The one-minute plunge in the price of gold on Friday at 6:43 a.m. MST seems to be getting a lot of attention from traders and the financial media. After drifting mostly higher overnight, the price of gold, as set by the Comex futures floor trading, suddenly dropped $28 dollars in the space of three minutes, with most of the drop occurring in the first minute. Not only did it blind-side everyone who trades gold and silver, but there lacked any fundamental news triggers which would have set off the dramatic drop in price. The question is, what happened, does it make sense, and can we profit from it if the drop in price will be reversed?
Long-time Comex futures traders and observers have been aware that from time to time, and it seems like more frequently in the last year, the Comex is susceptible to "bear" raids once the floor opens for trading (6:20 a.m MST for gold and 6:25 a.m. MST for silver). Even more pronounced is that a preponderance of these raids occur on Friday - when the rest of the world is at rest for the weekend - either right at the Comex open or shortly thereafter. Without introducing market manipulation theories into the discussion, I want to outline what happened, why it does not make sense that a plunge like this occurred and how to take advantage of a market "inefficiency" like Friday's.
The graph below shows visually what happened (contract time and volume data sourced from LemetropoleCafe.com):
As you can see, the waterfall plunge started at 6:43 a.m. MST with close to 8,000 contracts trading in the space of 60 seconds. Within a three minute time-frame, a total of 16,903 gold contracts traded. That is, 8.5% of the Friday's total volume in Comex gold traded in the space of three minutes. Inclusive of the post-floor electronic trading of Comex futures (Globex), total volume on Friday was a little over 196,000 contracts. Three minutes is less than 1% of the amount of time that the Comex + Globex is open. So 8.5% of Friday's volume occurred in less than 1% of the time the Comex is open. As I scoured all the news sources I use, I could not find any news released that would have triggered panic selling like this.
What does not make sense about this volume and price action is that the methodology of unloading a long position of this size runs counterintuitive to that which a rational, proceeds-maximizing seller would use when trying to unload a large, unwanted position. As a junk-bond trader, I learned early on that if you have a large position that you need to sell, you try to do it "gently" over the course of a day or a couple of days in order to prevent "disrupting" the market and to attempt to maximize your selling proceeds. You certainly would not dump a big position on the market out of the blue unless you were trying to disrupt the market. There's no question that Friday's gold contract dump created a sense of "shock and awe."
The second interesting aspect to Friday's market plunge is that the amount of paper gold sold in the space of three minutes far exceeded the amount of physical gold available for delivery on the Comex. Here's Friday's gold warehouse stock report from the CME which shows the number of ounces reportedly contained in Comex depositories:
Comex metals inventory is segmented into "registered" and "eligible." The "registered" gold is gold that has been designated by its owners as available for delivery. It also has been tested and certified by the Comex to meet the delivery standards required by Comex futures contracts. The "eligible" inventory is the metal that is kept safe in Comex depositories by the above-listed Comex custodians on behalf of the entities that own the gold and is not available for delivery.
I have marked the amount of gold that was available for delivery on the Comex as of Friday with a red box. 721,463 ounces of gold was currently made available for delivery. But the number of contracts that traded in the space of three minutes represented 1.69 million ounces of gold (each contract is 100 ounces.). In other words, someone sold the paper-equivalent more than twice the actual available amount of physical gold on the Comex in the space of three minutes. In theory, if one entity bought those contracts and demanded delivery at the end of the month (I'm assuming the buyer would swap his position into October contracts and stand for delivery), the Comex would stand in default.
More to the point along these lines, the December "front month" gold contract - as of Friday's close (pdf link) - has 230,456 of open contract interest. This represents over 23 million ounces of gold. As you can see from the chart above, the Comex in total only has 6.9 million ounces of gold. In theory, if every holder of those contracts decides to stand for delivery, the Comex is out of business. I'll point out that no other CME contract product has - or ever has had - this size of imbalance between the paper contracts outstanding and the deliverable underlying commodity upon which those contracts have a legal claim. The point here is that the imbalance between open paper claims on Comex gold and the amount of available gold to delivery is completely out of whack with rational market behavior and represents an extreme market inefficiency.
With a glaring imbalance between the amount of paper gold outstanding on the Comex - as represented by the open interest in futures contracts - it will only be a matter of time before the banks and funds who are short Comex gold futures will be overrun with delivery demands. Currently, December's open interest in gold futures is 33 times greater than the amount of gold available to be delivered. Historically this has never been a problem for the Comex. Historically, maybe 1-2% of the open interest in any delivery month stands for delivery. But the total stock of gold on the Comex has dropped from 11.7 million ounces in mid-December 2012 to 6.9 million ounces as of Friday. This is an astounding 41% drain of physical gold from Comex inventories. In the case of December, if just 3.5% of the contract longs decide to hold for delivery, the Comex will default.
Comex contracts do have a "force majeure" clause in them which enables the Comex to cash-settle its contracts. If you are willing to take a longer-term view of this situation and make a default/force majeure play, you can get long Comex gold contracts on days like Friday after a bear raid and keep rolling forward your position. If the Comex ever does face default/force majeure, the price of gold on the Comex will experience an incredible short squeeze up. If you make this trading play, you have to understand that you probably won't be able to take delivery of gold and will have to settle for cash instead. Given the enormous demand for physical gold from Asia and the BRICs, the day that a run on the actual physical gold at the Comex is a lot closer than it was when the precious metals bull started thirteen years ago.
But there are also ways to take advantage of days like Friday using short-term strategies. For instance, recognizing that Fridays have at least an 80% probability of being a down day on the Comex, we set up Comex gold contract short positions in our fund on Thursday evening and covered them after the hit on Friday. Doing this 'naked' is risky, but we are long plenty of physical gold in our fund in case the trading strategy backfires.
There are two other ways to profit from a market disruption event in the gold market like Friday's. First, I always advise people to accumulate physical gold over time as a hedge against the dollar devaluation policies of the Federal Reserve and our Government. Days like Friday are the perfect day to establish, or add to, your physical gold holdings. The second strategy is a short term-oriented trading strategy involving the VelocityShares 3x long gold EFT (UGLD). If you had bought UGLD on Friday within roughly 90 minutes after the NYSE opened, you would have had an average cost of entry of $15.30. If you sold your long position within the first 90 minutes of trading today (Monday) you would have made seventy cents on the trade. That's a 4.6% return in one day. I consider that a low-risk proposition given my view that what happened on Friday was not based on any real fundamental or technical events and that the next day would likely yield some form of "snap-back." In fact, up until the Comex opened today, the price of gold had clawed back all of Friday's price hit.
There are some other lower risk trading vehicles, like GLD or options on GLD, which can be used to take advantage of the Comex gold market on days like Friday but I have outlined the trading and investing strategies we use. There is no question in my mind that days like Friday represent an extreme market inefficiency with the Comex gold market that can be profitably traded - at least until enough traders in the market adapt to this inefficiency.
Additional disclosure: The fund I co-manage is long physical gold and actively trades UGLD and GLD.