The 'January effect' tells us nothing about US stocks
You shouldn't read too much into January's dismal stock-market performance. People always try to find patterns where none exist.
It's been a lousy January for US stocks. The S&P 500 slid by 3.6%. That's a bad omen, say market watchers. They point to the January barometer': the tendency for January's performance to be reflected for the year as a whole.
According to the Stock Trader's Almanac by Jeffrey and Yale Hirsch, January has foreshadowed the whole year's direction 88.9% of the time since 1950. The January effect has been observed in other markets too.
But before you get too nervous, it's important to take data like this with a huge pinch of salt. "It's Neanderthal statistics," as Mark Dow of the Behavioral Macro blog puts it. "You could say rain in Scotland predicts the money supply in the US, but that's just because rain always falls and the money supply grows."
Similarly, stock markets have spent most of the post-war period going up so it's no surprise that January correlates with the annual performance, because most of the time the market is going up.
This is clearly shown by the fact that since 1979, there have been 12 negative Januaries, but only four negative years, says CNBC.com's Alex Rosenberg. Positive Januaries are a much better predictor of the whole year's performance than negative ones precisely because stocks usually rise.
In short, correlation is not causation. People always try to find patterns where none exist, so as to make sense of the information overload. But if there really were a tradeable seasonal pattern, it would have been traded away by everyone jumping on the bandwagon, as several analysts have already noted.