November 30, 2004
Battered Commercials Still At Risk of Being Deeply Fried
By Brady Willett

The holiday delayed COT numbers were released yesterday.

Historically – at least during the current gold bull market – small speculators increase their long position and the commercials add to their shorts as the price of gold rallies. Well, for the week ended November 23, 2004 small speculators reduced their net long position by 6,354 contracts while the commercials reduced their net short position by nearly 6,000 contracts.  Given that gold rallied strongly on the week the COT data was exactly the opposite of what was expected.

Making Some Statements

‘As gold rallies strongly small speculators lock in profits’ is a statement that has never been made before -- until now. Moreover, and after much adieu, the statement ‘commercial short covering helps send gold higher’ can, finally, leave the realm of pipe dreams.  To be sure, after bleeding badly for weeks the commercials actually covered some of their short contracts for losses last week*.  It’s a mad, mad world.

Stranger still are the two extraneous events that impacted the COT statistics. These events helped drag open interest down by a startling 76,377 contracts.

The Important Moves of the Tiny

To illustrate how important small spec interest is in the marketplace consider the following: in 2004 the price of gold has fallen on a weekly basis by more than 3% only twice (weeks ended Jan 20 and May 11), and on each occasion the net long small spec position has dropped by 10+%.  Such is why last weeks data is perplexing: gold rallies strongly and the small specs run away with profits?

Prior to last week the last time small spec net long interest dropped by more than 10% while gold rallied was on May 27, 2003. The price of gold plunged on May 28.

Date Spot Gold Weekly Change in SS Weekly Change in  ($ ounce) net long position Spot Gold 11/23/2004 448.15 -13.30% 1.99% 5/11/2004 439.40 -10.16% -4.09% 5/4/2004 391.25 -19.59% -1.26% 1/20/2004 409.25 -10.64% -3.82% 10/7/2003 376.10 -13.24% -3.07% 5/27/2003 371.40 -16.69% 1.39% 3/25/2003 331.90 -20.56% -2.04% 3/18/2003 338.80 -12.52% -2.84% 2/11/2003 364.00 -25.23% -3.33% 10/15/2002 316.20 -16.68% -0.88% 10/1/2002 321.75 -11.74% -1.39% 7/30/2002 305.40 -20.66% -4.47% 6/11/2002 316.80 -27.69% -3.00% 4/9/2002 298.70 -10.88% -1.74% 1/29/2002 277.90 -21.02% -1.47% 12/11/2001 272.20 -29.28% -1.59% 11/20/2001 272.95 -33.90% -1.28% 11/13/2001 276.50 -17.16% -0.88% 10/23/2001 276.05 -23.32% -1.80% 10/16/2001 281.10 -13.84% -3.07% 7/31/2001 265.90 -36.32% -1.06% 7/10/2001 267.45 -12.39% -0.39% 7/3/2001 268.50 -26.89% -2.01% 6/5/2001 265.70 -29.43% -3.64% 5/29/2001 275.75 -10.70% -2.96%

Although analyzing commercial or tech fund influence in the marketplace is a more popular pastime than focusing on the small speculators, the fact is that recent moves by the small specs have the closest correlation to changes in the price of gold. For example, in 2004 changes in small spec interest have coincided with the to be expected change in the price of gold 61.7% of the time (as speculators add to their net long position the price of gold can be expected to rise and when the small specs reduce their net long position gold is expected to fall).  The comparable statistic for the commercials is only 52%.

Before belaboring the point consider the following chart and ask yourself one question: Are things really different this time?



Tainted Data?

Ordinarily if the small specs were reducing their net long position as gold rallies this would be incredibly great news for gold bulls.  Why?  Because during the current bull market when the small specs own – and continue to accumulate – an extreme net long position this has always presaged a serious price correction in gold.


Is it difficult to believe that the small specs are really locking in gold profits while the commercials begin frantically covering for losses? No. It is impossible.  What seems closer to the truth is that some small specs have moved capital out of the futures market and into financial instruments such as GLD.  If this is the case, the small spec position - while untrackable - could still have a weakening effect if the price of gold corrects (and small specs dump GLD). Incidentally, given that the highly anticipated launch of the first gold backed ETF in America attracted more than $1.3 billion in capital it its first three days of trading (Trimtabs), it is also likely that GLD has recently sapped some liquidity out of gold/silver equities (which have a combined market cap of around $137 billion).

If the launch of GLD was not enough to skew the COT numbers consider also that the CBOT celebrated its successful launch of electronically traded gold/silver contracts by waiving fees for the holidays (& PFG followed suit). The fees began being waved on November 22, or only 1-day before the latest COT numbers were tabulated.

The speculations are endless with regards to the exact impact GLD and the CBOT had on the COT numbers. Regardless, the impact was most noticeable in open interest. When open interest declines by 50+ thousand contracts it usually means that the price of gold is in free fall.  During the last week (ended Nov 23) spot gold rallied by $8.75 an ounce and open interest plummeted.



Outlook Remains the Same

Limitations of the COT data notwithstanding, the key point [again] is that the commercials remain extremely short. To be sure, even though net short commercial interest dropped from 217,563 to 211,606 contracts last week, because of the dramatic drop in open interest net short commercial interest as percentage of open interest actually increased from 38.6% to 43.5%.  The gold bull market is riddled with price corrections that began when the commercials were so extremely short.

Little Bobby is staying home from school today because he wants to guard his long Yen position against possible intervention.

Longer term the case for gold remains intact.  However, it is worth remembering that if U.S. dollar short/Euro long positions in the market prove overextended that the price of gold will correct.  As for the alternative to a ‘dollar rebound’ in the near term – which is a ‘dollar crisis’ - anticipation of such an event has grown so widespread that school children across America are betting their lunch money on whether or not the ECB will raise interest rates later this week (3-1 says they don’t).  Kidding aside, the story behind the dollar/gold relationship going forward is still the same: a dollar crisis or a gold bust.  Yesterday was the first day since the post-election decline in the dollar started that a slumping dollar was specifically blamed by many analysts for weakness in stocks and bonds.

To make a long recap of the COT internals short, my gut says that gold will top $500 before year end.  However, my gut is incapable of reading the COT numbers which continue to scream ‘correction coming!’


*  The commercials have reduced their total net short position on many occasions as the price of gold rallies. However, whether or not the commercials covered their positions for losses is often vague because of intraweek price changes in gold. For the week ended November 23 (last COT report) gold never traded below its previous COT close. As such, it can be said without doubt that ‘commercial short covering helps send gold higher’.