Among owners of businesses both large and small, it is generally conceded that hedging — protecting one’s self against commodity price increases — is an intelligent thing to do. Just as a man might insure his house against fire, he should also take steps to protect his business, rather than leave it exposed to the slings and arrows of commodity price fluctuations.

On this point, in principle, it seems that everyone agrees — except that so many companies don’t do it.

I was speaking recently with the owner of an ice cream manufacturing company. He is exposed to uncorrelated risk; that is, the price of the cream he purchases fluctuates. However, he cannot increase his prices to cover losses as the cost of cream rises. Like many ice cream manufacturers, he was taken by surprise when the price of butter passed $2 last year. He had not taken any steps to protect himself.

It’s not that he is unsophisticated in dealing with commodity markets. He knew the average price of butter in 2003 was $1.1450, and that the average from 2000 to 2003 was $1.2962. In fact, if you went back to 1998, the average price of cash butter at the Chicago Mercantile Exchange until 2003 was $1.2962. The resources that would go toward hedging could better be employed elsewhere in the business, he reasoned.

One way in which we try to come to terms with the plethora of information that surrounds us in our daily lives is by the recognition of patterns. Whether watching a football game or listening to a string quartet, our enjoyment is increased when we recognize a certain amount of repetition and are able to anticipate what will happen next. In dealing with commodities such as butter, we recognize patterns in charts and calculate the mean average over a period of time. The price may rise and fall, but the average mean is what the cost will turn out to be. But just how reliable is the use of an “average” when pricing butter?

Consider 1998. During the week of January 16, the price of cash butter at the Chicago Mercantile Exchange was at $1.1450. On June 5, it reached $1.7400. On September 25, it was $2.8000. Two months later, it was back down to $1.2175, which made for an annual average of $1.7780, an average more than $1 below the highest weekly average for the year. When you consider that the average from 1998 to 2003 is 1.2962, you see the average does not paint a realistic market picture.

Or 2001: During the week of January 5, the butter price was $1.1338. During the week of August 31, it was at $2.2083. The yearly average turned out to be $1.6630. During most of the year, prices were far above or below that average. For a company with uncorrelated price risk, $2 butter can be a catastrophe.  

I have also seen a preoccupation with averages paralyze purchasing managers with indecision. It’s similar to what a former major league baseball general manager called the fear of not getting the better end of the bargain. One team with a surplus of pitching needs a center fielder. Another team with an extra center fielder needs pitchers. The trade would be mutually beneficial. What keeps each general manager from pulling the trigger is the fear he will be strengthening the other team more than he will his own.

When hedging, the primary concern should be one’s own cost of production or profit margin rather than getting the better of the market. An average is only the midpoint, over a period of time, between the high and low prices at which a commodity has traded in the past. Tomorrow’s prices may be on a different order altogether.  m 

David Stark is a dairy broker with Chicago-based commodities brokerage Downes-O’Neill LLC.