January 2011 brought about a severe correction in the gold market. After the steep rally that ended on a multiple top and a new all time high in USD during December 2010, the new year started with a gradual slide, acerbating on news of a small hedge fund having overplayed its hand on a long/short calendar spread in gold futures: traders typically need a lead to follow the gold market lower. The greenback has been weakening since the beginning of the year, putting aside the inverse correlation between the gold price and the USD exchange rate. Main stream press also pointed to significant drops in the holdings of GLD, the major gold backed ETF. So, is this gold price correction about to end soon?
In a posting last Saturday, FOFOA commented on the lack of logic behind the interpretation of the change in gold holding by GLD.
Physical gold ETFs are designed to track the gold price, not to mirror it. As such the GLD ETF shares trade at a small premium or discount as compared to the inventory value.
Whenever demand for GLD shares is high, the custodian will create new shares and sell them at a premium relative to gold. Revenues are used to buy physical gold, to bring inventory in line with the increased market capitalization of the ETF. During a gold correction, offer exceeds demand and the custodian will put up bids at a discount to the inventory value of the ETF. The GLD custodians need limiting the cash overdraft and will sell physical gold to a few authorized participants (a limited number of brokers and banks), who may proceed in selling the gold. So any drawdown of inventory of the ETF is likely to be compensated by an increase of the holding of physical gold by some large investor. (The minimum quantity is a ‘basket’ of 100,000 GLD shares, the equivalent of 10,000 oz of Gold.) The whole procedure is a non-market operation.
Most generally paper gold investors follow the trend and are buying into rallies, selling into corrections… and are left holding the bag. Here, I wish to include a quote of Saturday’s posting by FOFOA, where he references “The GLD puke indicator” as found in a newsletter by Lance Lewis.
Now here's a bloodhound that might be on to a scent worth following. Lance Lewis, in his subscriber newsletter, follows what he calls "the GLD puke indicator" which tracks GLD physical gold regurgitations [emphasis mine]:
Just in case anyone missed it in last night’s letter, our GLD puke indicator that has nearly a flawless record at marking lows in gold triggered a buy signal yesterday after the ETF spit up 31 tonnes (and some blood) to trigger a 2.48% decline in its bullion holdings.
As we’ve noted before, one-day declines in the holdings of this ETF of over 1% have tended to be capitulatory in nature and have typically occurred near important lows in the gold price during gold’s secular bull market.
GLD inventory drawdowns and gold price corrections aligned
Consider that since the GLD ETF’s creation back in 2004, it has seen 1%+ one-day declines in its bullion holdings only 41 other times. When one goes back and looks at where these declines in bullion holdings have occurred, virtually all of them occurred “at” or were “clustered at” important lows in the gold price.
When we update this familiar (see above) chart for today’s 1%+ decline in bullion holdings, we can once again see where I have labeled the past eleven 1%+ declines in the ETF’s bullion holdings (plus today’s decline) with red dots and then placed a corresponding white dot below the price of GLD in order to show where that decline (or clusters of declines, as was the case in 2008) occurred relative to the price of the GLD, which is obviously tied to spot gold.
You will recall that we most recently used this indicator back on July 28th, 2010 in order to identify what was then the summer low in the gold price, and we used it again on October 7th, 2010 to recognize that a sudden 1 percent slide in gold from an all-time high was actually a just a one-day setback that led to new all-time highs being hit once again just a few days later.
The pattern you see emerge after today’s 1%+ puke, just as on those prior occasions, is that these “pukes” of bullion by the GLD ETF have always tended to occur at or very close to important lows in the gold price, and declines of over 2% have only occurred at MAJOR lows, such as the two major lows that were hit in 2008.
Note that one of those lows on September 9, 2008, which is the closest in size to today’s puke, also occurred just one day before a 5-day short squeeze/meltup of 30 percent in the gold price that kicked off on September 12, 2008. Perhaps the remaining shorts in the gold market will now pay a similar price for betting against a bull market?
Perhaps history will repeat and perhaps it won’t with respect to such a short squeeze, but given this indicator’s near flawless record at marking lows in gold, it's not to be ignored.As of now, the riots in