|| Current Members Log-In |||View Your Shopping Cart|
By William L. Jiler
As a student of commodity market analysis some 35 years ago, I recall my teacher, the late Victor Lea of the New York Institute of Finance, saying that commodity price forecasting would be simple if we only knew who was buying and selling futures contracts, and why. The U.S. Department of Agriculture (USDA) tried to partially address this problem by issuing a report roughly categorizing open interest positions. For many years, The Commodity Exchange Authority of the USDA issued a monthly publication entitled "Commitments of Traders Reports" which broke down month-end open interest of "Reporting" (Large) and "Non-Reporting" (Small and/or Foreign) traders. The statistical tables in the report indicated how open interest is allocated among large hedgers and speculators. And, by subtracting large traders' commitments from total open interest, the positions of small traders (both speculative and hedging) are also derived and presented in tabular form. In 1976, the new Commodity Futures Trading Commission (CFTC) continued these reports, and expanded them to include markets not previously covered, as well as new active futures markets. Because of computer and budgetary problems, the reports were discontinued from January 1982 through November of that year. The reports were resumed in February, 1983 to include statistics for December of 1982, and again expanded to cover the dynamic new financial futures markets.
At COMMODITY RESEARCH BUREAU (CRB), we developed techniques to utilize these statistics, and found them to be a valuable aid in the technical forecasting of futures price movements. The original research of approximately 23 years ago has not been significantly altered, and appears to be valid for the markets of the 1980's. The pertinent statistics have been gleaned from these reports and published in tabular format as a regular feature of the CRB FUTURES CHART SERVICE. These statistics are continually analyzed, and when they provide important clues to futures trends, the findings are published in the "Technical Comments" section of the weekly publication.
We will try to describe how these "Commitment" reports are analyzed and used to help forecast futures prices. We will also present the basic charts that can help you utilize the same techniques for "do-it-yourself" forecasting.
"Commitments of Traders" Reports
The official CFTC report is scheduled for release on the 11th day of each month, or the next business day if the 11th is a Saturday, Sunday or holiday. There are a number of terms used that require further explanation that will aid in your understanding of the report. Open interest refers to the number of futures contracts that have been entered into and not yet liquidated or fulfilled by delivery. A purchase of one contract, when balanced by a sale of one contract, is considered as a total of one open interest contract. A "Reporting" (large) trader is one who holds or controls a position in any one future of a commodity on any one contract market equalling or exceeding the quantities established by the CFTC as a reporting level; e.g. in March 1985 the reporting level of wheat, corn and soybeans was 500,000 bushels. In the tables, "OLD" refers to old-crop futures based on marketing seasons. In spreading, "ALL" futures in each commodity includes each trader's long and short positions without regard to which crop year/calendar year or market is involved. It is very important to note that because of intermarket spreading, long and short spread positions in a given futures market will not be equal. This causes confusion because percentage figures usually do not add up to exactly 100%. For your examination, a copy of a recent table on soybean open interest appears in figure 1.
Over the years, there have been adjustments in reporting requirements. For example, a number of changes were made as recently as Dec. 5, 1984, when reporting levels were raised for 19 out of 27 U.S. futures traded. Among the largest increases were Stock Index futures, raised from 100 to 300 contracts and T-Bonds from 150 to 300 contracts. Soybean oil, soybean meal, sugar and copper were raised from 100 to 150 contracts. These changes were made to reflect increased trading activities and should not distort the historical value of the statistics.
The CFTC has also continuously expanded the number of markets covered. Although they are not new markets, statistics for copper and gold were finally added starting with December, 1982 figures. While a two-year history of monthly positions may not prove to be a reliable benchmark of normal positions, we have found them to be particularly helpful in following recent market action. We are showing charts of these commodities in this article for future study.
To show how we use the government report to compile our open interest statistics, we refer you to the soybean table in figure 1, exactly reproduced from the February 1985 CFTC "Commitments of Traders" report. We used the figures under "Percent of Open Interest Represented By Each Category of Traders," and the "ALL" line from the first column. You will note that Commercials (Large Hedgers) were long 32.7% and short 39.1%. This indicated that they were net short 6.4% at the end of February.
In figure 2, we reproduced the Open Interest table from our weekly CRB Futures Chart Service, which shows how we presented all the statistics from the February 28, 1985 "Commitments" reports. Under soybeans, we rounded off the percentage figures from the official report to 33% long and 39% short. This made the net short figure 6%. We also showed the change from the previous month of +18%, indicating that they had increased their net long position by 18% (The actual number went from -24% to -6%). From the Non-Commercial Reportable Positions, we get the Large Speculators position. You will note the net position of -5% derived from Long or Short Only column in figure 1. The Small Traders statistics are from the Non-Reportable Positions.
Long Term End-Month Average Positions
Now we come to the very important long term end-month average or seasonally normal positions for each major category of open interest. These studies, presented in chart format at the end of the article, are the key to the analysis of the "Commitments" reports. Without a thorough knowledge of the average or normal open interest configuration, reported positions do not have much meaning. The charts will make this article invaluable for understanding the monthly reports and should have long-lasting use. We've been updating these averages for over 20 years, and surprisingly, the basic configurations do not change much for those markets with a 5 or 6 year history. Those few markets with only two or three years of history may prove to be a different story. They should be viewed with some suspicion, and updated as new statistics are received.
The Forecasting Methodology
Basically, we tried to determine the "forecasting" performance of the major identifiable groups of market participant—Large Hedgers, Large Speculators, and Small Traders. It was logical to assume that the larger and more sophisticated traders should have market insights that would enable them to predict futures price movements, if not infallibly, at least more accurately than the small traders who presumably included the "uninformed public." We also thought it was possible that the sizes of the various market positions, at different times, could well result in a type of self-fulfilling prophecy.
From the statistics in the "Commitments if Traders" report, we were able to approximate the net positions at the end of each month for Large Hedgers, Large Speculators, and Small Traders. We averaged their month-end statistics over a number of years to find out what their normal positions would be at any given time of the year. We then compared each group's actual position with their so-called normal position. Whenever their positions deviated materially from the norm, we took it as a measure of their bullish or bearish attitude on the market.
By studying subsequent price movements, we were able to establish "track records" for each of the groups. As anticipated, we found that Large Hedgers and Large Speculators had the best forecasting records, and the Small Traders the worst, by far. We were somewhat surprised to find that the Large Hedgers were consistently superior to the Large Speculators. However, the predictive results for the Large Speculators varied widely from market to market.
The differences between their current net open interest position and the seasonal norm supply us with a tangible percentage measure of the degree of bullishness or bearishness of each group towards a particular market to a certain extent. From these "net-net" figures, we obtain a configuration of market attitudes of the principal players. From our research and long experience we have drawn up some general guidelines:
The most bullish configuration would show large hedgers heavily net long more than normal, large speculators clearly net long, small traders heavily net short more than seasonal. The shades of bullishness are varied all the way to the most bearish configuration which would have these groups in opposite positions-large hedgers heavily net short, etc. There are two caution flags when analyzing deviations from normal. Be wary of positions that are more than 40% from their long-term average and disregard deviations of less than 5%.
How the Technique Is Used in Technical Analysis
Technical analysis for the purpose of price forecasting is the study of market action. Our definition of market action includes price, volume and open interest changes. There are many practitioners of technical analysis who use all varieties of charts, mathematics, oscillators, cycles and computer techniques. We also try to cover all aspects of technical analysis; however, our preliminary tools are bar charts showing high, low, close, and total volume. A line chart depicts total open interest. The "Commitments of Traders" reports are an extension or elaboration of open interest.
Although the analysis of these open interest statistics is a secondary technique, we have gained increasing respect for its validity as a forecaster of major price swings. It has helped us to anticipate turning points in the market, as well as confirming chart analysis and avoiding whipsaws.
We'd like to present some examples of how we utilized this open interest analysis in our "Technical Comments" section of the CRB Futures Chart Service. In late August of 1983, we turned bearish on sugar when it was over 10¢ a pound. Throughout 1983 and 1984, we advocated a bearish stand even though prices had dropped below 4¢ to 16-year lows. An important reason for our doggedness, in addition to the bearish chart, was our analysis of the "Commitments" report. For over two, years, the Large Hedgers' average net short position was over 20% larger than their previous 6-year average. Small traders, despite tremendous losses, averaged almost 20% higher net long positions throughout the entire debacle.
In August of 1983, Chicago wheat futures soared to new contract highs. The charts were very bullish, which we acknowledged in our "Comments" of August 12, 1983. However, we noted that the latest "Commitments of Traders" report sounded a negative note. Large Hedgers were 36% net short and Small Traders were 24% net long, both way over their 10-year averages at that time. Subsequently, the market topped out and prices trended lower for the next 6 months.
A study of the open interest configuration for corn and soybeans just prior to their spectacular bull move in the summer of 1983 will show how the analysis "did" and "didn't" work. It worked for corn, which showed Large Hedgers with net long positions well above normal and Small Traders net short. This bullish pattern was just the opposite of the soybean open interest. Here, Large Hedgers were heavily net short and Small Traders had a net long position of 20% versus a more normal 10% for June. Yet, both commodities enjoyed similar bull moves. An unforeseen drought that summer probably accounted for the strange results.
While we have shown only some relatively recent examples of this kind of open interest analysis, our experience with the technique goes back over two decades. The performance patterns are fairly consistent. Yet, we have to admit that there were exceptions that proved to be quite dramatic. Therefore, it is important also to utilize other available technical and fundamental tools to arrive at a high probability of success in forecasting prices. The nature of the events that shape price trends of futures contracts should keep even the most proficient of technical and fundamental analysts on their guard and flexible at all times. International developments, weather, and politically-motivated legislation are among the unpredictable forces that can change the direction of the markets in an instant. There is no master key that can unlock all the doors to successful price forecasting. Nevertheless, we believe that the proper interpretation of the "Commitments of Traders" reports is valuable and belongs on the analyst's key ring.
Copyright © 1934 - 2018 by Commodity Research Bureau - CRB. All Rights Reserved.
209 W. Jackson Blvd. Suite 200 Chicago, Illinois 60606-6940 USA
Phone: 800.621.5271 or 312.554.8456 Fax: 312.939.4135 Email: firstname.lastname@example.org
|Press Ctrl+D to bookmark this page - Set http://www.crbtrader.com as your Home Page|