Take a look at the equity weight over time of some multi-asset funds and you’d have thought that the manager was at best flipping coins and at worst drunk. Up one month, down the next, up again, down again. It is well documented that markets are unpredictable over short time frames, so can this be the right approach? I think not.
"Predicting short-term market movements is for the foolhardy"
I have espoused the merits of keeping it simple on numerous occasions, but nowhere is this mantra better demonstrated than with respect to tactical asset allocation.
Over the years, I have become increasingly cognisant of the parallels between tactical asset allocation and the operation of monetary policy (stay with me). As I gradually learned the hard way that predicting short-term market movements is for the foolhardy (one can never learn fast enough!), and began to focus all the more on the longer term, I realised that my monthly market reports didn’t change much month to month. What did change was the odd word here and there, with ‘strong’ replacing ‘modest’ or ‘am confident that’ instead of ‘believe that’. In other words, it was all about gradual change in nuance, as is the case with minutes of Fed or Bank of England meetings (that said, I very much doubt anyone will ever hang on my every word as they do with Yellen and Carney, other than my dog of course!)
I also found that as I matured as an asset allocation specialist, my proposed changes in equity weight or the weights of other asset classes became much smoother, in the same way that over the course of a business cycle central bank policy rates tend to move up gradually, stop, move down less gradually, then stop again. None of the constant chopping and changing you get with fund managers (unless your name is Jean-Claude Trichet that is!).
Most importantly, monetary policy tells you everything you need to know to be able to predict medium-term market movements with a reasonable degree of accuracy and thus to add value to portfolios through tactical asset allocation. And as far as monetary policy is concerned, all you really need to keep an eye on is the unemployment rate and inflation.
In other words, keep it simple - you really don’t need more information.
The CIA was particularly interested in this idea, and indeed wrote a paper in 2005 called, you guessed it, do you really need more information?3 In it, author Richard J Heuer Jr, referred to an unpublished 1973 manuscript by Paul Slovic entitled, “Behavioural Problems of Adhering to a Decision Policy”, which described an experiment to measure the utility of information. In the experiment, eight experienced horse handicappers were asked to choose their fifth, tenth, twentieth and fortieth most important variables found in typical past performance tables e.g. the jockey’s record, weight to be carried etc.
The handicappers were then given data for 40 actual races (sterilised to hide the race identities) and asked to predict the top five horses, in finishing order, as they were progressively given the 5, 10, 20 and 40 variables of their choosing. They were also asked at each stage to assign a degree of confidence to their predictions.
The results are shown in the chart below.
As the handicappers are given more data, the accuracy of their predictions does not increase, but in fact decreases slightly. Just as interesting, their confidence in their predictions increases materially. In other words, the more information the bettors had, the greater their overconfidence.
The problem is that while overconfidence was useful for early humans in facing down a sabre toothed tiger or a woolly mammoth, it causes all sorts of problems when it comes to making good investment decisions. And while our intelligence and knowledge might have increased immeasurably over the last 100,000 years, we still have the same brains and thus the same instincts. Financial markets are there to make you look stupid and understanding this would make most people better investors.
Back to simple asset allocation.
The next two charts illustrate the simple link between monetary policy, future returns from equities, and tactical asset allocation. Over an hour or two recently, I played around with data comprising only monthly central bank policy rates and monthly MSCI World data.
Chart 2 shows the clear link between the central bank policy rate (average of US, UK, Eurozone and Japan) and the performance of the MSCI World index, going back 20 years (both relative to trend).
Using this, one can derive an equity target weight1 for a hypothetical balanced fund that is based only on the central bank policy rate, then measure the performance of the tactical asset allocation (TAA) portfolio against a fixed weight strategic asset allocation. The results are shown in Chart 3 below.
Why have I chosen 53%/47% for the strategic asset allocation? Because I wanted to make sure that the volatilities of the strategic asset allocation index and the tactical asset allocation portfolio were exactly the same, and thus ensure that none of the outperformance came from higher beta (market risk).
If it isn’t obvious, 23%pts of outperformance over 20 years is nothing to be sniffed at. This could be the difference between portholes or balcony on that retirement cruise you are dreaming about. Furthermore, with one or two refinements that might take another hour or two, I suspect the model could deliver even better results.
Simple refinements of course, not complicated ones.
Published in Investment Letter, November 2017
The views expressed in this communication are those of Peter Elston at the time of writing and are subject to change without notice. They do not constitute investment advice and whilst all reasonable efforts have been used to ensure the accuracy of the information contained in this communication, the reliability, completeness or accuracy of the content cannot be guaranteed. This communication provides information for professional use only and should not be relied upon by retail investors as the sole basis for investment.
Comments