So-called rare or extreme events occur much more often than you might expect. Investment strategies should be set accordingly. Chris Batchelor explains.

Most investors care deeply about the global financial crisis (GFC), as it has left an indelible mark on their psyche – not to mention their net worth. Whilst markets have recovered strongly since the depths of the GFC, the shake-out left many investors suffering losses that won’t be recovered. Why did it happen and what has a large black bird got to do with it?

In sixteenth-century London the term ”black swan” described an event deemed to be an impossibility. The empirical evidence confirmed that all swans had white feathers. That was until 1697, when Dutch explorers discovered black swans on the Swan River in Western Australia. From then on the term ”black swan” came to describe an event perceived to be an impossibility but that was later found to be possible.

The GFC was a black swan event: a scenario that most financial modelling deemed impossible, or so unlikely as to be almost impossible, but which actually eventuated.

One of the most important assumptions underlying financial modelling is that investment returns follow a normal distribution. This means that the probability of returns that are significantly different from the average return is very low. Events that are at the far ends of the distributions (the “tails”) have a very low probability of occurrence, and are deemed extreme events.

There is a mounting pile of evidence that extreme events happen more frequently than probability models imply, and that their impact is more severe. Since 1987, 10 major market events can be identified, including Black Monday (1987), the Asian crisis (1997), Long-Term Capital Management (1998), the tech bubble crisis (2000) and the GFC (2008). These events are characterised by increased volatility, increased risk aversion, negative returns and increased correlation across asset classes.

Researchers have observed that during these extreme events correlations between asset classes move towards +1. A correlation of +1 between two assets means that the asset’s prices move in perfect sync with each other. A correlation of -1 means movements in asset prices are precisely the opposite of each other. Zero correlation means there is no relationship between asset price movements.

Less-than-perfect correlation between different asset classes is the justification for spreading investments across asset classes to reduce risk. When one asset is falling the effect is offset by another one that is rising (or at least falling less). The tendency of correlations to move toward +1 is referred to as the contagion effect – asset classes move down together and the benefits of diversification are reduced. During extreme events, the classic risk management strategy of a broadly diversified portfolio across asset classes may not provide the defence mechanisms investors anticipate.

Given that we have just been through a very extreme event, how likely is it that we would experience another one any time soon? Probability theory suggests that such events should only occur a couple of times every hundred years. The empirical evidence suggests that they occur much more frequently than that, possibly as often as every three or four years.

Given the likelihood that most investors will experience another extreme event or two in their investing lifetime, what protections are available? There are no simple answers. Many of the classic risk management strategies work well in normal times but breakdown when markets undergo extreme stress. Other hedging or insurance strategies can be costly to implement. The benefits may not justify the cost.

Investors and their advisers need to be aware of the possibility of extreme events and cognisant of the possible impact when they are devising an investment strategy. Knowledge may be your best defence.

Chris Batchelor is senior technical writer for Kaplan Professional.