My table shows monthly returns in the last 10 years for three of my no-thought portfolios: value (the 20 highest-yielding stocks); momentum (the 20 biggest risers in the previous 12 months) and defensives (the 20 lowest-beta stocks). All have a seasonal pattern.
| Monthly returns on stock strategies | ||||
|---|---|---|---|---|
| Defensives | Momentum | Value | FTSE 350 | |
| Jan | -0.4 | 1.3 | 0.8 | -0.1 |
| Feb | 1.3 | 1.7 | 2.5 | 0.7 |
| Mar | 0.8 | 2.3 | 0.3 | 0.4 |
| April | 0.9 | 4.2 | 4.7 | 2.5 |
| May | -0.1 | -1.3 | -3.5 | -1.2 |
| June | -2.0 | -2.2 | -4.0 | -2.4 |
| July | 1.3 | 1.5 | 0.5 | 1.2 |
| Aug | 0.5 | 0.6 | 3.1 | -0.3 |
| Sept | 0.7 | 0.1 | -2.4 | -0.4 |
| Oct | -0.1 | 0.4 | 0.0 | 0.8 |
| Nov | 0.4 | 1.6 | -4.3 | -0.3 |
| Dec | 2.5 | 3.2 | 2.2 | 2.5 |
| Based on our no-thought portfolios since September 2005 | ||||
Defensives tend to do relatively well in the summer. On average, they have beaten the market in each of the five months from May to September, a time when the overall market does badly. But they tend to underperform in April and October when the market does well. This is what you'd expect from a low-beta portfolio. It's consistent with investors' appetite for risk being seasonal; it increases in April, causing a shift out of defensives, but increases in the summer causing a rotation back towards them.
Value stocks, however, have the opposite pattern. They do badly in the summer (except, oddly, in August) but better at the start of the year. This suggests investors want to take on cyclical risk early in the year - our portfolio comprises 'deep value' stocks which tend to be cyclical - but avoid it in the summer.
Momentum stocks tend to beat the market in all months, except for May and October. But their outperformance is bigger in the winter. It averages 1.4 per cent per month from November to April but only 0.2 per cent per month from May to October. This suggests that momentum stocks on average have a highish beta, tending to outperform when the market does well. (This fact does not, however, explain momentum's great returns; pure high-beta stocks have actually underperformed the market in the last 10 years.)
Now, these results in themselves aren't terribly powerful because 10 years is a small sample. However, the fact that the FTSE 350's seasonal pattern in this period looks similar to the longer-term pattern for the All-Share index makes me suspect that these results might be representative of a longer-term picture. And while nobody holds precisely my value, defensive or momentum portfolios I suspect that my results for these are similar to other value, defensive or momentum strategies.
The point of all this is not to recommend that you now dump defensives and get into momentum and (in a few weeks' time) value stocks. Dealing costs would erode much of the profits you'd make from doing so. Instead, there are two points here.
First, all this corroborates the theory that appetite for risk is seasonal: defensives do relatively well in summer and cyclicals relatively badly, consistent with appetite for risk falling in these months, albeit from high levels.
Secondly, if we do see defensives underperform in the next few months and value and momentum do well over the winter we shouldn't interpret this as a permanent shift to a bull market and to 'risk on' trading. If history's any guide, it'll simply be the usual historical pattern.
