Risk Analysis

A radically different approach to risk analysis, aimed to minimise risks while maximising returns

No-one ever complains about upside volatility

Monthly volatility is irrelevant for longer-term investors

Volatility is not constant, but  fluctuates with the investment cycle

Volatility can be counter-productive

The arrows show how future risks and returns differ from those back-tested with hindsight after a stock market crash.

NB: The arrows point in the opposite directions at the top of bull markets.

Volatility defined as Risk

Standard industry practice among financial advisers assumes that risk <> return is a spectrum along which investors can select the setting that suits them best. As suitability is a formal requirement of the FSA for advisers dealing with private clients in the UK, heavy reliance is placed on a precisely definable measure that can protect advisers against claims of negligence, made with the benefit of hindsight about future investment performance. Normally risk and reward are respectively defined as the volatility and average rates of monthly returns over the past 5-10 years. This diagram illustrates the practice. There are four problems with this.

Firstly, volatility is even-handed. Volatile investments are just as likely to rise as to fall. Defining volatility as risk is therefore a misuse of language as it biases investors to become fearful. Traditionally more volatile investments like shares have tended to rise - with intermissions - over the very long-term, so such fearful investors would suffer opportunity cost and erosion of real value through inflation.

Secondly, monthly volatility is irrelevant as a measure of risk for longer-term investors, except in respect of the part of their portfolio that they might have to liquidate quickly in an emergency. While short-term traders may turn over their portfolios several times a year, the typical longer term investor will hold his investments for many months or years, so random changes from one month to another are not as meaningful as longer-term prospects, as discussed below.

Thirdly, volatility is not constant, but varies with the investment cycle. At the start returns are positive, while volatility is low. Then volatility rises as prices rise further, but volatility keeps on rising as prices crash, and only peaks when prices have established a low.

Finally and worst of all, volatility analysis can be counter-productive. While seductively scientific, this often leads investors into inappropriate asset allocation as indicated by the arrows if 'safe' assets are over-priced and 'risky' assets are under-priced which happens when investors are particularly risk averse, as they have become in recent years. It can also encourage bad market timing because volatility is generally high when after markets have fallen and before they rise. Volatility can therefore become a kind of contrarian indicator for investment prospects.  

Portfolio stress-testing is now possible since the Lehmann crisis has shown what the worst is.
Stress-Testing by Maximum Drawdown

Private investors could usefully learn this lesson newly imposed by financial regulators on banks. What would happen to my portfolio if the worst happened? This has become possible because the Lehmann crisis has now shown what the worst is. It is what happens when the international monetary authorities fail to do their duty in maintaining confidence in the banking system. That has only happened in 1907, 1931 and 2008, so this provides a useful once-in-a-generation stress test. By definition, that means it is unlikely to happen again soon - principally because precautions against repetition have now been widely taken. Nevertheless it remains a recent and valid test.

The table below shows the losses incurred by portfolios containing different proportions of real and monetary assets on a temporary and long-term basis when measured in dollars during that crisis. These are represented by total returns on a global index of shares and ten-year government bonds denominated in local currencies. Here temporary means from the peak in October 2007 to the low in February 2009 and long-term means four years later. That period is selected because it represents the duration of the most frequent investment cycle.   

How much can you afford to lose? Check it out here
  Global Markets (US$) 2007-1011  
  Proportions   Drawdown  
  Shares Bonds   Temporary Long-Term  
  10% 90%   -3% 23%  
  20% 80%   -9% 18%  
  30% 70%   -14% 12%  
  40% 60%   -20% 7%  
  50% 50%   -26% 1%  
  60% 40%   -32% -5%  
  70% 30%   -38% -10%  
  80% 20%   -43% -16%  
  90% 10%   -49% -21%  
  100% 0%   -55% -27%  
  Source: Investors RouteMap      

Risk depends on market conditions

Is it really risky to invest at the bottom?

Forward-Looking Risk Analysis

Investors RouteMap analyses three alternative dimensions of risk.

Overvalued assets may continue to show good momentum, and conversely assets that represent good value may display poor momentum. Risk is therefore minimised, when trends are favourable and valuation is cheap. That is when the prospects for future returns are also best. Therefore risk is minimised by maximising returns. 

As a result of structural factors, such as the systematic under-recording of profits in some countries, it may often be less sensible to look at valuation of shares compared to peers and more meaningful to consider valuation relative to their own past, as Investors RouteMap does.