However, the White Queen disagreed: “Why, sometimes I’ve believed as many as six impossible things before breakfast.”If we have learned anything over the last few years of crisis and economic depression, it’s that the apparent “impossible” can happen.
Here are a few.
1. Despite the accolades heaped on the UK’s regulatory financial framework, the UK experienced a run on one of its banks for the first time in 150 years.
2. In the US, Lehman Brothers, a blue chip investment bank deemed too big to fail, failed. RBS, a pre-crisis leviathan of the global banking community, collapsed into a cesspit of its own making, along with most of its competitors.
3. The interbank market, arguably one of the most important financial markets for the global economy, literally seized up as it became paralysed with fear.
4. Previously conservative and cautious central bankers started experimenting with extraordinary monetary measures including the creation of cash, mainly because they couldn’t think of anything else to do.
5. US government debt – in the pre-crisis period, the premier risk-free asset class – suffered the humiliation of a ratings downgrade.
6. And finally, that Southampton football club could establish themselves in the Premiership.
Most of these events (except for the last one) and other crisis-related events not listed here, were, in the words of the great philosopher Donald Rumsfeld, “unknown unknowns”.
That is, they were events that were deemed so unlikely that they did not warrant consideration prior to the summer of 2007.
And yet they happened.
When the crisis was at its peak, many very stupid financial market participants started talking about experiencing “six sigma” events.
Basically six sigma events should have happened perhaps only once or twice in human history, and yet we were apparently experiencing one of these a day at the height of the crisis.
What these events really revealed was the true depth of their ignorance and hubris.
Sadly the sort of catastrophic events that are normally thought to be impossible, or at least very unlikely, naturally have a devastating impact on investors’ portfolios.
But the normal approach to protecting an investment portfolio from these tail events (so called because they appear only in the extreme of any probability distribution) is simply to hope that they do not happen.
However, this is not how we approach other fairly remote but possibly devastating events elsewhere in our life. For example, we insure our homes against the risk of fire and so on.
Unfortunately, one of the reasons why investors do not insure their portfolios against “impossible” events is that the cost is normally prohibitive.
At Cass Business School’s Centre for Asset Management Research (CAMR), we have been investigating whether it might be possible to employ simple investment techniques for limiting these downside risks, from wherever they may materialise.
The most promising approach that we have identified is referred to as “trend following”. The approach involves identifying whether a trend in an individual asset, or financial market index, is positive or negative.
Money is invested in the financial instrument when its trend is positive, naturally, but once the trend turns negative, the money set aside to invest in that asset is placed into a risk-free asset class, normally cash.
So it is a binomial decision process: an investor is either 100 per cent invested in the asset or 100 per cent in cash.
So how is the trend identified? There are an infinite number of ways that this could be done, but we have found that the following rule works very well: if the price of the financial asset today is above its 10-month moving average, we define that as a positive trend and invest.
However, if the price of the financial asset today is below its 10-month moving average, we define that as a negative trend and invest in cash instead.
Our research results are based upon financial market data spanning the period from 1993 to 2011 on components of five broad asset classes – developed and emerging market equities; developed economy government bonds; commodities; and commercial property.
In total we gathered return data on 95 sub-components of these five broad asset classes.
Figure 1: Average annual return, 1993-2011
Source: Cass Business School
The first two bars in Figure 1 show the average return achieved from a simple buy-and-hold strategy where we invested 20 per cent in each of the five broad asset classes, and from a trend-following strategy applied to the five broad asset classes.
With respect to the latter, the maximum invested in one of the broad asset classes was 20 per cent when it was in a positive trend and 0 per cent when it was in a negative trend, with 20 per cent of the portfolio invested in cash instead.
As such, if all five broad asset classes were deemed to be in a negative trend, then the fund would have been invested entirely in cash.
The average return from the trend-following strategy was 9.2 per cent a year compared with just 6.7 per cent a year for the buy-and-hold comparator.
This additional return over nearly 20 years really adds up.
The remaining bars in Figure 1 represent a comparison between an equally weighted buy-and-hold approach to investing in, for example, a set of developed economy equity markets, and a trend-following, equally weighted, equivalent approach.
In every case, the trend-following approach produces a massively superior return performance.
This is particularly true in the case of the set of emerging equity markets in our sample, where the buy-and-hold approach produced an average annual return of 5.48 per cent while the trend-following approach generated 12.6 per cent.
But what about cutting off the tail? Figure 2 shows the maximum drawdown for each of the strategies shown in Figure 1.
The maximum drawdown is the maximum peak-to-trough fall that a portfolio experiences over a given period, in this case from 1993 to 2011.
Figure 2: Maximum drawdown (%)
Source: Cass Business School
The maximum drawdown for the buy-and-hold strategies is high, ranging from 46.6 per cent to 67.2 per cent.
But the downside risk, represented by the left hand-tail of the return distribution, is massively reduced for all the trend-following approaches, with the exception of government bonds.
For the strategy applied to the five broad asset categories, it is as low as 7.4 per cent.
Overall, trend-following seems to enhance returns and cut off the majority of tail which causes such distress to investors.
It is results like those presented in Figures 1 and 2 that have convinced us that trend-following may be an ideal approach for risk-averse investors, particularly those approaching retirement or those in retirement that may be in income drawdown.
It is effectively long-only investing but with built-in risk management.
Something that you and Alice might have thought impossible before reading this article.
Andrew Clare is a professor at Cass Business School. The views expressed here are his own.
