Investing is not like other activities. In most walks of life, if you have strong evidence that something is a good idea, you should do it. In investing, however, this obvious principle doesn’t necessarily hold. Which is the problem we face now.
History tells us that now is the time to buy equities.
Since 1966 the All-Share index has given an average total return in real terms of 8.1 per cent from Halloween to May Day, but has lost an average of 0.6 per cent from May Day to Halloween. All of the market’s massive rise in the last 50 years, therefore, has come only in the winter and spring. This is no mere quirk. Ben Jacobsen at Tilburg University and Cherry Zhang at Nottingham University Business School have shown that there’s a similar seasonal pattern in almost all national stock markets since data began – which in the UK case is 1694.
There’s a good reason for this pattern, pointed out by the University of Toronto’s Lisa Kramer and colleagues. Our moods are seasonal. As the nights draw in we become anxious and depressed. This pushes share prices down to a level from which they subsequently rise. And in the spring, prices get a further leg up as the lighter nights increase our optimism.
The evidence that we should buy now – on a six-month view – is therefore overwhelming.
Which is precisely the problem. Investors should not leave money on the table. If they know that a strategy works they should follow it. In doing so, however, they will bid prices up and so eliminate those potential profits. Anomalies – deviations from market efficiency – should therefore disappear.
From this perspective, the stronger is the evidence that a strategy works the less inclined we should be to follow it, because it is more likely that investors have wised up.
And investors do at least sometimes wise up. For example, in the 1980s they realised that small stocks had outperformed big ones for decades. So they piled in, pushing up their prices. If you had thought in the late 1980s that there was overwhelming evidence that small caps outperformed you would have been dead right. And you’d have suffered a decade of underperformance as they did badly in the 1990s.
In the same way, John Cotter and Niall McGeever at University College Dublin have shown that several anomalies that were profitable in the 1990s have since ceased to be so.
The same thing might even be true for the whole market. In 1985 Rajnish Mehra and Edward Prescott pointed out that, for most of the 20th century, equities had done much better than theory predicted. This might have contributed to the subsequent bull market, which left shares so expensive in the late 1990s that they have since done worse than gilts.
Common sense, then, is sometimes right: investors don’t leave money on the table. Which means another piece of common sense is wrong: we should not always act on strong evidence.
Perhaps, then, Halloween is not the buying signal the evidence suggests it is.
Or is it? Investors don’t always learn from evidence and so price away anomalies. Defensive stocks, for example, have continued to do well even though we've known for years that they do so. This tells us that there are in fact big obstacles to learning.
Some are institutional constraints. To profit from the Halloween effect you must short-sell in the spring and go long in the autumn. Many fund managers, however, cannot do this. Equity-only funds must remain more or less fully invested. And many others face limits on how much they can short-sell or borrow. This is one reason why it took many years for the equity premium to disappear after it had been discovered.
Another obstacle is that the Halloween effect tells us to go against our instincts, which are to be optimistic in the spring and pessimistic in the autumn. This is difficult. As Harvard University’s Matthew Rabin has shown, we tend instead to project our current tastes into the future.
A further barrier to learning is ideology. The Halloween effect has a devastating message. In telling us that share prices are strongly affected by moods it says that investors are not the sober, rational fact-based experts that fund managers pretend they are. And as the American novelist Upton Sinclair said: “It is difficult to get a man to understand something when his salary depends upon his not understanding it.”
These obstacles to learning might explain why the Halloween effect hasn’t disappeared recently; the UK market did badly last summer, recovered nicely between Halloween 2018 and this May Day, and has fallen since.
My hunch is that the Halloween effect still exists. But that’s all it is – a hunch. Hard facts are difficult to find in finance – and they would often be useless if we did find them.
