Move, January: At least two other months have greater predictive power for stock market returns than you do.
January has a reputation for being able to predict the direction of the US market during the next 11 months of the year. This supposed capacity is known as “January forecast” and “January barometer”.
You will see many references to this indicator in the coming days, now that January is officially in the record books as a “low” month – with the S&P 500 SPX,
falling 1.1%. I have already written that the January Predictor is based on an unstable statistical base. However, the financial headlines will boast about the supposedly negative implications of the January decline for the remainder of 2021.
So, let me point out some other ways in which Predictor is not worth following.
What’s so special about January?
A good place to start is to remember that January 2020 was also a low month (down 0.2%) and yet the subsequent 11 months produced a gain well above the 18.4% average (assuming dividends were reinvested).
This is just a data point. Another clue that there is nothing special about January is that other months have even greater predictive “powers” in predicting the direction of the stock market in the subsequent 11 months. Since the creation of the S&P 500 in 1954, in fact, June has the strongest forecasting capacity, followed by February. January is in third place.
Why, then, don’t you read about a June Predictor or a February Barometer? My guess is that supporters are less motivated by statistical rigor than by stories and narratives that capture their attention. From a behavioral point of view, the calendar year is a more natural period to focus on than the periods from February to February or June to June. But psychological significance is different from statistical significance.
The importance of real-time testing
There is another telltale sign that the January Indicator is not all it seems: it does not pass tests in real time.
By that, I mean tests carried out after it was initially “discovered”. If the January Predictor had passed these tests, we would be much more confident that it is not just the result of a data mining exercise in which historical data is tortured long enough to make a pattern emerge.
But it was not able. As far as I can tell, the January Predictor real-time test starts in 1973. This is the first mention of it on Wall Street, according to an academic study on the subject. Unfortunately, his track record since then is much less impressive. Since 1973, in fact, not only is it not significant at the 95% confidence level that statisticians often use when determining whether a standard is genuine, it is also not significant even at the 85% level.
We should not be surprised; in fact, January Forecast is in good company. Consider a study published last May in the Review of Financial Studies. He examined 452 alleged statistical patterns (or “anomalies”) that previous academic research had discovered. The authors of this recent study were unable to replicate these results in 82% of cases. The remaining 18% turned out to be much weaker than originally reported.
No correlation between the magnitude of the January high and the return in the next 11 months
Another clue that the January Predictor is based on an unstable statistical basis is that there is no correlation between the strength of the market in January and its gain in the subsequent 11 months. If there were such a correlation, we might be able to invent a plausible narrative about investor confidence at the beginning of the year, carrying on for the rest of the year.
But there is no such correlation. Because of this absence, in order to believe in the January Predictor’s effectiveness, you would have to believe that an S&P 500 gain of just 0.01 carries both predictive power and a 13.2% gain. This undermines credulity.
By the way, I chose that 13.2% in my illustration because it is the biggest gain in January for the S&P 500 since its creation in the mid-1950s. That came in 1987. From January 31 of that year until the end of 1987, the S&P 500 lost 9.9%.
To profit from a statistical standard, you must follow it religiously for years
Finally, even if the January Predictor was based on a solid statistical base, you would need to act on it for many consecutive years to try to profit from it in a rational manner. A good rule of thumb in statistics is that you need a sample of at least 30 before the standards become meaningful. In the case of the January Predictor, it means that you would need to follow it for three decades. In addition, during those 30 years, you would not make any other transactions, except switching to a 100% capital allocation every January 31 when the stock market rises in January, and to a 0% allocation if the market in January it is low.
Without patience and discipline, you are doing little to improve your chances above a heads or tails.
The end result? For all intents and purposes, you cannot conclude anything from the stock market decline in January on what it will look like on December 31.
Mark Hulbert is a regular contributor to MarketWatch. Its Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be contacted at [email protected]
More: Get ready for a 10% drop in stocks, driven by the 3 ‘Rs’, warns Bank of America
Read too: This surprising strategy to synchronize the stock market has an academic seal of approval