Summing Up the January Effect, Looking to February and Beyond

January was unusual not just due to the market's losses, but more so by the magnitude that underperforming stocks from December outperformed. Typically in January, the most beaten up stocks from December, whether due to tax loss selling or institutional window-dressing selling, outperform the best performing stocks from December, by about 3.4% over the last 35 years. Not every January, but often enough and consistently enough to be expected. This is the essence of the "January effect", a rebound from the artificial selling pressure of yearend.

This year, the rebound effect was much larger, roughly 8.5%. That is, the weakest stocks from December, various housing-related, financial, restaurant, airline and specialty retailing stocks, among others, crushed their equity peers last month. As I said, some rebound effect is to be expected, but could this magnitude been forecasted? and exploited?

Well, yes and no. I looked at the last 35 years, and how the relative performance of the biggest past month's losers did relative to the biggest past month winners, depending on how the market did in December. If the market was up in December, the Loser/Winner spread in January averaged 2.7%, but if December was a down month, like in 2007, the spread soared to an average 6.3%! Granted there's only been 9 negative Decembers during that time, so the statistic isn't that robust, though statistically significant, yet this year's result falls inline with the past.

Of course, to truly exploit this effect means basically selling all your attractive stocks and buying all the down-beaten junk, then at month end flip back out of them and back into the good stuff. This much trading is expensive and definite, whereas the rebound effect has some element of uncertainty. These costs double with a long/short strategy. So, despite the "gross" appeal of such a tactic, the "net" effect is to sit tight.

Given January's outcome, what to expect for February and beyond? First, as might be expected, the Loser/Winner spread tends to reverse, so the January's winners tend to revert and become losers again, and vice versa. However, beyond that it is too tough to tell. No clear trend emerges or more importantly, there is not enough data over the last 35 years to make a reliable guess. So, best thing to do is not change plans, and stick to a strategy that buys stocks based on the combination cheaper relative valuation, with good growth, rising expectations and longer term relative strength (among other traits) and avoid the opposite. Disregard January's performance and stick to your knitting.