• Tacit Thought

Bertrand Russell’s Turkey

January, 2023

The British philosopher and mathematician, Bertrand Russell, once told a sad and rueful tale about a Christmas Turkey. It was a parable to illustrate the problem of learning and particularly learning from experience.

A family bought a turkey. Every day, before school, the children would wander down to the enclosure where the turkey lived and gave him his breakfast. Every day the turkey wandered freely around the garden and every night the turkey curled up in his bed happy in the knowledge that he would enjoy a fine breakfast the following morning. Days, weeks, and months passed, and the turkey grew fat, happy and contented.

Then, one day, it was December 25^th.

Anybody who has ever considered an investment will be familiar with the regulatory injunction that the “the past is no guide to the future,” and future returns cannot be guaranteed. Russell’s turkey would doubtless agree.

The problem with this in the real world is that if the past is no guide, it leaves us all adrift in a sea of constant uncertainty.  How useful would a physician be if every time a patient presented with a set of symptoms the acquired knowledge and experience of the physician counted for naught?

In practice, we all develop rules or short-cuts, heuristics is the technical term, very much based on the idea that the tomorrow will, in fact, be like today and indeed, the day before.

How then do we bridge the gap between this problem of induction, what Mervyn King, the former governor of the bank of England, describes as “Radical Uncertainty,” and making effective decisions for our clients’ funds based on our experience in the real world.

One answer to that question is that we build a model of the market, that does assume that the past is a guide to the future. Of course, that model is statistical and probabilistic in nature, but it provides a guide both to underlying trends in financial markets, and the parameters around which we should build our risk-management; how we handle King’s radical uncertainty.

The behaviour of the FTSE 100, the most widely known and followed index in the UK, is an interesting example.

The index was established in the early 1980s and represents the top companies by value listed in the United Kingdom. In that period we have 445 monthly observations or 37 years of financial return data.

The mean rolling annual total return in that period is a surprisingly high 9.54%. The distribution of those returns is also interesting. Half of the time the FTSE has returned between 0.74% and 19.24% on an annual basis; 25% of the time it has returned between 19.25% and 57.75% and finally; 25% of the time the return has been between 0.73% and minus 32.3%.

So anybody contemplating an investment in the Footsie has a good idea of the range of returns the index has delivered over that period. From our perspective it is the 1 in 4 risk of a negative return that gives a good idea of the nature of the risks entailed in investing in the UK’s top shares.

Digging into the data still further, we find that the golden years for the FTSE 100 was the decade starting in October 1988 ending in October 98 just prior to the market peaking at the end of the century.

Why was it a “golden” period? For two reasons: firstly returns were high and stable, between 10% and 17% for that period and secondly, volatility was low and declining. The volatility (or riskiness) of investment returns halved over the period and there were no periods of longer than one year generating negative returns.

It all went wrong, oddly enough, around Christmas 1999, to the great surprise of our “inductivist” turkey, ending what Mervyn King called the “NICE” decade of the 1990s – the period of  “Non-inflationary, constant expansion.”

Looking back, the 1990s were not a good guide, by any means, to the 2000s.

Two issues stand out:

Periods of negative returns became much more common in the UK after 2000. Had you held a basket of UK equity from May 2002 to May 2005 your three-year holding period return would have been negative, at worst, by minus 11.75%

Secondly, since 2000 the mean annualised rate of return has become progressively lower whilst at the same time the level of volatility of those returns has increased.

It would be wrong to draw too firm conclusions from this set of data, but a couple of points are worth noting:

• The realised return on UK equity does seem to be declining and risks expanding.
• Whilst the past is a “limited” guide to the future, it is useful for investors to understand the behaviour of the market over extended time periods to gain some knowledge of the risks. UK equities have delivered annual returns as high as 57.75% but as low as minus 32.3% since the 1980s.
• In the long-run, equities tend to return high positive returns, but the “long-run” may be longer than you think.

Our response to this type of data is to try and ensure that portfolios are properly diversified across assets and between economies.

Your response should be to ensure that we are up to date with your circumstances and your time horizon.

After all, no-one wants to be a turkey.