Is there any merit to the theory that “as January goes, so goes the rest of the year” in the stock market? Well, sometimes; but not always. In fact, the trend is supported about six times in 10 — and those odds are not much more favourable than flipping a coin.
Supporters of the January Effect subscribe to the idea that a market rise in January indicates a rise for the full year — and, conversely, a drop signals a lower market a year later. However, the truth is that the other “signal” months of April and November have better records in forecasting year-ahead results using one-month price changes.
From the past three signal months, we get a mixed outlook: stocks had dropped in November 2011, issuing a bearish signal; in April 2011, the signal was bullish.
All this is based on what has happened to the S&P 500 composite index since 1974. But these signal months are also signals in the Canadian market. (The U.S. leads; Canada follows.) The two markets are more closely correlated than any other pair in the world except for Germany’s and Switzerland’s.
That’s why this April either will provide confirmation or non-confirmation of the current bull run. To indicate a continuing rise, the S&P 500 composite index should close in April above 1363.61, the level at which it ended April 2011.
The January Effect is, perhaps, losing its power because so many more investors are aware of it. That happens when a single market indicator, technical or fundamental, becomes widely and closely watched. The fewer people who know about a useful indicator, the more likely it is to work.
There is another difficulty with January Effect signals. For months that produce successful signals, the size of the month’s gain or loss has nothing to do with the year’s gain or loss. A big January gain may precede a small 12-month rise — and vice versa. To take an extreme example, the market had gained only 0.2% in January 1986, but the year’s gain was 29%.
Successful April and November signals have a closer relationship to the outcome in the subsequent 12 months than January, although the margin is narrow. The odds that April signals are accurate are about seven in 10.
Bearish signals in April that lead to a drop in the next 12 months have had a 0.44 correlation between the size of the April drop and the drop over the following 12 months. That is the highest linkage in these signals — and it is not statistically significant.
Still less useful in the past couple of years is the idea that the 10 best-performing industries in the S&P 500 composite index during January will be market leaders for the next 12 months.
For example, only one of the top 10 top-performing industries in January 2010 — automobile manufacturers — was anywhere near being a top performer in the subsequent 12 months, placing 11th in gains over that year.
The January 2011 result was better. One industry in the top 10 industries that month — pipelines — was among the top 10 performers over the next year.
Perhaps January 2012’s top performers will do better. The top 10 industries were: real estate services, diversified metals and mining, multi-sector (financial) holdings, investment banking, life sciences tools and services, construction and farm machinery, aluminum, other diversified financial services, fertilizers and (tied for tenth place) auto manufacturers and Internet retail. IE