RealMoney Silver
Go
Home | TheStreet Picks | RealMoney Ideas | Earnings Calls | Analyst Upgrades/Downgrades | Columnist Conversation | Bios | Getting Started
Help | Advanced Search | Logoff


Take Commodity Speculators Into Account
By Howard Simons
RealMoney.com Contributor

5/13/2008 12:54 PM EDT

One of the more execrable aspects of modern government is that it intrudes upon our privacy and liberties while collecting and redistributing worthless information and withholding from us things we actually might find useful.

A case in point on worthless information: Grab a jar at random from your kitchen pantry and read the label. There you might find, to your astonishment, no doubt, that vinegar has no fiber and no saturated fat. Who knew?

Now try to find data on large non-commercial traders, a.k.a. "speculators" and long-only commodity index funds in the Commodity Futures Trading Commission's weekly Commitment of Traders (COT) report. Wait a minute, you say. Both classes of traders make directional bets on the price of a commodity, accepting the linear profit profiles and lognormal distributions of returns intrinsic to such bets, so they should be reported equally, right?

Well, no. According to the CFTC, the long-only indexer is a commercial trader, while another trader with reportable positions that may be either long or short is a non-commercial trader.

Restated, the pension fund that has bought into commodity investments is not a speculator, even though its bets must be exerting upward pressure on prices. Meanwhile, the old-school commodity trading adviser is a speculator, even though he may be selling futures in a losing cause. And what is this with "reportable positions"? Anything less than 350 contracts in crude oil, our topic here today, is non-reportable. For those of you with futures accounts, pop in a 340-lot order and see if your clearing firm considers it a small trade.

The Trend Is Your Friend

All directional trading systems can be classified into one of two broad categories, trend-following and mean-reverting, with trend-following being by far the more popular. As an aside, the buy-and-hold investor in stocks is really nothing more than a long-term trend follower. If the underlying value of a commodity changes, price has to change to catch up to it, and then trend-followers adjust their position to chase price.

The end result is that trend-followers always are going to be late to the party; their biggest long or short positions will be established well after the reasons for buying or selling have disappeared. This simple reality gave birth to a minor cottage industry of reading these COT reports to see when the large non-commercial traders were at an extreme position.

The theory, promulgated by those whose memories are restricted to their own winning trades, was to sell long non-commercial imbalances and buy short non-commercial imbalances. We will return to this below.

Unclassified Traders

If, however, we classify large institutional investors as commercial traders, the whole system breaks down. The CFTC reporting period closes on a Tuesday for release late Friday; the first day you can act on the data is Sunday night/Monday morning. If we map a trend oscillator lagged over this four-day gap against the net position of non-commercial traders, we see a fairly tame position by these traders during a runaway bull market.

Non-Commercial Traders and Trend in Crude Oil
Click here for larger image.
Source: Bloomberg

Does this mean the world's speculative community is not interested in crude oil or has transferred all of its activities away from exchange-traded instruments and into the cash market? Of course not; what it does mean, however, is that we have blinded ourselves to the actions of highly speculative institutional traders by classifying them as commercial traders. Their net long positions have to be offset elsewhere, such as short positions by non-commercial traders.

If this sounds like we have a government that does not know a speculator when it sees one, you are correct. Of course, baseball's antitrust exemption stems from a 1922 Supreme Court ruling wherein Justice Oliver Wendell Holmes opined that baseball was a sport, not a business. Is this a great country, or what?

What your eye cannot tell you is when the relationship between non-commercial positions and the trend changed. An econometric analysis places that time in late-August/early-September 2006; the pre- and post-segments are different at a 99.88% confidence level. This September 2006 date is several months after a March 2006 column discussing changes in crude oil futures' forward curve and the impact of commodity index funds thereon.

Mapping Traders Against the Trend

Now if non-commercial positions follow the trend, as they appear to do, we should be able to demonstrate this visually with a set of bubble charts. Let's map the trend oscillators against the net positions for commercials and non-commercials. Non-reportable positions are omitted from the analysis. Positive values of the trend oscillator are depicted in blue, negative in white, and the size of the bubble corresponds to the value of the oscillator.

In an ideal world, everything in the southeastern quadrant should be blue, except for the extreme positions indicating the last fool buying the top; everything in the northwestern quadrant should be white except for the bottom-sellers. This ideal map is not adhered to strongly with the four-day lag noted above.

Net Crude Oil COT Positions and Trend: Four-Day Lag
Click here for larger image.
Source: Bloomberg

We can play around with different lead/lag combinations and arrive at the conclusion below for a seven-day lead, which maps the trend oscillator for Friday, April 25, against the May 6 report highlighted in red: Non-commercial trend-followers' positions are nothing more than a lagging indicator of the trend.

Net Crude Oil COT Positions and Trend: Seven-Day Lead
Click here for larger image.
Source: Bloomberg

Worse, and obviously a larger subject, the data for different markets fail econometric tests for what is called "cross-sectional" pooling. Restated, the rules you develop for one market are not applicable to others, and any trading rules developed that display back-test success are nothing more than the product of data-mining. We digress.

Better Reports, Please

Political decisions can be no better than the data supporting them; the usual outcome is that they are considerably worse. Present calls for limits on speculation, something akin to what India did with various agricultural futures, should be based on information as to who is doing what in the markets.

It would be nice if the CFTC provided us with some useful information in energy and metals markets as to what the long-only index funds are doing and a breakdown of positions now classified as non-reportable. Otherwise, we are left to speculate, and that is bad, is it not?

RELATED STORIES
India's Groundbreaking Play to Stem Inflation
Need a Commodity Hedge? Check Out BVN
More Heavy Lifting for the Dollar Bulls


Howard L. Simons is president of Simons Research, a strategist for Bianco Research, a trading consultant and the author of The Dynamic Option Selection System. Under no circumstances does the information in this column represent a recommendation to buy or sell securities. While Simons cannot provide investment advice or recommendations, he appreciates your feedback; click here to send him an email.

Read our conflicts and disclosure policy.



Terms of Use | Privacy Policy

© 1996- TheStreet.com, Inc. All rights reserved.