It is quick and easy to calculate but is too often seized on to make black and white ‘cheap or expensive’ calls when there are plenty of shades of grey in between.

P/Es takes the share price and divides it by earnings per share. The lower the P/E, the less you pay for the earnings.

My particular bugbear is that pundits talk about markets being ‘historically’ cheap or expensive without defining their time frame.

Equities have traded on significantly differing P/Es over the decades making it easy to choose an appropriate timeframe to ‘justify’ a contention. The magisterial Barclays Equity/Gilt study tells us that in the UK the market traded at a single figure P/E during the 1970s, between 10 and 15 for the 1980s but then shot up to the previously unprecedented high 20s during the crazy 1990s before collapsing back to a single figure P/E by 2009.

The average P/E over the last 50 years is thus around the current level of 13x but the market has traded well away from the average for many years at a time.

Commentators often use the 20 year average of 14.5x as their ‘long-term average’ which takes in the inflated 1990s P/E but not those of the far lower 1970s and 1980s, thus making the current valuation look ‘cheap’ whereas I would say it is ‘fully valued’. Purists would in any case say investors should use a ‘Schiller P/E’ using 10 year smoothed earnings but that is a discussion for another day.

So is it fair to say low P/Es are good and high P/Es bad? Err not quite, it depends on the type of company you are valuing and the stage of its earnings cycle. The logical argument is correct for ‘secure growth’ economically insensitive companies with predictable earnings such as consumer staples and utilities.

However, you normally buy cyclical companies (industrials, retailers, auto manufacturers) when P/Es are high, paying a premium because the earnings denominator is very depressed and about to see a very strong recovery. You can extend this analysis in aggregate to markets depending on their composition.

This is why Europe at 13x forward earnings is arguably cheaper than the UK which trades on a similar multiple but comprises a higher proportion of mature ‘secure growth’ companies rather than cyclical/recovery plays with depressed earnings which are more prevalent in Europe. The concern about the US is that it is trading on 15x earnings (the 30 year average) but with a market composition that gives it less of an ‘earnings beta’ to a global recovery than Europe and Asia.

According to the boffins at Schroders, US earnings are in a far more advanced stage of recovery and are around 15% higher than their 2007 peak while European earnings are still around a third below. To say that US is expensive and everywhere else is cheap is a bit black and white and ignores the ‘when Wall Street sneezes the rest of the world catches a cold’ effect but nonetheless it is fair to say that it is the US stock market around which there are the greatest valuation concerns.

Nifty Fifty

An interesting historical precedent is the ‘Nifty 50’. This was a collection of US stocks which from the mid-1960s to the early 1970s outperformed the broader market by an annualised 15% per annum.

These were the classic ‘secure growth’ companies which investors flocked to in challenging economic times driving up their P/E valuations to a massive premium to the rest of the market. They were touted as ‘buy and hold forever’ stocks but it ended in tears during the subsequent 1970s bear market as the P/Es of the Nifty 50 contracted back to market levels, causing no end of damage to their stock prices.

Stockmarket geeks will be interested to know that subsequently some of these companies have flourished (Coca Cola, IBM, J&J, Walt Disney and Wal-Mart) whereas some have since being consigned to history including Emery Air Freight (now a small part of UPS) and Joseph Schlitz Brewing (now owned by Pabst Brewery Company).

The analogy is with the ‘expensive defensives’ of today where high quality stocks with robust balance sheets, strong and sustainable business models, high margins, predictable and growing earnings have outperformed strongly since 2009 and been driven up to premium valuations. In doing so they have become a ‘crowded trade’ with the market arguably significantly overpaying for this secure earnings growth.

For several quarters we have been suggesting a diversification of equity fund holdings to include funds with more of a recovery and cyclical bent and we continue to favour this strategy.

Robert Pemberton is investment director at HFM Columbus Asset Management