Thursday 30 October 2014

PE & Prospective Returns

Last week we had a look at a pretty reliable ratio that looked at a shares market capitalisation and how this could be used to identify periods of over and under valuation. This week we have been looking at some more, primarily the PE or Price to Earnings Ratio  that is widely used  throughout  financial services.

The PE ratio is derived from the current price divided by the earnings per share (EPS) , so if the price is 100p for a share and the earnings per share the company has generated 10p then the PE Ratio is 10.This means that investors are willing to pay 10 times the annual earnings for a share; a higher share price 200p and the same EPS would  by 20 - similarly a decline in EPS would also see the PE Ratio rise .

Conversely the tried and trusted strategy of buying a share with a low PE Ratio, caused either by a falling share price or decline in Earnings Per Share when  investors perceive  the shares prospects to be poor has often yielded positive results.

The PE ratio is not perfect lots of things can alter the value of a share and the earnings per share, from negative sentiment to  one off events that effect the earnings for a year; however it is useful to see where  we  are in comparison to history.

In this instance rather than look at an individual share we thought we would construct an average from some of the larger companies (based on market capitalisation)  in the FTSE-100 to see if this would give us any clear indication of market direction.

We used 25 different shares across 19 different sectors that have all had a wide and varied history over the last 20 years or so. The  graph below shows how the PE peaked in 1999/2000  and is currently higher than 2007/2008 market  high. 



The following graph shows the percentile rank for the PE e.g. 90% means that the current level is in the top 10% of high readings of the period analysed. We then overlaid a graph of actual 5 year annualised returns (inverted) from the FTSE 100 which shows  quite a close correlation and suggest that we should not be expecting too  much form the index.









No comments:

Post a Comment