Rethinking Risk – 2 Practical Measures for Individuals
When I first came across the concept of risk, as understood in investing, I was bewildered, and in many ways, I still am.
Within the world of investing, when one speaks of risk, invariably, the word ‘volatility’ will crop up. Volatility is, “a measure for variation of price of a financial instrument over time.” (source: http://en.wikipedia.org/wiki/Volatility_(finance))
Now, if I understand the mathematics correctly, it measures the standard deviation (deviation from the mean) of price over a period of time. I won’t go into the exact mathematics because 1) I don’t entirely understand it, and 2) I don’t like saying too much about things I don’t understand. What I do understand, is that volatility does not measure loss.
This is significant to me, because I place much higher priority on how much I could potentially lose, than on how much a price deviates from its historical mean.
The question is which would I be worried more about? A) Losing money, or B) watching price fluctuations. Warren Buffett’s first rule of investing states, “Don’t lose money.”
Which I think is a reasonable point of departure.
Measuring Risk
I love Singapore’s libraries. They’re air-conditioned bastions of comfortable sofas and plush leather seats. Plus, they have really clean toilets, which is a big plus for a public space. If you can avoid detection by the grumpy security guard, you can catch a nap. Or if you surreptitiously camouflage your cable to a nearby power socket, you get to charge your ipad while putting in new floors in your Tiny Tower! (I’m currently at a modest 133 levels)
Sometimes I find a book that puts a little wind in my sail, like 7Twelve by Craig L. Israelsen. In one particular chapter, he talks about measuring risk, and proposes two more measures of risk, on top of volatility.
Worst-case return
Answers the question, how bad can it get?
The worst case return is calculated given one’s holding period. Over a period of time, say a 5-year time frame, we define a holding period, of say 6 months, and generate the 6-month return, with the starting point that moves daily. That gives a range of 6-month return numbers, and from there, it’s a matter of identifying the worst 6-month return based on historical data. This sets up an expectation for how bad things can get.
In 7Twelve, Israelsen uses a holding period of 3 years, and calculates worst-case returns based on cumulative return (the total gain or loss from the start of a period to the end of a period). As Singapore has few funds with enough data to support a 3-year holding period, I’m pretty much limited to funds with a 10-year track record. For purposes of most investors, I’ll go with 1-year, 3-years and where possible, 5-years.
Frequency of loss
Answers the question, how often will it be bad?
The frequency of loss is calculated given one’s holding period. Over a 5-year time frame, and given a holding period of 6 months, we can look at how many times this 6-month return is negative. This sets an expectation for how often you’d expect to see an investment dip into negative territory, over a 6-month holding period.
Isrealsen used 3-year rolling holding periods in 7Twelve, and within the limitations mentioned previously, it’s a perfectly legitimate holding period to use.
From the Individual’s perspective
As an individual, I have neither the training nor the aptitude to make full sense of the complicated technical terms that professionals employ. There’s no shame in that – humanity wasn’t built on the merit of math alone. The challenge is to find simpler ways to express important concepts like loss, and risk.
As always, if there are better ideas out there, I’m keen to hear them.