Finance and portfolio theory did huge damage to everyone in the investment field when they redefined risk to mean volatility. The sooner we abandon this fiction and its black-box fantasies the better for all concerned.
Everyone apart from a few financeland boffins understands risk to mean what it says in the dictionary: “The chance of loss or injury.”
Volatility, the variation in investment returns, has only an indirect relationship with risk. It is only identical with risk in portfolio theory models that were never more than hypothetical and have never proved to have beneficial effects.
Because parts of these academic models were widely adopted by asset managers, they had to define risk as volatility. Regulators did not have to, but went along with the fashion – another instance of “regulatory capture” or Stockholm syndrome, which an unkind observer could describe as the FSA’s modus operandi.
The reason for the risk volatility mistake is the desire to measure things in the belief that this enables us to control them. Volatility can only be measured using statistical tools that are valid only when dealing with “normal distributions” that do not represent what happens in investment. Hence the absurdity of the “6-sigma event”, one where returns fall outside six standard deviations of the mean. Dozens of these occurred during the credit crunch, making nonsense of models that said such an event should occur only once in four billion years.
More fundamentally, the assessment of volatility using probabilistic tools is only valid if the future resembles the past. They do not work where there is genuine uncertainty. The absurd pretensions of portfolio theory, with its calamitous rear-view mirror perspective, misled everyone into believing there was less risk out there than there really was.
The right procedure for getting people to understand risk is to forget the algebra and ask them to quantify their needs – definite amounts of capital or income at a future date.
The next question is “How much would you suffer if you had less?” It is only in relation to this potential for loss and injury that anyone can really
judge the merits of an investment. If the potential for gain and loss is assessed by deriving a range of possible outcomes from the range of
historical returns from the major asset classes (as per the Barclays Capital Equity-Gilt Study) then you have as good a handle on the risk-return balance as you can get but reality will sometimes deviate from these projections, as it did in 2008.
As Warren Buffett has often said: “More maths doesn’t improve investing.” And his mentor Ben Graham, the creator of value investing, also rubbished the use of calculus or higher maths in assessing investments. Markowitz’s portfolio theory is admired for its maths rather than for its utility.
Investment is only partially susceptible to quantification because investment is not a science but an art. Indeed, one of Buffett’s best followers,
Robert Hagstrom, has written a book provocatively entitled, Investing – the last liberal art.
To prove the point, take this from Buffett himself: “We’ve never succeeded in doing a good deal with a bad person”. What have volatility or stochastics got to do with this most fundamental source of risk?
Chris Gilchrist is director of Churchill Investments and editor of The IRS Report