Why Volatility isn't an adequate measure of risk

Nowadays, the most popular indicator of a security's risk is volatility. As we all know volatility measures how much the price of a security has fluctuated in the past. People see volatility as a risk indicator because, they say, if you buy a high volatility security you will expose yourself to a greater drop in the security value than if you buy one with low volatility. That's true, but the price movements that occur in the short term hardly tell something about the intrinsic risks of a business.

It's no surprise that this statistical concept has become so popular in modern markets dominated by traders which operate with very short time horizons. Those market players generally don't act on a value basis and don't know the characteristic of the business they buy or sell, instead their operations are based on their ideas of future price movements. To them, a simple price fluctuation based indicator like volatility appears very useful. They use it to estimate the potential loss/gain of a trade.

Unfortunately for them they learned the hard way one of the shortcomings of volatility: what happens tomorrow may be really different than what has happened in the last day, month or year. In 2008 while stock kept falling volatility drastically increased, reaching peaks not seen since 1987 in September-October after Lehman Brothers bankruptcy. Short term investors experienced bigger adverse developments in market prices than what they had estimated using volatility, beta and VAR. That's why many traders and hedge funds were utterly crushed by the market in the meltdown after September 15th.

The volatility of the S&P 500 is now higher than in 2006-1H 2007, but it's price is much lower. If one values risk on volatility he would come to the conclusion that the S&P is now much riskier than in 2006-1H2007. This is against any sound reasoning. How in the world can you assume a higher risk if you buy something at a much lower price?

Let's suppose than in year T you estimate that the stock of X corporation has an intrinsic value of 100. It's market price is 110 and it's volatility 20%. In year T+1 it's price falls to 60, it's volatility increases while it's intrinsic value is little changed. By buying something worth around 100 at 60 you would have an adequate margin of safety reducing significantly the risk of permanent losses, still volatility indicates than the risk of the security has increased! The intelligent investor understands that this is total nonsense and uses other reasonable methods of risk measurement.

The last problem with volatility is that it doesn't tell nothing about the intrinsc business risks. You will have to study the business until you understand it and read the financial statements to have a grasp of the intrinsic risk of a complex thing like a corporation.