Cash Buffer after markets downturn?

kidmugsy wrote: »

From The Retirement Cafe: 'Lastly, let’s talk about “optimal asset allocation.” Gordon Irlam published a study in which he noted that, while it would be tremendously helpful to know the optimal allocation, his estimate of the 95th percentile confidence interval for equity allocation was roughly between 10% and 80%. Based on that analysis, I can’t know with any certainty at all whether rebalancing would move me closer or farther from optimal. Tweaking an asset allocation within 5% tolerance is, I believe, an example of the massive overconfidence prevalent in many areas of retirement planning. We feel certain about things that aren’t certain, at all.'

Let me repeat: 'the massive overconfidence prevalent in many areas of retirement planning'.

One lesson that has been true historically, and which may remain true in future, is that the returns on equities tend to be higher when the equities are bought (or held) when valuations are low, and tend to be lower when the equities are bought (or held) when valuations are high.

Are equity valuations high or low at the moment? In the US, at least, they are high.

I am struggling a bit to deduce what the positive alternative proposed strategy is.


One area where I believe the quote is misguided is that the purpose of setting up an allocation is not to try and predict the optimum. This cannot be known until after the event, and could well turn out to be either 100% equity or 0% equity. Trouble is that beforehand we dont know which.


The key benefit of setting an allocation is that it provides a control point which can be reviewed regularly but not too frequently. The alternative is to go with whatever seems right at the time, chopping and changing depending on the pessimism or euphoria of the latest blog one reads. Having set the allocation the obvious implication is that if price movements or withdrawals change it rebalancing is required unless one has a good reason to move to a different allocation.


It is no good saying 'the massive overconfidence prevalent in many areas of retirement planning' unless one can be more specific and say what parameters and strategies should be used instead of the prevalent ones.


On US equity valuations I read that the long term S&P 500 Price/Earnings ratio is about 15. The latest figure is 19.1. So rather high. However perhaps this is due to a relatively small number of large tech and internet stocks and that the bulk of the US is reasonably priced. The FTSE All Share P/E is 13.4 so perhaps rather low. As it so happens my geographic allocations have moved off plan so that US is too high and the UK too low, which demonstrates a good reason for rebalancing.