It's that time of year again, with New Year's resolutions already seeming an impossible target and the cold winters nights setting in with no Christmas break to look forward to... However investors are traditionally very happy this month with December and January traditional outperformers. So I thought I would have a look to see whether I could explain how such a seemingly predictable phenomenon can persist.
Before we have a look at the data I’m going to analyse some of the potential explanations so we know what we’re patterns we’re looking for that could confirm or contradict them. The business year is cyclical with most trading being done in the fourth quarter thanks to the commercialisation of Christmas. However analysts are well aware of this, and their forecasts represent this and their predictions should be as accurate as every other time of the year, so what are the common explanations? A good example is the recent announcements of the quarterly results of various supermarkets with Tesco undershooting their forecast and rival Sainsbury’s over-performing causing their share prices to go through similar negative and positive shocks respectively.
A major reason often cited is that investment/fund managers report their end of year figures December/January time and like to show that they have been holding the big winners of the year. This could potentially lead an uplift in the best performing shares of the year and of the wider market stimulated by the increased buying activity (a similar but opposite and milder than the panic effect in market crash conditions). The knock-on effect I would expect to see if this did happen would be for February to be a poor month. Why? Because once the reporting window is past the fund managers need to redistribute their portfolio, taking positions for the new year and this should unwind any effects on the surge in December.
Another potential explanation, and in my opinion a more likely scenario, is that CEO’s of businesses downplaying expectations before Q4 comes around. It is in management’s interest to make sure that analysts and investors underestimate the value of their stock to help preserve their jobs and increase their potential bonuses when the upside surprise kicks in at the end of year. After all it’s much better to deliver bad news when people are just getting back from their holidays in the summer rather than the time that you are expected to be delivering your headline results. If this theory holds merit I would expect to see underperformance in the months preceding December as executive level management employ techniques to reduce expectations.
So now that we know what we’re looking for, let’s have a look at the data. I’ve selected some of the world's most important indices to look at: DJIA, NASDAQ composite, ASX, FTSE 100, and the NIKKEI 225. I’ve taken the monthly performance data over the last 40 years to produce the graph below:
The graph clearly shows that December, January and surprisingly April have consistently outperformed over the last 40 years.
Revisiting my earlier predictions it is interesting to see that February is a very ‘average’ month performing at 0.57% compared to the average of 0.68% which indicates that investment managers window dressing their portfolio does not hold too much sway. November is a relatively strong month however the preceding months are typically poor performers with September being particularly bad. Which suggest that there is some merit to the lowering expectation theory.
Although the data agrees with the original predictions I made about performance I’m not entirely sure that’s the whole story. Firstly analysts compare like for like quarters, so it would be difficult to bamboozle through off their predictions with poor results in Q2/Q3. Secondly while this effect would make sense if you anticipate a small tenure at your company, it would be next to impossible to pull off year after year.
So the question still remains, why do we see out performance in the months of December, January and April but poor returns in the summer/late autumn? My answer is in two parts and perhaps controversial for economists but I believe it is credible. Firstly, news flow is not random. This may fly in the face of the efficient market hypothesis which is accepted by many, however I believe companies’ management are much more likely to release bad news around the summer as there is weaker trading for most companies in the summer (obvious exceptions such as airlines aside). If such a cyclical company is struggling to stay afloat, it is much more likely to enter administration when cashflow is tight in the summer months.
Positive news flow is more likely in the winter. Management want to take credit for their good actions and results when the markets are paying full attention and not on their summer holidays! More exciting announcements/forecasts are reserved for year-end (both calendar and tax year).
Finally, significant negative events have a habit of occurring in the summer. Looking down the listof stock market crashes over the time period I examined these events have coincided with. Using the data to estimate this; I have calculated that this accounts for between 20 - 25% of the overall deviation so while not the full story it is definitely a contributing factor.
