The recent evaporation of Amaranth Advisors, a $9 billion commodity-oriented hedge fund, provides plenty of fodder on this topic. We have commented previously on the explosive growth of hedge funds: now there are an estimated 9,000 such funds with approximately $1.2 trillion under management. In a post-9/11 world, many investors have tried to cope with higher perceived risks by seeking more exposure to alternative assets such as hedge funds, commodities, real estate, venture capital and others. Ironically, strong economic fundamentals worldwide have simultaneously generated a substantial amount of excess liquidity that needs to be invested, and as this capital has sloshed its way into newer investment vehicles in search of a few extra basis points of return, it’s debatable whether investors and managers alike truly comprehend the risks. Many of these alternative funds, for sure, offer beneficial risk dispersion. But as the Amaranth saga unfolded, a book by Nassim Nicholas Taleb, Fooled by Randomness, provides great insight into how such meltdowns can afflict even the most revered of hedge funds, including the 1998 collapse of Long Term Capital Management. Taleb’s writing comes across as occasionally condescending (he runs a hedge fund, after all), but it is an intellectually rigorous examination of risk that many investors would find well worth reading.

In very simple terms, one of Taleb’s arguments is that most of the world of finance (and its definition of risk) is improperly built upon a convenient statistical device many of us remember from school, the bell curve. The bell curve says that when one measures most data sets, the results tend to hover near the middle. If we were to evaluate a class of schoolchildren, most of the measured heights would be close to the average, although there would be a handful of relatively tall or short kids. But in this example, extreme outliers would be so rare that they could probably be ignored: the odds of finding a 9-foot tall kid are basically 0.00%. So the bell curve works very well in describing data in the physical world – height, weight, even test scores – but it may be a very poor tool for finance. As history shows, extreme outcomes do occur. After World War I the German mark moved from four per dollar to four trillion per dollar. The diligent work of several Nobel Prize economists at LTCM failed to anticipate the unlikely correlation of numerous investment markets. Because in the short run financial markets are so susceptible to human emotion (i.e. fear and greed), extreme outcomes are possible and cannot be ignored. The bell curve, unfortunately, tends to seriously underestimate the risk of such extreme outcomes, and while the shortcomings of this tool have been highlighted for decades, it remains standard teaching inside our nation’s MBA programs. Taleb admits that because we know so little about the probability of some very remote events, certain risks are just not worth taking at all. And so Taleb counsels plenty of diversification to his clients and readers.

One could imagine that Amaranth’s downfall was simply due to a bad, leveraged bet in the natural gas market by a young rogue trader. But when considered in light of the explanation given by the firm’s founder, the structural weakness of the Amaranth’s risk models is highlighted: “We viewed the probability of market movements such as those that took place in September as highly remote…But sometimes, even the highly improbable happens.” Furthermore, the founder defended the “well-credentialed and experienced risk professionals” who monitored the firm’s portfolio risks.

These comments highlight a troubling overconfidence at Amaranth (and perhaps other funds) that rely heavily on theoretical risk models to create “optimal” investment strategies. But markets don’t always cooperate. As Taleb might argue, traditional risk models can create a false sense of security because of the difficulty of assessing the probabilities of rare, unknown events. When mixed with occasionally illiquid markets – as well as the substantial portfolio leverage at firms like Amaranth and LTCM – the consequences can be disastrous. Warren Buffett once wryly remarked that it’s only when the tide goes out that you learn who’s swimming naked.

It’s worth briefly repeating that our theory of risk management at Harris Associates shuns the theoretical. We judge the most important risk our clients face as the permanent loss of capital, and so our philosophy and process are geared to emphasize absolute rather than relative metrics. We insist on buying businesses at a substantial discount to intrinsic value; we rely on our own experienced analytical team that is charged with considering the downside (through a regular Devil’s Advocate process) as much as the upside; we structure portfolios to match our clients’ needs and risk appetites; and our partners and employees have a significant portion of their own savings in our mutual funds. It’s also important to note that we shun leverage and the use of exotic derivatives embraced by many hedge funds: we humbly recognize that even the most sound long-term strategies can be undone by temporarily illiquid markets and creditors who might force liquidation at inopportune moments. Without such distractions, the growth in our clients’ capital over

the long run is more directly tied to the fundamental growth in per share value of our portfolio holdings, a relationship that properly recognizes that our defining skill is the assessment of the intrinsic value of operating businesses. Even if the stock market fails to recognize the full value of the enterprise in a timely fashion, our clients end up with a true ownership stake in a valuable, growing business. Such a long-term investment strategy, as we know, does not perform well in each and every season. Our approach, however, has consistently emphasized this discipline, and we have been willing to endure stretches of mediocre performance in order to maximize long-term results without, in our opinion, taking on undue risk.

Today, the investment environment remains benign: slower yet positive economic growth, low unemployment, interest and inflation rates, and a relatively favorable tax regime. On a company-specific level, although a slowing in the rate of growth will pressure record-high profit margins, we continue to emphasize better quality, larger businesses with strong balance sheets and dominant market positions that continue to trade at parity with lesser quality peers. Our current market outlook is unchanged by movements in the commodity markets or Amaranth’s demise. Frankly, we have no clue whether the next monthly move in commodity prices is up or down, but we continue to believe that technological advances and free markets generally solve long-term resource supply problems. In the meantime, we expect those markets to remain volatile, and we would not be surprised to see further stress among hedge funds as recent inflows of cash put pressure on some managers to adopt more aggressive strategies.

While our demeanor has been bolstered by recent market trends, we continue to see meaningful discounts to our estimates of value among our portfolio holdings; we fully expect those discounts to narrow over time. Our investment strategy is certainly not without risk – we humbly admit our holdings face countless economic, financial, and competitive hazards each day. We do our best to protect against such risks in a variety of common sense ways that have served our clients well over three decades. But we believe the risks our clients face are entirely different compared to those who play the current hedge fund game with simply a theoretical or mathematical perspective.

Edward S. Loeb