Model Portfolio solutions increasingly dominate the retail investment universe, with investors often simply categorised into one of (up to) ten pre-filled portfolio pots (or silos in the modern jargon) depending on how they risk score using standardised tests that may even claim to be psychometric (all part of that important personal service you understand). Unfortunately there are deep flaws within every step of this process but the purpose here is to focus very straightforwardly on the model portfolio bit because we are convinced this will cause significant issues for many and may even be the source of the next major investment scandal, which will leave people asking of the investment industry “Did they even know what they were doing?” to which the correct answer would unfortunately be that in many cases: “No they didn’t”.
Let’s explain briefly before we examine some simple charts. The underlying process behind most model portfolios is Modern Portfolio Theory, or MPT, which compares returns with risks across a range of asset classes and produces asset allocations designed to maximise returns relative to each level of apportioned risk. On a superficial level it produces portfolio allocations that will rarely startle an experienced investment manager, who frankly would likely start building a portfolio from similar assumptions. The key words there though are start and assumptions, after which the boring, hard work begins.
Unfortunately among those whose career specialty is not investment management this process is seen increasingly as a panacea for hard work and deep thought and, disturbingly, we see more and more intermediaries who claim to use a “Scientific Investment Process”, which I’m afraid they aren’t. Not even slightly.
Let me quote one of the leading text books on the subject of portfolio theory, demoralising thoughtful pupils before it expects them to plow through hundreds of pages of frightening looking maths:
“Basic Assumptions:
- All investors visualize each investment opportunity….by a probability distribution of returns.
- Investors risk estimates are proportional to the variability of returns. (Note: They mean volatility is the key risk measure)
- Investors are willing to base their decisions on only the expected return and risk statistics (so without understanding the Why/Context behind those figures!)
- For any given level of risk, investors prefer higher returns to lower returns….In other words, all investors are risk averse rate of return maximizers.” (We agree here)
However shortly afterwards we are told:
“Portfolio theory is admittedly based upon some simplifying assumptions that are not entirely realistic.” (No kidding!)
And that:
“All the assumptions underlying portfolio analysis have been shown to be simplistic, and in some cases overly simplistic.” (so not the scientific principles the Large Hadron Collider was built on then?)
The latter point would be semi-acceptable, occasionally, if investors were utilising the full power of the maths across specific individual investments but in most cases they are using just the most basic asset allocation outcomes, after which they feel free to choose from a menu of funds that fit the appropriate asset sector description. This is what we focus on below (although frankly much more detailed MPT analysis can throw up results as misaligned as those under discussion, which will be a topic for a following post).
For the sake of this exercise below are the graphs of two portfolios (as well as the UK stock market and a leading UK Balanced Benchmark) over a nine year period which would come in as a balanced to slightly lower risk end of portfolios, utilising well known and widely used investments and based around an asset allocation of:
45% Fixed Interest, 20% UK Equities, 20% International Equities, 10% Alternatives, 5% Cash (All charts Source: Financial Express/Andrews Gwynne)
Click/Tap to enlarge
The two portfolios labelled Higher and Low Vol have an identical basic asset allocation, yet for the Low Vol(atility) portfolio the managers are more conservative, risk averse, highly macro (economics) aware and fundamentally value oriented. The Higher Vol portfolio the managers are more adventurous, growth oriented and happy to take on more risk while worrying less about the big picture.
Interestingly you will see the Low Vol portfolio outperforms for most of the period due to the low draw-down during the GFC (we believe a cornerstone of long term planning is always minimise draw-downs…not providing a relative performance), while over recent years the Higher Vol portfolio has caught up. But it looks volatile, and hence risky, doesn’t it?
So as a thought experiment lets take the Higher Vol portfolio, remove all the fixed interest elements and increase the equity elements to 85% of the portfolio and then compare with the original:
Click/Tap to enlarge
You will note that despite higher draw-downs during risk-off periods for the 85% equity portfolio the two track remarkably closely. Indeed the correlation between the two is comfortably above 95% over most measured periods, despite the original having 45% fixed interest and only 40% equity.
(At this point it is worth pointing out the obvious, that many risk assets are correlated like never before, hence unless you are using genuinely “alternative” strategies your portfolio risk at an inflection point is probably MUCH higher than you think!…..This will be the subject for another post soon)
Now lets do the same exercise for the Low Vol portfolio (Take away all the Fixed Interest and go 85% equity) and compare with the Higher Vol balanced mix:
The two charts track very closely again, although in this case the 85% equity portfolio suffers lower volatility and hence lower implied risk than the High Vol balanced approach, while the underlying correlation is actually well below 90%.
Lets look at a Risk/Return Scatter Graph for this period for all of these portfolios:
Click/Tap to enlarge
Over the period all four portfolios provide similar returns, although the Higher Vol 85% equity portfolio has done it at considerably more risk than the Low Vol Balanced approach.
However if we look at the last few years only, which is what most models do (simply because the data crunching gets formidable) the outcome is very different:
Using this as a risk/reward guide you might easily find yourself overloading on risk, which is fine if everything keeps going up, but!
Ironically all the data covering returns from Fixed Interest over the last few years is becoming effectively null and void in terms of predicting future returns as in a Zero rate world yields can only compress so far, and hence prices only rise so far. So with each price increase the income available falls until the total return from any bond becomes effectively flat, at a mathematical null point (or in reality the risk/return gets so skewed you wouldn’t rationally invest, if say Italian Government Debt got to negative interest rates……..Oh!……What a very messed up world we live in.)
The key point is however that in reality, with risk assets moving in dangerous harmony your basic MPT portfolios ARE NOT DIVERSIFYING RISK, indeed they may in fact be magnifying it. The consequences of that is that in another market rout many a 3 or 4 risk portfolio is likely to behave like a 9 risk portfolio, with inevitable regulatory interest. The problem is clear and WILL impact many.
Yet with intelligent informed investing you can firstly diversify within basic asset classes by understanding the underlying investments in detail. You can than add alternative strategies that are genuinely diversified by understanding how they work, when they correlate, when the don’t and if either of those two events is by historical accident or a causal link.
All this requires hard work, study, humility, experience and an open mind to learning.
Yet how can so many be making this mistake, surely most investors are qualified in the maths involved? Sadly very few are in anything other than a brief overview. While there are exams offered by the mainstream professional bodies for those servicing none institutional clients (from the Personal Finance Society -PFS- and the Chartered Institute for Securities and Investing – CISI) they are hard focused exams, have very low pass rates and are rarely taken. At least however they ask when the theory works and when not, testing to see, often via detailed essay with reference to historical reality, if candidates understand the limitations of the process – of which there are many. Unfortunately the part of the investment profession that always covers this topic is the institutional side who take the CFA (Chartered Financial Analysts) exams, which covers MPT but using a multiple choice style approach which assumes a) That it is intellectually valid, and b) That it works. Which unfortunately it isn’t and doesn’t.
Put another way the young guns of best paid and allegedly qualified part of the UK investment industry believes in investment fairies. Interestingly all of the fund managers we use and trust KNOW its total B@££@<&s and would never dream of using it. Funny that.
Joking aside please be aware MPT isn’t a “scientific process”, not even slightly, and as we approach the end of a very tired multi-asset bull market, with worsening (mildly terrifying) internals and a global economy looking rather weak and vulnerable many investors may discover their portfolios are unexpectedly risky in the worst way possible.
(It is worth noting that all the managers used in the above examples are well known to us and most have been utilised by us in real portfolios so we understand the why they do what they do. Furthermore you will note all the portfolios outperform two key benchmarks. A very simple truth is good managers who have survived a decade or so in the business and have a repeatable, predictable and tested process tend to beat markets reliably. Indeed we find it hard to build example portfolios utilising funds we know that don’t significantly outperform passives over most time frames.)