Lies, lies and damned P/e ratios

On 12.11.13 the price earnings ratio on the FTSE 100, a pretty common general valuation tool, was reported as being 13.9 in the FTSE Actuaries Share indices published in the FT. This caught my attention because it was lower than I had thought; sure enough, the equivalent figure for the previous day was 15.8. Had the main UK market just got 12% cheaper, although the index itself had barely moved?

I have not checked this with the compilers, and this may not provide all of the explanation, but I don’t think it can be a coincidence that on 12.11.13 Vodafone, representing something like 6% of the market, announced its interim results and hence its new 12 month earnings per share (EPS) was incorporated in the ‘e’ part of the P/e ratio.

To have such a large impact, the Vodafone EPS must have increased considerably (to increase the ‘e’ and therefore lower the P/e) and, indeed this is what we find when using the basic reported EPS. Prior to 12.11.13 the P/e ratio of the index must have been based on the Vodafone reported EPS for the 12 months to 31.03.13 of 0.87p, which was adversely affected by an impairment charge. Using the new interim numbers (and calculating the 12 month reported EPS) this number rises to 41.78p; some turn around!

Of course, this is really nonsense in terms of understanding what is going on. On an adjusted (management basis), which we use for our models, the Vodafone EPS actually fell slightly, from 15.65p to 15.44p.

This caused me to wonder what was the real underlying P/e of the market and how much it had changed since the end of last year. Using adjusted EPS we calculate that the FTSE 100 at 29.11.13 is on a P/e of 15.3, compared to the figure reported in the FT of 13.6. The equivalent figures for 31.12.12 are 11.0 and 11.5 respectively.

What to make of all this? Well, it is all a bit of a minefield, to be honest, and comparison isn’t helped by changes in the makeup of the index over the period. Whilst the index weighted by market capitalisation is on 15.3 times, on average the p/e of the FTSE 100 is currently c, 16.5 times (it was 15.3 on 31.12.12.), confirming what we know, that, larger companies tend to be cheaper at the moment. The top twenty largest companies by market capitalisation, excluding the banks, are on an average P/e of 14.6 times.

With all these confusions, it may be best to stick to dividend yield as a valuation metric (as in our asset allocation model). The FTSE 100 was yielding 3.72% on 31.12.12 compared to 3.52% now; with the market having increased by 11% since the end of last year, dividends seemed to have grown by 5%, a not unhealthy increase.