“Sell in May, and go away. Come back on St Leger’s Day” is a famous stock market adage that has endured for decades, despite being bad advice.
The origins of the phrase date from the time when the City was full of toffs who became more preoccupied with the social whirl in the summer – the Chelsea flower show, Wimbledon, Royal Henley, Royal Ascot, the Epsom Derby, Cowes (not necessarily in that order) , and finally the St Leger classic at Doncaster – than earning money in the stock market.
In theory, the idea of selling in May in the pre-Big Bang 24-hour trading environment seems sound. In practice, even before the Big Bang in 1986, it was about as reliable a piece of advice as you get from the average astrologer.
Using data from Datastream, the City’s foremost provider of historical financial data, I can reveal that in the 21 years before the Big Bang, there were 15 occasions when the All-Share index for the London Stock market was at a higher value in mid-September – which is when the St Leger is run – than it was at the beginning of May.
In other words, in 21 years, selling in May and coming back on St Leger’s Day would have been a money making strategy just six times.
It is true that there were occasions when the strategy worked very well; most notably in 1974, when the All Share index fell 41.6% between May and September, but that surely had far more to do with the aftermath of the three-day week in the UK when Britain was brought to its knees by a dispute between the government and the coal miners.
What’s more, pursuing the policy the following year would have meant missing out on a 107.4% increase in the stock market between May and mid-September.
After the Big Bang in October 1986, the sell in May policy has been just as flaky. Datastream’s data shows that in just 14 of the 47 years since 1966, the market was cheaper in mid-September than it was in May. That’s less than one year in three.
I am not sure it is statistically viable to do so, but taking an average of the All Share index change between May and September each year since 1966 gives an average (mean) gain of 8.1%. Comfortingly for statisticians out there, the median change is also 8.1%.
So, the numbers have spoken: To sell in May, and go away, is a strategy that does not pay. Usually.