A debate is raging in the superannuation industry and it centres on the definition between growth and defensive assets.
Those classifications have acted as signposts on the pathway to deliver better risk-adjusted (or more precisely volatility-adjusted) long-term returns for so many years that the terminology has become standard.
Question marks have been raised at the edges. A PricewaterhouseCoopers paper released in June questioned the wide 60-76 per cent allocation to growth assets employed by typical balanced super funds (included in SuperRatings’ performance tables) when the funds themselves apply the labels at their own discretion.
But there are far deeper flaws in the approach.
It fails on two accounts: the growth-defensive split does not take into account the underlying drivers of risk factors which apply to each asset class and it does not take into account the specific individual risks faced by each investor.
Identifying the underlying risk factors
Rather than argue about whether relatively new asset classes such as infrastructure, distressed debt, commodities and various hedge fund strategies should be classified as “defensive” or “growth”, investors would be better served by looking at the underlying drivers of risk.
These can include: economic growth; valuation; inflation; liquidity; credit; political risk; momentum; manager skill; option premium; and demographic shifts.
Many of these same risk drivers lurk across multiple asset classes making the diversification based on growth and defensive assets in many portfolios a superficial one.
How many investors have been disappointed when growth assets fall in value and their apparently diversified portfolio also plunges in value? How many investors are aware that equity and bond returns have been correlated for long periods of time such as from the 1960s to 2000s, across both rising and falling markets?
The answers lie in each asset class’s underlying risk factors rather than the simple growth-defensive labels which are still used as a proxy for diversification.
Our goal, as an industry, should be to identify the specific risk factors displayed by each asset class – and likely distribution of future returns – at any point in time. It is an approach which requires robust testing and modelling rather than generic rules of thumb. Only then can effective portfolios be built which meet the retirement goals of investors with minimal risk.
A range of research studies have found merit in this approach.
A 2009 study by Briand, Nielson and Stefek applied an equal weighting across 11 style and strategy risk premia from 1995 to 2008 to produce similar returns to traditional 60/40 portfolios but with 65 per cent less volatility. A 2013 study by Podkaminer found that a simple factor portfolio achieved slightly higher returns than traditional portfolios with about one-quarter of the volatility.
The merits of good advice
However, those studies still define risk in a way which investors rarely find useful: volatility.
A more personal definition of risk encompasses human capital (factors which affect an investor’s current and future income such as wage inflation or employment) and liabilities (an investor’s current and future consumption which includes factors such as inflation and longevity).
The advice process can help reveal exactly what is important to each investor. A more nuanced blend of financial assets (rather than those based on “growth” or “defensive” asset classes and an assessment of how much an investor can take in a bad year) can then be used to manage these specific human capital risks and to help fund their liabilities.
It is a process which puts the emphasis on the potential risks and payoffs of a more tailored investment strategy rather than on particular asset classes and their blunt definitions.