Warren Buffett Takes a Stand on Risk
by Pat Dorsey | 05-02-01 | 06:00 AM
Last weekend thousands of faithful investors made the trek to Omaha, Neb. for the Berkshire Hathaway BRK.B annual shareholder meeting, at which Warren Buffett held forth on stocks, accounting, and the meaning of life. As usual, he dispensed a variety of bons mots, one of which sparked quite a debate among the stock analysts here at Morningstar. First, the less-controversial quotes--or at least, the ones that didn't raise a ruckus hereabouts:
On the long-term returns of equities: "People who expect to earn 15% on their investments over the long term are living in a dream world."
On sell-siders/institutions and accounting ethics: "Institutions are more concerned with form than substance."
On hyping tech stocks: "The majority of the wealth created over the past two years has been not by great performance, but by great promotion."
On investing strategies: "People who talk about 'value' and 'growth' stocks really don't know about investing." (Buffett's partner in crime, Charlie Munger, has phrased this same concept more colorfully: "The whole concept of dividing it up into 'value' and 'growth' strikes me as twaddle. It's convenient for a bunch of pension fund consultants to get fees prattling about and a way for one advisor to distinguish himself from another. But, to me, all intelligent investing is value investing.")
Now, here's the Buffett quote that stirred up a veritable hornets' nest of debate here at Morningstar (my e-mail inbox is still smoking): "If someone starts talking to you about beta, zip up your pocketbook…. There are two kinds of risk: 1) The risk that you could get a better return on your money elsewhere (opportunity cost) and 2) the risk of permanent capital impairment (financial health)." (For those of you who are unversed in Modern Portfolio Theory--and you ain't missing much, by the way--"beta" is simply a way of measuring volatility, which is often used as a proxy for risk. For more on beta, click here.)
Innocuous though it may seem, Buffett's definition of risk flies in the face of just about every piece of academic finance published in the last couple of decades. From the academic perspective, beta is intuitively attractive because it measures the likelihood that an asset will go down (or up) more (or less) than a specified benchmark, which helps investors get a handle on whether they might have to sell the asset at a loss. Moreover, it has the very desirable mathematical quality of being quantifiable. (Concepts like opportunity cost and financial health are somewhat fuzzier.)
From Buffet's perspective, however, volatility just doesn't enter into the equation, since he's willing to hold stocks for long enough to make the risk of exiting at less than the stock's fair value relatively low. From a purist's viewpoint, this is interesting, since it means that risk essentially boils down to valuing the company well and making sure it's financially healthy enough to survive. But needless to say, this viewpoint ignores the very real stomach-churning feeling that afflicts many of us when we watch our investments tank--and it ignores the irrational behavior (selling at the bottom) that this feeling can cause.
For what it's worth, I haven't quite made up my mind on whether volatility is as unimportant as Buffett thinks, but it's certainly a refreshing alternative to the almighty beta. In any case, I thought you'd enjoy a recent interchange among my colleagues on the issue, and I've included a much-edited version below:
From the pro-beta camp: "I agree completely that risk to the Oracle--one of the wealthiest individuals in the world who will never have a need to sell a stock except when it is the most opportune time--is adequately defined by the two statements. Perhaps for the other billions of people who own securities, the price that we sell it at--and it may be before we die--does make a difference, and that's why volatility matters."
A rebuttal: "If you have a three- to 10-year time horizon, volatility doesn't much matter. Volatility only really matters if you have a time horizon of two years or less--maybe three years tops. The first part of Buffett's definition of risk, opportunity cost, encapsulates this anyway, because it basically says that if you pay too much for a stock, you are taking on too much risk."
The Beta Boys respond: "Opportunity cost by definition is the cost of not doing the next best alternative. We can only know what our next best alternative is in hindsight, which doesn't help anyone make a decision to buy or sell a stock today. On the second point, time and time again, people sell stocks because the price falls too much for them to stomach, even though they initially believed the company was solid and told themselves they had a long time horizon. Unless you have a rock-solid stomach, if your $100 stock falls to $25, your 10-year time horizon just went out the window and you sell. Most people know they shouldn't, but they do anyway. In all seriousness, it seems knowing the volatility of a stock could help investors with various time horizons make better investment decisions, especially those of us who like to sleep at night."
The Beta Bashers riposte: "If you are on top of a company's fundamentals and you buy in at a price below fair value, you have eliminated all the risk you can as an investor. This of course assumes you are using the investment vehicle the way it is intended--long term. Buying in at a price that produces optimal expected future returns compared with other investments--in other words, taking seriously the price risk/opportunity cost part of Buffett's risk measure--puts you most of the way there. I think people (me included) just buy stocks when they are overvalued too often. Volatility isn't the problem, valuation is. If you don't feel comfortable valuing it, don't buy it."
And a last word from a colleague who feels that volatility matters: "Although you are correct in saying 'you have eliminated all the risk that you can as an investor,' people generally invest their money for real purposes. Those purposes are usually time-specific: buying a house, retiring, etc. To say that 'volatility isn't a risk unless you aren't investing right' is a cop-out. Maybe we are getting too far into semantics, but just because you can't control a risk doesn't mean that it isn't a risk." |
