Floyd Upperman wouldn't trade without studying the Commitment of Traders reports, and he doesn't think you should, either. The author of COMMITMENT OF TRADERS: STRATEGIES FOR TRACKING THE MARKET AND TRADING PROFITABLY outlines his trading methods in articles on his website and the November issue of STOCKS & COMMODITIES as well as in his book.
The Commodity and Futures Trading Commission (CFTC) publishes the Commitment of Traders (COT) reports. The reports break down open interest held by commercial, non-commercial and non-reportable participants in certain markets. The CFTC defines open interest as "the total of all futures and/or option contracts entered into and not yet offset by a transaction, by delivery, by exercise, etc." The markets covered are those in which twenty or more traders hold positions at or above levels that the CFCT sets. Markets included in the reports comprise such diverse issues as wheat, frozen concentrate orange juice, U.S. treasury bonds, VIX futures, the U.S. dollar index, and, of course, the major U.S. indices.
The CFTC makes the reports available at the www.cftc.gov website. Although beginning with a monthly report in 1962, the CFTC gradually increased the frequency of the releases over the years. Since 2000, the CFTC has published the report weekly, releasing it on Friday at 3:30, with the information compiled as of the Tuesday preceding the release. Since 1995, the information has been provided free, in both short and long forms. An example of a short form can be seen in the wheat contract on the Minneapolis Grain Exchange as of January 3, 2006.
COT Report for Wheat Futures as of January 3, 2006:
Because Upperman came from a farming background and understood grains, he made his first commodity trade in the wheat market. He remembered conversations in which wealthy farmers had talked about their use of futures to hedge their risks. He knew that those wealthy farmers wanted to lock in profits and do it at the point in the growing season when there's the most doubt about the outcome of the crop. With a first successful trade behind him, he began employing his area of expertise--analyzing data for high-tech companies--to find patterns that would help him pinpoint successful trades. He did find patterns but didn't always understand what was behind the patterns. It was when he found the COT reports that all those areas of expertise began to jell into his trading method.
Upperman believes that the "commercial" positions in wheat listed on the COT report illustrated above could be broken into two parts: the commercial producers and the commercial consumers. In the wheat market, the producers would be the large farmers. Upperman asserts that these producers nearly always take short positions because that's the only way they can lock in their profits. They want to be able to sell their wheat at a certain price. Commercial consumers are those buying the crops, and they almost always take long positions, so that they can lock in delivery at a certain price. These commercial traders, both producers and consumers, are the ones in the know, the ones most likely to have advance knowledge of supply and demand in the crops.
Most times in the past, their long and short positions were balanced, but when there's an imbalance, it's the non-commercials taking the other side of the trade. Those non-commercials consist of funds and large traders when you're talking about wheat. Non-reportables might consist of small commercials, small funds, small traders and speculators, not big enough to be considered either large commercials or non-commercials.
An imbalance among the commercials demonstrates something about anticipated supply/demand imbalances. A large net short position shows that commercial producers believe that supply will outstrip demand, for example, because producers have been more worried about locking in their prices by going short than consumers have been about locking in the prices they'll pay by going long.
However, Upperman cautions that the mistake that most traders make is in following the commercials. The commercials scale in and out of their positions early, he claims, with the process sometimes taking a year to unfold. Besides, they rarely liquidate their positions. Upperman discovered that the non-commercials, in contrast, almost never take delivery of the product, and so will almost always liquidate their positions, looking to book profits. The funds have a higher correlation coefficient with the trends in prices, he claims, because they're the ones who will be liquidating their long and short positions. The commercials are in the market for different reasons, to hedge the prices they'll receive for their product in the case of the commercial producers and hedge their need in the case of the consumers. They serve as the balance between cash and futures rather than as market movers, Upperman claims.
Because the commodity index funds are growing, they might be changing the dynamics, Upperman states, thereby making the COT reports even more powerful tools than they were in the past. Upperman suggests that rather than following the commercials scaling into the markets, traders should be following or anticipating what the large funds are doing, since they're the ones who will be liquidating positions.
However, he recounts tales of rare instances when commercials have taken extreme positions. This does alert market watchers to an imbalance and warns them that non-commercials might be forced to make a move. The raw COT data isn't going to give traders enough information to determine when that occurs, he cautions. Such extreme positions will be different for each market, requiring the study of historical data.
Upperman uses statistical skills and proprietary studies to graph historical data from the commercials on a bell curve. He's looking for those instances when commercial positions are in the tails of the distribution curve, the times when they're at an extreme. This occurs 0.03 percent of the time. Major tops or bottoms may be being made when commercial positions move into the tails, but at the point they move into the tails, traders don't know how the imbalance will work out.
What is clear is that the large traders, taking positions opposite that of the commercials, must also have imbalanced positions. For example, if commercials have a large net short position, producers have surmised that supply may overwhelm demand, as noted earlier. While this is occurring, non-commercials are likely to be in large net long positions, opposite that of commercials. According to an article available on his website, Upperman looks for instances when the commercials reverse from those extreme positions as buy or sell signals. Such a signal was given April 7, 2000 in the Nasdaq when commercials moved positions from the tail on one side of the bell curve to the opposite tail.
Commodities are always cyclical, Upperman claims, unlike equities that are not always so. There's always at some point a reversion to the historical mean price. Upperman thinks that supply constraints may be changing that picture for crude, with a mean that might be increasing.
It's easy to understand Upperman's breakdown of producers and consumers among the commercials in commodity markets, although not always so easy to follow the rest of his arguments without a complete study of his book. Identifying commercial producers (taking short positions to lock in the prices they'll receive) and consumers (taking long positions to hedge the prices they'll) doesn't prove as intuitive when studying the stock indices. Upperman claims that the endowments, pension funds and biggest unions serve as the commercial producers and consumers when talking about the equity indices such as the SPX.
In neither of his articles, however, does he reveal the levels he considers extreme for the COT data for the SPX or other equity indices. As of January 3, commercials were net short the SPX, 77.2 percent of open interest being short and 69.1 percent being long. Without the knowledge of historical trends or access to Upperman's bell curve, it's difficult to know where such a position falls along that curve, although the position was net short more than it had been in recent weeks.
Traders who want to find out for themselves whether that position was extreme might check out Upperman's book, with Upperman reportedly discussing his methods of graphing the COT data in that book. Another possibility is to read Larry Williams' new book on the COT data. Williams has reportedly developed a method of determining proxy COT data for individual stocks. Industrious and talented statisticians also always have access to free historical data through the CFTC, where they can gather information needed to create their own studies.