As mentioned previously, my wife’s pension is invested using a ‘tactical asset allocation’ function dreamed up by my good self. It basically uses the long term average of the FTSE’s real CAPE (real as in adjusted by RPIX). More specifically it uses the long term average of what I call CAPE10, which is the average of the last 10 years CAPE values.
I’m not sure if this is any better than just using a longer earnings average for CAPE (i.e. CAPE is also known as PE10, the current price of the market divided by the average of the last 10 years real earnings, so you could use PE20 or PE30 as has been done in some studies to good effect). However I haven’t seen anyone else use it so it’s nice to be in virgin territory, even if the difference is likely miniscule and possibly negative.
Anyway… the current CAPE10 value is 15.8. I have estimated the long term average CAPE10 to be 17.59. That’s pretty approximate as it’s derived via various adjustments from the long term average CAPE of the S&P 500, i.e. the US market.
So my market prognostication is that:
- The FTSE 100 is current ‘undervalued’ by 16%
- The current ‘fair value’ is about 6,170
Therefore I think that future returns are likely to be slightly above average.
Richard Beddard says
Very interesting. Are you referring to Shiller when you talk about studies using PE20 and PE30 or do you know of other studies?
Deep Value Trader says
PE20 and PE30 are mentioned in Rational Pessimism, a study into predicting equity returns using Tobin's Q and PE ratios. A copy is here:
I have seen them mentioned elsewhere but I can't remember. Basically when you average real earnings over a longer period the predictions become more accurate, which seems sensible enough.
I will at some point post the calculations for the valuation above, but time as ever is in short supply!
Here is another, and newer, study on predicting future returns based on Tobin's Q and PE1, PE2, …, PE9, PE10, PE15, PE20, and PE30:
Hi Anon, thanks for the link.