Bernie Madoff Offered Glorification of Smoothness: John Dorfman

Commentary by John Dorfman

Jan. 16 (Bloomberg) -- I admit to feeling a twinge of schadenfreude in regard to the Bernie Madoff scandal.

Madoff, a former chairman of Nasdaq, is accused of running the largest Ponzi scheme in history, swindling investors out of as much as $50 billion.

The reason for my emotional reaction is simple. In the past few years, two of my clients and several prospects have asked me, in effect, “Why doesn’t your firm have nice smooth returns like that guy Madoff?”

Now they know. The government says Madoff was paying off early investors with the funds incoming from later investors. Madoff, who has posted $10 million bail, awaits trial.

One quality that made Madoff’s investment vehicles attractive was the metronomic uniformity of his performance. From 1991 through 2003, Fairfield Sentry Ltd., a fund managed by Madoff, reported annual returns of between 8 percent and 18 percent every single year.

From 1990 through 2005 it reported an average monthly gain of 1 percent, with 96 percent of all months profitable, and with a maximum monthly loss of 0.56 percent.

I would argue that investors place too high a value on such level returns, also known as low volatility, or low standard deviation.

What matters most, in my judgment, is the compound average annual return achieved over a reasonable length of time -- say, five or 10 years.

Lumpy 15 Percent

As Warren Buffett, one of the world’s most successful investors, wrote in a 1996 report, “I would much rather earn a lumpy 15 percent over time than a smooth 12 percent.”

Some academics and investment practitioners place a great value on low standard deviation of returns. They reason that returns that don’t vary greatly from one year to another are more likely to be replicated.

The best returns, on a multiyear basis, come from investment managers who make bold decisions, and who often invest contrary to prevailing wisdom. People who follow these practices rarely have even, predictable annual results.

Give me a Ken Heebner, whose CGM Focus Fund was down 49 percent in 2008 but up 79 percent in 2007 and 66 percent in 2003. In spite of volatility, his fund ranks in the 97th percentile over the past five years and returned a compound average of 26 percent per year before taxes in the 10 years through 2007.

Other managers I have admired over the years, including John Templeton, Peter Lynch, Michael Steinhardt and John Neff, also had considerable variation in their annual returns. Theirs is the approach I seek to follow.

Gauging Risk

Measuring risk is tricky. The widespread use of statistical measures such as beta and standard deviation has resulted in what I consider a cult of smoothness. It focuses on what can be measured, but what is easiest to quantify isn’t necessarily the best measure of risk.

Consider two hypothetical investment accounts. One of them - - we’ll call it Low Volatility Partners -- consistently loses 1 percent of investors’ capital each year. The other, containing the common stock of Buffett’s Berkshire Hathaway Inc., had returns of 35 percent in 1997, 52 percent in 1998, a loss of 20 percent in 1999, a gain of 27 percent in 2000 and a gain of 6 percent in 2001. The compound annual return for those five years was 17 percent.

Many academics and some practitioners, with a straight face, would say that the Low Volatility Fund was the less risky of the two. To me, such a conclusion represents a triumph of quant-think over common sense.

Actual Results

A measure of risk I like better is the biggest decline a fund has experienced since inception. For funds with track records of 10 years or more, I believe this gives a better handle on risk because it deals with actual results rather than a theoretical measure.

Then there is character risk, as investors in Madoff’s enterprise found out to their sorrow. No one is a perfect judge of character, but it is better to entrust your investment portfolio to a person you know well and trust than to someone whose numbers for the past few years look hot.

Just as many investors are seduced by the low volatility of some managers’ annual returns, so are many drawn to invest in the stocks of companies with level annual earnings. Procter & Gamble Co., Kimberly-Clark Corp. and Kellogg Co. come to mind.

In the past five years, Procter & Gamble has sold for 22 times earnings on average, Kimberly-Clark for 17 times earnings and Kellogg for a multiple of 19. They commanded a premium in the marketplace because of the presumptive steadiness of their earnings.

Schnitzer vs. Procter

For my part, I’d rather have a Schnitzer Steel Industries Inc. or a Seaboard Corp. Their earnings bounce around a great deal and so do their stock prices. But over the past decade (1999 through 2008) Schnitzer stock gained a compound annual average of 24 percent a year and Seaboard shares were up an average of 12 percent.

By contrast, Procter & Gamble and Kellogg returned only 5 percent a year (including reinvested dividends) over that decade, and Kimberly-Clark 2.4 percent.

Disclosure note: I own Berkshire Hathaway, Schnitzer Steel and Seaboard personally and for clients. I have no investment position in the other stocks discussed in this column.

(John Dorfman, chairman of Thunderstorm Capital in Boston, is a columnist for Bloomberg News. The opinions expressed are his own. His firm or clients may own or trade securities discussed in this column.)