Monday, January 28th, 2013

The Myth of the January Effect

As we approach the end of January, investors should prepare for a host of articles commenting on the January effect in equities. It’s a reliable column-filler, and the default conclusion tends to reinforce the readers’ prejudices. In a rising year, bulls will look for any evidence of continued momentum. In a bear market, a weak start to the year is yet another reason to stay depressed. The January effect should be a useful indicator – what a shame it doesn’t exist.

Most articles make the elementary mistake of comparing returns for the month of January with returns for the year as a whole. This appears to be perfectly fair, but in reality it’s a completely useless statistic. By the time investors know what the results for January are, it is too late to buy equities in January. The proper comparison is between what happens in January and the next eleven months. In the jargon, this is about the persistence of “out of sample” returns.

The numbers which follow are based on data from 1995 onwards. This is not an exhaustive sample, but it is large enough to draw some basic conclusions. For US equities, the correlation between returns in January and returns for the whole year (i.e. the conventional way of publishing) is 33%. For investment returns, correlation stats of less than 40% (negative or positive) are generally regarded as underwhelming. So even on the conventional basis this is nothing to get excited about. Using the eleven-month data, the correlation falls to just 9% – effectively non-existent.

The same is true of most other major equity markets. For Japan, the equivalent figure is +18%, for the UK it is -1%, and for emerging markets -8%. So the January effect does not exist – or at least it hasn’t over the last 17 years. But there is an important exception: Europe. For Eurozone equities, the correlation of returns between January and the next eleven months is 38%, and for Pan-European equities it is 46%. So if you feel bullish, make sure you buy European equities, because of -not despite – the fact that they have risen nearly 5% YTD.

Leave a Reply

Full Blog Archive:

Other Recent Posts