Don't Misjudge Risk

Don't Misjudge Risk
By Nathan Parmelee

 

A few weeks ago, while doing my morning news cruise, I stumbled across an investing advice piece in USA Today about whether or not it was time to sell Microsoft (Nasdaq: MSFT). As a Microsoft shareholder, I was naturally curious, but when I came to the section on risk, I was mystified.

Using historical volatility to assess future risk
This article looked at the geometric mean of returns back to 1986, and the standard deviation using historical trading results. Over that time, Microsoft averaged a return of 42.3%, but with a standard deviation of 51.8. This means that historically, the share performance has been volatile. I don't disagree with the math, but I don't see how this is useful in predicting Microsoft's risk today. The piece also assumed a steady dividend. I do agree that Microsoft is unlikely to cut its dividend, but the company has actually been increasing its payouts. I find this far more important to assessing future risk, because dividend-paying companies are generally less volatile, and for most of Microsoft's history, it didn't pay a dividend.

This approach assumes that a company is just as risky when it's large as when it was small, ignoring changes in business. If we follow this logic, we would have to assume that Wal-Mart (NYSE: WMT) is riskier than it appears, based on trading data from the 1970s, when the company was a small rural operation. And I'm stumped about why investors should care about Nokia's (NYSE: NOK) trading volatility from when it was more focused on monitors and TVs than on mobile phones.

A better approach
Investing is about making money, and at the most basic level, assessing risk is about determining the probability of losing money. That's why risk is more about valuation and a company's ability to generate economic returns on its invested capital, and less about the changes in its stock price squiggles in the past. As a general rule, when you're valuing a company and you see a high level of growth for a long period of time, that's risk. On the other hand, if a company only needs to grow at very low rates and earns returns on invested capital above its cost of capital, you're looking at a fairly attractive investment, regardless of the shares' previous volatility.

This is why I find Pfizer (NYSE: PFE) interesting, with a near 4% dividend yield and a price-to-free cash flow around 18. Despite the fact that it has drugs coming off patent, it won't take a lot of growth over 10 years to justify the company's valuation. But as much as I appreciate the management and the business at Fastenal (Nasdaq: FAST), I'm not a fan of its price-to-earnings (P/E) ratio of 38 and its price-to-free cash flow of 112. That's also why Google (Nasdaq: GOOG) is less risky today at $365 than it was in early January, when it was $100 higher. That's not to say I don't consider Google a risky investment, but I think it becomes less risky as the price declines, even if the variability in the shares, as measured by the standard deviation, is increasing.

Foolish final thoughts
Having given my treatise on why paying too much is the ultimate risk, there is a very good use for standard deviation -- determining how much risk you have taken on to obtain the gains you have recognized. It still looks at the past, but it's only measuring the risk taken, not the potential risk in the future. So the next time some wise guy tells you a company is risky because its shares are historically volatile, you'll know to tune him out. All that doesn't matter if you're paying a reasonable price for a good company.