like to find specific rational explanations for things. In the early 1970s, one of the most commonly cited causes for stagflation was the failure of the Peruvian anchovy fishery in 1972, which was due in large part to a severe El Nino event. Anchovies were a major source of livestock feed, so the effect was to push up food prices. Another contributor was the success of the Organisation of Petroleum Exporting Countries (OPEC) at constraining oil supplies, which also boosted prices. To the monetarists, the cause of stagflation was the inept government printing too much money.

While the exact causes are debated, though, one thing was for sure: the effect of stagflation was to make people unhappy. There's nothing like rising prices, and a risk of losing your job, to upset the electorate - which is why politicians keep a close eye on the misery index. They know that money is an emotional business. It may also turn out that the cause of inflation has less to do with things like oil, anchovies, or money supply, than with basic human emotions.

In 1971, the Israeli psychologists Daniel Kahneman and Amos Tversky published a paper that explored the difference between intuition, which they called System 1 thinking, and reasoning, or System 2. System 1, according to Kahneman, is 'fast, effortless, associative, and often emotionally charged'. It is also governed by habit, which makes it hard to change or control. System 2, in contrast, is 'conscious, it's deliberate; it's slower, serial, effortful, and deliberately controlled, but it can follow rules'.

The paper, entitled 'Belief in the Law of Small Numbers', presented empirical results showing that their experimental subjects, when acting in System 1 mode, could not make accurate estimates of probability. They made extremely basic mistakes, and appeared to have no grasp of the rules of chance. This might not be surprising, except that the people they were studying were experienced statisticians.

The Law of Small Numbers in the title was a reference to the Law of Large Numbers. This is the name of a theorem, stated without proof by Girolamo Cardano and finally proved by the mathematician Jacob Bernoulli in 1713, which says that the accuracy of a statistical sample improves with the number of samples - or as Quetelet said: 'The number of individuals observed.' For example, the quality of an opinion poll will be much better if it is done for a thousand people, than if it is done for ten people.

Statisticians know this very well - or at least the System 2 side of their brain knows it very well. But Kahneman and Tversky found that, in practice, they didn't need large numbers of samples to jump to a conclusion - instead they 'view a sample randomly drawn from a population as highly representative, that is, similar to the population in all essential characteristics', System 1 was leaping in with the answer before System 2 had even fired up its pocket calculator. As a result, people could be easily fooled.

During their long collaboration, Kahneman and Tversky found a number of results that directly question the neoclassical assumption that we make decisions rationally. If there is such a thing as the average man (or woman), then it turns out he has some distinct psychological quirks. For example, he has an asymmetric attitude towards loss and gain - he is roughly twice as sensitive to losses - so tends to avoid taking risks. He is biased by recent events, so if the market has been going up recently, then he expects that trend to continue. He dislikes change, prefers to keep what he has than trade it for something similar, and hates to give up a long-held belief. He underestimates the likelihood of extreme events, and overestimates his own ability to deal with them.

One might think that a group of people would make better decisions, and in some ways they do. But as Kahneman explains, 'when everybody in a group is susceptible to similar biases, groups are inferior to individuals, because groups tend to be more extreme... In many situations you have a risk-taking phenomenon called the risky shift. That is, groups tend to take on more risk than individuals.’ Groups also tend to be more optimistic, suppress doubts, and exhibit group-think. In larger informal groups such as markets, this can translate into herd behaviour, in which investors all rush into the market, or out of it, at the same time.

The work of Kahneman and Tversky helped create the field of behavioural finance. The area has recently been supplemented by the even newer field of neuroeconomics, which uses techniques like brain scans to find out how our brains handle economic decisions. For example, scans have shown that the offer of a reward affects different parts of the brain depending on whether the reward is immediate or delayed. The former triggers a stronger response, which may explain why many people don't set enough aside for retirement. In fact, studies of patients who for neurological reasons are unable to process emotional information show that it is extremely hard to make decisions without some emotional input. If we really did have infinite computational capacity but no emotion, as the neo-classical model demands, we would be incapable of buying a pair of socks.”