172 The Financial Times Guide to Business Numeracy The Efficient Market theory Throughout most of the history of the stock markets, most people thought that if only you could properly understand the patterns in stock price movements, you could predict how they would move in the future. People like Charles Dow were taken very seriously in their day. In the late 1950s a young American undergraduate, Eugene Fama, was helping a professor to produce a stock market newsletter. Fama's job was to analyse the historical price data to identify `buy' and `sell' signals based on recurring patterns. After a while Fama saw that although you could find repeating patterns, they did not predict future price movements well at all, and later he published an academic paper, `Random Walks in Stock Market Prices', which suggested that stock price variations were randomly distributed and therefore unpredictable. Fama was not suggesting that most share prices go up and down like yo-yos. Most of the time, the price changes are tiny in the short term. It is these tiny changes, argued Fama, that are random and unpredictable ­ they take a `random walk'. To understand the `random walk', imagine you're standing in a field, and you flip a coin for each step you take. `Heads' you turn right, and `tails' you turn left. After you have taken 25 steps in this way, on average you will be only five steps away from where you started (Figure 7.7). If you were a share price, however, an investor would like to be able to predict the point when you were furthest away from your starting point. Fama said this isn't possible, and proposed that the reason for these price changes were that investors' views are constantly changing in response to the flow of new information about companies and markets. Figure 7.7 A random walk