Selling shares in May isn't always the way: Research into stock market history unearths some surprising facts, says Quentin Lumsden

IT IS THE nature of stock markets that no one knows what they are going to do. That has to be the case, because if anyone knew with any degree of reliability they would be able to use the derivatives markets to amass a vast fortune.

So all predictions should be taken with a large pinch of salt. However, there is a seasonal pattern. Stock markets tend to weaken in May, June and July and be strong in December, January and April.

This phenomenon is well known to market professionals, hence slogans such as 'Sell in May and go away'. This has been quantified by Nick Glydon, an analyst at the stockbroker Credit Lyonnais Laing. He has looked at the performance of the stock market since 1945. On average it has dropped by nearly 1 per cent in each May, June and July.

The best months are April, January, December and August in that order. But the April outperformance has been less marked in the past 10 years, when January and December have been the best.

In effect much of each year's rise is concentrated in a five-month period, and particularly December and January. If the FT-30 was calculated since 1945 using only the seasonally strong 1 December to 1 May period, the index would have been over 6,600 last November against its then 2,350.

I have performed a variant on these calculations looking at the US, British and Hong Kong markets. For the UK I found that selling in May was a poor guide to strategy. The dominant feature was the returns for buying in October and holding until the following May.

Based on figures since 1975 this worked well in the US and very well in Britain, and was almost a licence to print money in Hong Kong.

My conclusion is that investors should consider buying derivatives such as FT-SE 100 call options on 1 October and again on 1 December in every year for the British and Hong Kong markets with a view to sales in the first week of May.

Meanwhile, what about selling this May? Mr Glydon is in favour. He makes two points. First he says that if there is a big move in the summer - by 10 per cent or more - it is three times more likely to fall than rise, so why take the risk?

And, contrary to expectation, if the market has already been weak between January and May it is more, not less, likely to weaken further in the summer.

For a different perspective I spoke to Robin Griffiths, an analyst at the stockbroker James Capel, who has also studied market seasonality. He agrees that markets tend to weaken after May but says this has to be set against the overall trend - in a bull market it may rise less. This year he does not expect it to work because of the influence of bond markets.

He believes that bond markets have hit bottom, at least for the time being. He expects a strong rally in bonds and in turn a summer rally in shares led by stocks such as brewers and the out-of-favour drug stocks.

I believe strongly that it makes great sense to buy between October and December. If shares have weakened first, as Nick Glydon expects, that buying opportunity will be all the better.

(Graph omitted)