Business Spectator Home Page

Over the last one to two years Australian and global equity markets have recorded largely unprecedented losses and volatility. The first chart below shows the VIX Index that depicts the volatility implied in S&P500 Index Options. You can see the extraordinary spike in the probability of loss since the middle of 2007 (refer to the right-hand-side of the chart). Periods of extreme risk were also evidenced during the 2001 tech wreck.



The second chart illustrates daily movements in the S&P500 Index. (US equities account for around 40 per cent of global equities, which in turn capture about 23 per cent of all super fund savings.) The losses are spectacular: -33 per cent during the 1987 crash, -49 per cent as a consequence of the deflating of the 2001 tech bubble (which Australia was largely insulated from), and -44 per cent and counting since the global credit crisis gathered pace. (It is also interesting to observe here that equity markets were very, very slow to internalise the impact of the credit crisis and lagged debt markets by a significant period of time.)

There is a rarely-discussed argument here that as financial markets have become increasingly integrated as a result of globalisation and the emergence of the internet, they have also become more unstable as fickle investor preferences are rapidly transmitted around the world with contagion-like consequences. This is also sometimes referred to as an adverse feedback loop. Of course, to the extent that there have been structural 'regime shifts' in the risk-return characteristics of the global equities asset-class there is an even stronger case for revisiting the strategic asset-allocation decisions that have been made in the past.





So what has happened in Australia? The chart below illustrates the performance of the Australian equities market since January 1985, which is broadly consistent with the tenor of super fund liabilities. Observe how massive losses of between 40 and 50 per cent have suddenly transpired during relatively short periods of time. It is pretty scary to think that around 27 per cent of all Australian super fund money is pegged to the vagaries of this one line. And a chart depicting the performance of global equities, which accounts for another 23 per cent of super fund capital, would look uncannily similar.




To shed further light on the riskiness of equities, I have illustrated the daily returns attributable to the well-diversified All Ordinaries Index since 1985. While other asset classes, such as 10-year government bonds, 90 day bank bills, or residential real estate, display return volatility of less than 10 per cent per annum, the Australian equities market has experienced extreme losses over both short and long horizons. Unsurprisingly, its volatility is also multiples that of the aforementioned investment categories. If only to emphasise this point, you can see in the chart below that since the beginning of 2008 there have been four individual days on which the entire ASX All Ordinaries Index has fallen by more than 5 per cent (and one on which it fell by more than a stunning 8 per cent).



Since the early 1970s many academics and institutions have made the mistake of assuming that equity returns were 'normally distributed' (that is virtually never subject to extreme shocks like the 1987 crash, the 2001 tech wreck, or the 2007-09 credit crunch) and that markets were 'frictionless' (in that investors always benefit from perfect liquidity and price-discovery).

But we are increasingly learning that equities markets are not always 'efficient' in the sense popularised by University of Chicago economist Eugene Fama nor are they always liquid. The religiously-held assumption of market efficiency profoundly altered the investment landscape and led, for instance, to trillions of dollars being invested in the “index” funds provided by State Street, Vanguard and others. (Ironically, if all investors indexed their capital, the market would become perfectly inefficient.)

The market efficiency paradigm critically hinges, however, on the belief that investors are in aggregate rational and not subject to systematic behavioural biases. That is, investors’ expectations are, on average, accurate. More recently, though, pioneering academics such as Kahneman, Tversky, Shiller and Thaler have applied principles from psychology, sociology and anthropology to prove that in practice people behave in a manner that can deviate strikingly from the predictions of the efficient markets hypothesis (and rational expectations in particular).

This makes intuitive sense if we cast our minds back through history and consider the speculative fads and crashes since the Dutch tulip mania, the emergence of junk bonds in the early 1980s, the related 1987 stock market crash, the late 1990s tech craze and the inexorable 2001 tech wreck, through to the most recent credit crisis.

It is now accepted by many economists that behavioural biases plague human decision-making under uncertainty and can cause extreme asset price bubbles that inevitably deflate.

In parallel with these innovations in behavioural finance, academics have started to accept that capital market returns are not 'normally distributed', but rather characterised by so-called 'fat-tails'. The presence of these fat-tails or 'black swans' in asset returns, which imply that extreme events can occur with far greater regularity than the predictions of a 'normal' distribution, is also consistent with the tendency of investors to irrationally herd in one positive or negative direction, which can in turn perpetuate clusterings of extremely positive or negative outcomes, such as those observed in equities markets over the past 18 months.

Finally, to borrow the words of the famous Yale endowment manager, David Swensen, we have discovered that the liquidity that we pay a premium for in listed markets is rarely there when we most need it (such as in times of crisis).

While it is not fashionable to highlight equity market risks the fact is that virtually all commentators, economists, stock brokers, investment bankers and fund managers are 'long' the asset class. And so it is not surprising that there have been innumerable 'dead cat bounces' since the stock market correction began in late 2007.

Nonetheless deep optimism still abounds. I recently heard one 'expert panel' predict that the ASX200 would rise by 26 per cent in 2009. This seems somewhat hard to fathom in an environment where Australia is expected to experience a potentially protracted and globally synchronised recession. When questioning this projection I have met with the refrain, “Oh but the equities market 'leads' GDP.”

As one test of this claim, the chart below maps the performance of the ASX All Ordinaries Index against changes in GDP during Australia’s last serious recession in 1990/91. It can be seen that over the course of 1990 Australia’s GDP fell by 0.5 per cent. During the following 12 months GDP contracted by another 0.5 per cent. It was only in 1992 that the economy started to recover and recorded healthy growth of 3.5 per cent.

With this in mind, the news for equities investors is not good. Australia’s share market displayed no ability whatsoever to anticipate the economic rebound. The All Ordinaries Index peaked at just over 1700 points in January 1990 and it was not until April 1993 that it managed to breach this barrier again. Indeed, over the three year period January 1990 through to end 1992 the Australian share market actually fell in value by 7.5 per cent. Along the way there were, of course, numerous dead cat bounces.




In closing, one might reasonably ask: How should super funds strategically (ie, in 'equilibrium') allocate their capital? The actuaries that advise our super funds – known as 'asset-consultants' –typically leverage off the Nobel Prize winning work of Harry Markowitz to identify the long-term 'mean-variance efficient' selection of asset-classes that minimise expected risk for a given level of expected future return. Simply speaking, Markowitz’s insights provide a precise mathematical overlay for Miguel de Cervantes’ age-old aphorism “don’t put all of your eggs in one basket”.

In the chart below we have used Markowitz’s optimisation techniques to estimate a typical super fund’s “mean-variance efficient” asset-allocation based on their target level of return. For any given level of return, the chart shows the asset weights that minimise expected risk. The time horizon that we have employed – 1982 to 2008 – is the longest period over which we can measure credible total returns to all of the major domestic asset-classes: namely Australian equities; 10-year government bonds; 90 day bank bills; Australian Listed Property Trusts (LPTs); and 'net' Australian residential real estate (where we have cut the gross rental yields in half to control for transaction costs).




While there are more sophisticated ways to carry out this analysis, the chart affords a broad indication of the conservative investor’s preferred asset-allocations assuming that the risk, return and correlation characteristics attributable to each investment category over the last 26 years are a fair proxy for what they might look like in the future. Any deviations from these historical point estimates involve highly subjective assessments about the future, which few if any of us are really equipped to make.

The conclusions deriving from this long-term analysis are interesting: depending on where the super fund’s target nominal return sits on the x-axis, the equilibrium allocation made to Australian equities is between 3.6 per cent and 14.9 per cent. That is, far smaller than the 27 per cent weight that the typical Australian super fund attributes to Australian shares. Similarly risky LPTs are afforded just a 3.4 per cent to 4.4 per cent weight. By far the most valuable sources of long-term super fund returns are 10-year government bonds, cash and residential property given assumed growth aspirations of 10 to 12.5 per cent per annum. Indeed, it appears that fixed income investments have been especially unrepresented in our super funds’ asset-allocation decisions judging from the latest portfolio weight data that I previously provided.


Applying standard mean-variance analysis to data from the major asset-classes over the last circa three decades one cannot help but arrive at the following conclusion: Australian super fund members deserve an explanation as to why they are being exposed to high levels of potentially avoidable risk, which has only been further highlighted during the recent capital market cataclysm. At the very least, greater transparency is required as to precisely how funds are making their strategic asset-allocation decisions.