By Jerome Booth
Asset allocation as generally practiced today is widely accepted as having major deficiencies. My book “Emerging Markets in an Upside Down World” argues that it is massively distortionary, affecting hundreds of millions of people.
Indices are seen as the building blocks of asset allocation, yet this is based on some simplistic erroneous assumptions. One is the idea that observable (i.e. past) volatility is always a good proxy for risk. A second is that as one adds stocks (within a specific asset class) to build a portfolio, one cannot get any diversification benefit after a certain point, and cannot obtain lower volatility than that of the index. Yet this is not necessarily the case unless stocks are added randomly.
Another misguided assumption is that the relationships between stocks in an asset class can be ignored and asset price behaviour can be described solely in terms of stock-specific volatility and correlation to the index. In sum, there is little more than ‘computational ease’ to justify the use of asset classes (as represented by indices) as building blocks in asset allocation.
Further, the evolution of what constitutes an asset class appears to be a function of marketing. If enough purveyors of a new product can convince a critical mass of institutional investors and their consultants that something is an asset class, then so it indeed becomes. This has resulted not only in the perceived investment universe becoming a massive distortion of reality, but worse, the gap may even be growing as the world is changing quickly, arguably much quicker than, and in different directions to, the evolution of the set of accepted asset classes.
So how should investors react to this distortion?
One approach, which in my book I have called the Entrepreneurial Approach to asset allocation (as opposed to the Universal Approach) is to stop pretending one is optimising allocation within the set of all possible investments, and instead assess each new opportunity with reference to one’s existing portfolio and investment objectives. Many of the best allocators already do this, albeit whilst paying lip service to optimisation. This means assessing all investments using the same criteria. It also, by the by, means throwing out the Capital Asset Pricing Model, which is anyway based on entirely circular logic – as proved by Richard Roll in 1977.
One can also use factor analysis, increasingly popular as an alternative or supplement to traditional asset allocation. This involves ditching some of the traditional divides between asset classes, which is good, but is still based on extrapolation of past security data. Also, there are real problems in deciding what factors to use. Macro-economic factors can be chosen, however – which, in my view, is a step forward.
Asset allocation is today an exercise in simplification – how to cope in a complex world without thinking too much. The absence of macroeconomics at the centre of asset allocation, whose engine instead is various simplistic rules fuelled by extrapolation from past data, creates periods of relative stability interspersed with periods of huge systemic risk.
We need to put macro-economic analysis back at the heart of asset allocation – and add some historical and political analysis to boot. Mathematical simplicity may be intellectually more attractive than the social sciences, but life is not that simple.
Worse, focusing on past trends, even if they are macro-economic ones, is simply dangerous when there are major structural imbalances in the global economy capable of creating very significant market disruptions … like now. Such an environment requires strategic thought, more often than once a year, but with the ability to change portfolios rapidly if needed.
I advocate using macroeconomic analysis, but as we all know economists are famous for their disagreements. And as Niels Bohr once said: “Prediction is difficult, especially about the future”. Hence I advocate scenario planning but also real-time thinking. And because we worry about agency problems and behavioural biases causing decision-makers go off the rails if we let them change their minds more than once a year, we also need to focus on agency problems and behavioural biases and adjust for them.
It’s not easy but, then again, the real world is complicated. And, after all, the individuals who entrust their savings to institutional investors deserve better.