The markets are full of chatter about the ‘January effect’ or the ‘January barometer’: the idea that January’s stock-market performance forecasts the overall year’s. Pay no attention.
Human nature tends to want to find patterns to explain the world, especially if there’s a chance to cash in. As Fidelity’s Tom Stevenson points out in The Sunday Telegraph, “experiments [show] that the neurological impact of spotting patterns or sets of circumstances that in the past have made us money is exactly the same as that triggered by a hit of cocaine”.
And it doesn’t take much to make us think we’ve spotted a trend. Consider a coin flipped six times in a row, says Stevenson. The odds of getting the tails and heads sequence HTTHTT are exactly the same as those of producing HHHHHH, yet the “streak of heads makes you feel quite differently about what is going on”.
The S&P 500 index, for instance, has moved in the same direction in the full year as it did in January on 17 of the last 25 occasions. That’s a bit better than a coin toss, but hardly impressive when stocks tend to go up most of the time anyway.
It’s a similar story with other famous stockmarket barometers, such as the Superbowl (American football) indicator, Republican or Democratic presidents, and the hemline or lipstick indicators.
Correlation does not necessarily mean causation – and the spurious correlation often isn’t especially impressive anyway, as in the example above. It’s the economic backdrop, not the patterns we insist on seeing, that really matters.