Date posted: August 3, 2008

A Bad Case of Anchoring

Last weekend lifeguards had to rescue dozens of people at the beach near where I live. A combination of a strong wind from the south-west and spring tide caused dangerous undercurrents that carried people away or prevented them from reaching the beach again. All this despite warning signs. Many people regard the sea as a big pool and are oblivious of the forces of nature. Spring tide? What’s that?

As one lifeguard explained, some people may spend 15 minutes or longer struggling to get back to the beach, without getting anywhere nearer, instead of letting the current carry them along until it gets weaker, because they want to get out of the sea where they went in. This is a typical case of anchoring, but in a different domain than finance.

In behavioral finance anchoring refers to the tendency to base a decision upon an arbitrary initial reference point. Anchoring is most common in situations with which people have little experience. If a shop assistant tells you that the pashmina shawl you hold in your hands costs 3200 rupees, the best price in Jaipur, but that you can have it for 2500 you may think you’re getting a bargain. When you want to leave the shop to think it over and the shop assistant offers it to you for 2000 rupees you may be inclined to buy it, because 2000 rupees, that’s a real bargain! Unless you know that real pashmina is quite expensive and so is 2000 rupees for a shawl made from viscose…

In their seminal 1974 paper "Judgment Under Uncertainty: Heuristics And Biases", Kahneman and Tversky describe one experiment whereby they spun a wheel containing the numbers 1 though 100. First participants in the study were asked whether the percentage of African countries in the U.N. was either higher or lower than the number on the wheel. They were then asked to give an actual estimate. Oddly, the number on which the wheel landed had an effect on the subjects’ answers. The median estimate was 25% when the wheel landed on 10 and 45% when it landed on 65. Kahneman and Tversky concluded that the random number had an “anchoring” effect on the subjects' responses, drawing their answers closer to the number they had just seen - even though the number bore no relation whatsoever with the question.

So, at the moment some people think the time has come to buy into banking stocks again. After all, they have fallen sharply over the past year, haven’t they? Well, if last year’s price is your anchor, or last year’s P/E ratios.

Dan Ariely gives another example in his highly enjoyable book Predictably Irrational.

Anchoring has become a popular area of research by the way…