In Professor Kahneman and Tversky's 1974 paper, they describe anchoring bias as this:

"In many situations, people make estimates by starting from an initial value that is adjusted to yield the final answer. The initial value, or starting point, may be suggested by the formulation of the problem, or it may be the result of a partial computation. In either case, adjustments are typically insufficient. That is, different starting points yield different estimates, which are biased toward the initial values. We call this phenomenon anchoring."

An experiment was done to prove this theory. There were two groups of students given the following arithmetical expressions respectively and were to give an estimate within 5 seconds.

Group A: 1 x 2 x 3 x 4 x 5 x 6 x 7 x 8

Group B: 8 x 7 x 6 x 5 x 4 x 3 x 2 x 1

Group A made a median estimate of 512, while group B made a median estimate of 2,250. The motivating hypothesis was that students would try to multiply the first few factors of the product, then adjust upward. In both cases the adjustments were insufficient, relative to the true value of 40,320; but the group A's guesses were much more insufficient because they started from a lower anchor.

Similarly, investors always look at the historical price of a stock as the reference point and act on it. Proton used to be the darling of the stock market with prices around RM8 - RM10 in the early 2000. However, due to Asean Free Trade Agreement (AFTA) and other competition, Proton's market share has plunged from 60% to 24% since 2000. When the stock price declined to RM6 in January 2006 many investors thought it was a bargain (as they reference from the high of RM10) and started to accumulate the stocks. Little did these investors know that Proton later fell to below RM2 two years later.

In another example, investors like to anchor on the 52-week high and 52-week low of stocks and make reference from these two numbers. They tend to think that a stock has the potential to get back to its 52 week high which often leads them buying into over-valued stocks.

Investors like to predict stock prices based on their past performance. However, if you are the proponent of efficient market hypothesis where it says stock price follows "Random Walk" theory, there is no way you can predict the future price. Just like the fair coin game, the previous flips have no relation to the subsequent future flips.


Happy investing,

Pauline Yong