Saturday, March 20, 2010

Pulling Back Curtains

(click on image to enlarge)

Here is some important information:

TRYING TO MAKE SENSE OF WHAT’S HAPPENING ON THE CME (AGAIN): (By J. Kaczor) In addition to its many other services and programs, the Chicago Mercantile Exchange (CME) provides a daily spot market for cheddar cheese, butter, and nonfat dry milk for any person or company who wishes to buy or sell a carload of these products on short notice. The prices that result from these trades (or offers or bids) are used directly or indirectly in formulas that determine levels of milk prices throughout the country. (Basically, $.01 per lb of cheese equals $.10 per cwt of Federal Order Class III milk (or Class 4b for California.)) The CME also offers a market for monthly Class III milk futures contracts which reflect a collective opinion on where Federal Order Class III milk prices could be over the next twenty-four months. Futures contracts are mostly used for hedging purposes, and are not used in a direct way to set milk prices. A carload of cheese is specified to be 40 to 44 thousand lbs; two Class III futures contracts represent 400,000 lbs of milk, enough to produce 40,000 lbs of cheese. In the sense that changes in the price of cheese affects opinions of future milk prices, the two markets are related.

In 2007, the U.S. Government Accountability Office (GAO) was asked by a number of U.S. Senators to look into complaints about possible manipulation of the CME spot cheese prices. The main reason for those complaints was the fact that cheese prices on the CME spot market are used by almost all participants in the cheese industry to determine their selling prices, but is a “thin” market, susceptible to manipulation. Typically, less than one percent of the total amount of cheese produced in the U.S. is traded on the CME, and only a small number of traders participate. The GAO completed its study and concluded that the oversight provided by CME and the Commodity Futures Trading Commission (CFTC) had improved to the point where it was believed they were capable of detecting and deterring manipulation. The CFTC is authorized by sections 6(c) and 6(d) of the Commodity Exchange Act to investigate actual or attempted price manipulation and to prosecute those who are found to have done so.

Because of the relationship between the CME cash market for cheese and the Class III futures market (“The price paid for a Class III milk futures contract is based on the monthly Class III milk price released by USDA, which is heavily influenced by the price of cheese at the CME spot cheese market.” GAO Report, page 14), CME’s market surveillance analysts look for activity on the cash market that may not reflect true market conditions and which would benefit a trader’s futures contracts. The example given by GAO was “trading activity on the CME spot cheese market that might directly benefit a trader’s futures positions, such as selling cheese in order to lower prices and benefit a short position in Class III milk futures.” The possibility of that happening recently came to mind, based upon a distinctive sequence of activity reported for a single trader on the CME over a period of twenty-one weeks.

The period referred to runs from late October 2009 through the present. (Trading data from earlier weeks was not available for this article.) The sequence of trading for this company was from that of a heavy buyer to a heavy seller. The chart on the right shows the daily prices for blocks and barrels on the CME from September 1, 2009, through March 18, 2010, and the daily trading activity reported for a Division of Davisco International, Inc. Davisco operates three cheese manufacturing plants, the largest of which is located in Jerome, Idaho. Together, these three plants produce an estimated 750,000 lbs of cheddar cheese daily.

Block cheddar prices had risen from $1.0875 per lb on July 14, 2009, to $1.27 per lb on September 4th where it leveled out for a time, and then proceeded to rise fairly steadily to a peak of $1.72 per lb on December 2nd, and then leveled out at $1.70 per lb for the next three weeks. Following is a recap of trading activity reported for Davisco from October 29th through December 22nd.

• No trades were reported in October for Davisco until the 29th, at which time 37 buys or bids were reported on their behalf over twelve trading days, which represented 44% of all such activity during that period. The block price rose $.07 per lb during that period.

• Following a two week period of inactivity, 70 buys or bids were reported for Davisco over seventeen trading days, which represented 61% of all such activity during that period. During this period, the block price reached a high of $1.72 and held no lower than $1.70 per lb.

The block price began to fall on December 23rd. No trading activity was reported for Davisco from December 23rd through January 5th. Following is a recap of trading activity reported on their behalf from January 6th to March 18th:

• Davisco became a seller, selling cheese or offering prices which lowered the daily price at the end of the day’s trading. A total of 39 sales were reported for Davisco over sixteen trading days, which represented 48% of all such activity during that period. The block price initially fell, but ultimately rose by $.075 per lb during that period.

• Following a two week period of inactivity, a total of 165 sales and 9 offers to sell were reported for Davisco over twenty-three trading days, representing 93% of all such activity during that period. The block price fell by $.2475 during this period.

Those 165 carloads sold during this 31-day period totaled 6.6 million lbs of cheddar cheese. According to information provided on Davisco’s website, that amounts to about one quarter of all the cheddar believed to have been produced in their three plants during that time. That’s consistent with a report that buyers of cheese on the CME are saying the cheese they’ve gotten last week is “quite fresh, with most of the ages ranging from 5 to 10 days old.” Last week Davisco was reported to have sold 56 of the 60 carloads of cheddar that were traded.

A broker commented on Wednesday’s trading: “Even with increased interest from eight different traders on the CME spot market yesterday, a lone seller was able to keep prices unchanged with 5 blocks and 11 barrels exchanging hands. Class III milk futures responded in kind by removing additional premium from the market…” So, what’s going on? It looks like Davisco’s buys and bids in December provided significant support for cheese prices at that time, and Davisco’s sales and offers to sell since the middle of February have had significant affect on moving cheese prices lower.

So what? Some who have commented on recent CME cheese price movements point out a number of overreactions for block prices in the past fifteen months: a low of $1.04 per lb in January 2009, a sharp rise to $1.33 a month later, a drop to $1.0875 in July, the rise to $1.72 in December, and finally back to about where the prices were last September. Prices have been volatile. They result from bids and offers made in an open exchange monitored by interested participants. As mentioned earlier, the job of determining if trades on the CME reflect true market conditions (supply, demand, extraordinary conditions, expectations, etc.) rests with CME market surveillance analysts. According to what GAO found, that determination includes possible direct benefit to traders’ futures positions, which could imply manipulation. GAO explains manipulation this way: “Generally, price manipulation is any planned operation, transaction, or practice that intends to and causes or maintains an artificial price – that is, a price that is higher or lower than it would have been if it had reflected the forces of supply and demand.”

The relationship between CME spot cheese prices and Class III futures prices is clear but not precise. When cheese prices rise, prices for Class III futures rise; when cheese prices fall, prices for Class III futures fall. The chart of Futures Contracts (on the right) shows the daily changes in the values of the March, April, and May Class III milk futures contracts on the CME from September 2009 through this March. As you can see when comparing this chart to the chart earlier in the article, as the cash price for block and barrel cheese on the CME trended up and down, the value of futures contracts throughout 2010 also trended up and down.

The vast majority of futures contracts are purchased by hedgers who hold them until the final price for the month is determined, at which time they close their positions (long or short) with offsetting sales or purchases. Another very important segment of futures activity is that of speculators who provide essential market liquidity by accepting the risk of losses by selling and buying contracts to and from anyone for any month at whatever price then prevails.

The rise and fall of cheese prices along with the rise and fall of futures prices shows the possible amount of risk and reward for speculation. Has the example given by GAO of what CME and CFTC analysts look for as evidence of price manipulation occurred anytime in the past six months: selling cheese in order to lower prices and benefit a short position in Class III milk futures – or the reverse, buying cheese in order to raise prices and benefit a long position in Class III milk futures? That’s a question worth considering.


  1. Where else or why other than market manipulation would someone try to sell for a lower price what they produce. It doesn't take a Harvard degree to figure this one out. Milk processor 101: sell a small amount of what we produce, buy all our imputs cheaper! Only thing they are forgetting is "they" are starving the goose laying thier golden egg!

  2. It also seems odd that the dairy trading pits cannot attract more speculators and non-processor traders that would help provide the needed liquidity. It seem obvious that they wont trade on fundamentals when they know that the processors are going to trade on inside info or inventory pricing needs instead of fundamentals. I quit hedging when markets went against fundamental reason, and I decided the markets were being fixed. Last year I found out DFA was convicted of doing just that, and paid a big fine.

  3. "Typically less than 1% of cheese produced in the US is traded on the CME". Yet in one trading cycle mentioned in this article one trader generated 93% of the sales. So milk prices are predicated on Davisco's whims and this is not suspicious? Investigations take place when there is a hint of impropriety? Uh hint hint hint! 6 months of suspicious activity should be enough (sic) to make anyone with a brain take notice, but as usual here we sit with no oversight and business as usual.