This time its different……..!
Many of the indicators that I know well and trust continue to tell me that the outlook for UK equities is not great; at best the major index (FTSE-100) is fair value and at worst it is set up for negative returns over the next five years.
I have held this view for some years and still riskier assets have continued to make modest progress, so should I not be questioning their effectiveness? The answer is that I do, but my belief remains that they are accurate and that it is only the distortion of Quantitative Easing and ultra-low interest rates that have prolonged the recovery in equities since 2009 for way longer than previous cycles; at some point QE is going to end and its likely that sellers will return and find that there are few buyers for their holdings.
The longer this continues the higher the probability of a sharp correction; we should also consider the possibility as to whether it has already started – the graph below shows the last two market highs and how they compare to current conditions we currently seem to be tracing the same pattern as 2007.
Alone this is not enough – what else should we be looking at?
Using a list of twenty five bellwether shares first lets establish what are the most historically reliable indicators of future market returns for the subsequent five years; the accuracy is shown as a percentage with 100% the most accurate
- Price to Sales (77.12%)
- Price to Earnings (48.88%)
- Price to 10 Year Average Earnings (50.92%)
- Price to Cyclically Adjusted 10 Year Average Earnings (46.67%)
- Price to Peak Earnings Peak Earnings is highest EPS achieved (36.84%)
- Tobins Q (62.42%)
Interestingly its Price to Sales that is the most accurate – possibly because Accounting interferes with it the least! The close correlation between the average PS ratio of the Bellwether Shares and the subsequent returns from the FTSE-100 can be seen below.
However relying on this single indicator alone would again be dangerous, and the others whilst not as accurate are not entirely useless!! To show this I have used linear regression to create a statistical model that not only removes the risk of using a single item but also improves the accuracy of the predictive powers. This model has been suggesting negative returns for almost two years now, and the fact that these are yet to happen suggests that any downward leg in the FTSE-100 may well be more severe.
We can also look at simple averages of the key ratios across different time frames using the matrix below – only the PE and PE10 suggest that the FTSE-100 is below fair value, and only over a three year average is the index undervalued – using an average of all indicators across all six time frames the index is just over 3% over valued. However in a worst case scenario the PS ratio suggest that the index is 15-26% higher than it should be; if you remember this is the most accurate indicator and 26% too high equates to a fall of around 3.58% per annum over five years which is not too far from the statistical models implied returns.
There remains more than enough evidence to justify a cautious approach
Model Portfolio Update
I refuse to be drawn into purchases of equities at present without an extremely compelling case – there are still some out there but not many! – and remain committed to a strategy of hiding out in Index Linked Gilts with the lions share of the Model Portfolio.
This strategy pushed things forward well in the early part of the year but with the shares previously purchased struggling a little now its been a negative few months; that said it remains well ahead of the FTSE-100 Total Return as it continues its mission to avoid significant losses and make small frequent gains!