Weekly Review
After four consecutive weekly price declines, gold and silver finished higher this week, as a result of an explosive Friday morning rally tripped off by the monthly employment report. Gold finished $31 (2.6%) higher for the week, while silver ended 19 cents (1.2%) higher. Given gold's relative outperformance, the silver/gold price ratio widened out a full point to nearly 76 to 1.
The standout price feature was the spike up in all precious metals and corresponding spike down in the US dollar as the report was released and the high volume flat line price movement throughout the remainder of the day. This is a price pattern increasingly common over the years and occurs in both directions, up and down. When the price spike is down, reports circulate that the COMEX commercials who deploy HFT (High Frequency Trading) are selling massive amounts of contracts to bomb gold and silver. When the price spike is to the upside, there are not nearly as many reports blaming the COMEX commercials.
The truth is that the price spikes in either direction are always the result of HFT trading and massive futures contract positioning changes on the COMEX, but rarely do the reports on these occurrences get the whole story straight. On sudden price plunges, it is always the managed money technical funds doing the selling, with the commercials doing all or most of the buying. Sure, the commercials may get the snow ball rolling down the hill by selectively and strategically selling some small quantities of contracts to induce the technical funds to sell in much greater quantities. But on big high volume price plunges, by the end of the day's trading, it is always the technical funds which have sold on balance and the commercials which have bought. Always, as proven out in the COT report.
The reverse is true on days like yesterday, when prices suddenly soar in the space of minutes and then flat line in high volume the rest of the day. You can be sure that the managed money technical funds bought massively yesterday and the commercials sold just as massively. I hate to rely on the preliminary open interest report for yesterday's trading, because such reports are subject to sharp revision; but I have little choice since the final open interest numbers won't be published until Monday morning. Based on yesterday's price action and trading volume, I would estimate an increase in technical fund buying and commercial selling of 20,000 to 30,000 net contracts in COMEX gold futures and, at least the preliminary open interest would seem to confirm that, as total open interest in gold reportedly increased by nearly 19,000 contracts.
http://www.cmegroup.com/daily_bulletin/current/Section62_Metals_Futures_Products.pdf
Bottom line, the surge in gold prices yesterday was almost exclusively the result of technical fund buying and commercial selling on the COMEX and, at least for the day, it was a case of both the technical funds buying high and the commercials selling high or at least much higher than the previous days. This is the price and market trading pattern so clearly delineated in the market structure analysis as indicated in the COT report.
Higher prices from here will, undoubtedly, feature more technical fund buying and commercial selling; lower prices, the opposite. The day will likely come when this isn't what drives gold and silver prices and it will be a day recognizable by almost all who follow COT and physical market developments. I don't know how close that day may be, but there have been some truly extraordinary developments in gold, namely, those concerning the current COMEX June gold delivery process that bear mention.
More gold has been issued and stopped for delivery in the first five days of the June contract than ever in my memory. In addition, more than 400,000 oz of gold have been deposited into the big gold ETF, GLD, over the past three days and I don't believe that includes metal owed as a result of yesterday's trading. Given that gold prices were flat to lower for weeks until yesterday, the surge in metal deposits into GLD, coming on top of the 2 million ounces deposited during May (also on declining prices) is extraordinary (get used to me overusing that word). It still looks to me like a big gold buying moose is on the loose in GLD, but let me focus on the COMEX June gold deliveries. But remember, both in GLD and in the June COMEX deliveries, we are talking of cold, hard physical metal, not paper derivatives contracts.
In fact, the most extraordinary development of all is that the physical quantities of gold changing hands in GLD and in the COMEX June deliveries haven't had any observable effect on the price of gold; given that at least until yesterday, gold prices hadn't moved higher despite the physical developments. And if yesterday's price surge wasn't related to the employment report as a cover story to rig prices higher through HFT computer algos and not what I am about to discuss, then it's time for me to get my eyes examined.
Because the COMEX June gold deliveries have been so unusually large, I will endeavor to report on them in the most objective manner possible. In the first five days of delivery on the June contract (there are still weeks of deliveries to go), a total of 11,765 contracts have been issued (delivered) and stopped (accepted), equaling 1.1765 million oz of gold (100 oz in a contract). In dollar terms, this comes to more than $2 billion. To keep things in perspective, there are more than 8.6 million total gold ounces in the COMEX warehouses.
While the number of gold deliveries issued and stopped so far is the highest ever (I believe), it is also somewhat overstated in that when a delivery is made, if that delivery is immediately redelivered, that redelivery is counted as new delivery. It's possible that as many as two or three thousand contracts may have been redelivered, thus reducing the total real amount somewhat. I know the source data can be complicated, but if you scroll down and look under gold deliveries for June, you might get a sense of what I'm talking about. But I'll try to summarize things.
http://www.cmegroup.com/delivery_reports/MetalsIssuesAndStopsYTDReport.pdf
There were an unusually large number of clearing firms who, on behalf of clients, took delivery of gold and then redelivered this month, more than I have ever seen. I'm just speculating, but having done this on behalf of clients many years ago, I think what I'm about to say to be informed speculation. Because there was such an unusually large number of open contracts going into first delivery day for the June contract, a good number of smaller market participants may have sensed a developing tightness for delivery and decided to stand for delivery themselves as a way of testing if those holding short contracts had the metal to deliver. This is how markets work. Up until this point, the shorts appear to have had the metal (or were able to get it) and as those testing the shorts got metal on delivery, the test was completed and redelivery was made. That was just one unusual feature to the June delivery process.
As I indicated last week, the biggest participants in COMEX gold deliveries are usually JPMorgan, HSBC and Bank of Nova Scotia, all in their own proprietary or house trading accounts (although JPMorgan, as the largest dealer in gold and silver, also does a large customer business). The June delivery process to date has reconfirmed the three largest participants are as I indicated. So far, JPMorgan has taken delivery on 5173 gold contracts or 44% of the total 11,765 total contracts issued and a much larger percentage when redeliveries are netted out from total deliveries. On the issue side, HSBC and Bank of Nova Scotia have issued a combined 6110 total gold deliveries. It's not hard to see who the big dogs are on the COMEX.
There is a 3000 contract position limit in the spot moth of gold on the COMEX and I've seen a single trader issue more than that in the past, as HSBC has done this month in issuing 3586 contracts so far. It doesn't appear to be a problem to me because if someone has the physical goods to deliver, I can understand that issuer being allowed to deliver more than the supposed position limit (as long as it's not disruptive to price).
On the other hand, I do have a bit of trouble understanding why any buyer or stopper of gold, in this case JPMorgan, would be allowed to buy much more than 3000 contracts. I mean, what is the legitimate business necessity for anyone to buy more than the position limit in any one month? I'm sure JPMorgan did receive a hedge exemption from the spot month gold position limit, I just don't know why. I did write to the director of Market Oversight at the CFTC about this earlier this week, but received no response. (Nor do I receive Christmas cards any more from the agency – or ever, for that matter). There are still more than 3000 contracts still open in the June delivery month (adjusting for Monday's notices), so more gold deliveries lie ahead. Additionally, more new June contracts have been purchased over the past week, highly unusual in its own right, indicating new physical demands for gold.
If you had told me a couple of weeks ago that 11,765 total gold contracts would be issued in the first week of delivery, with more than 3000 contracts to go and that 400,000 oz of gold would be deposited into GLD this week, to say nothing about JPM stopping almost twice the spot month position limit, I would have surmised that this would have caused price fireworks. We did see some price action yesterday, but as I said, it certainly didn't look related to the June deliveries or the deposits into GLD.
As much as I would have been surprised, the craziest thing is that more than ever, this confirms that futures contract positioning on the COMEX exerts a greater influence on price than anything else; physical market developments included. What makes it crazy is that this is a textbook example of artificial price manipulation a case of the futures market tail wagging the physical market dog. This is as far as you can get from the intent of commodity law.
Sticking with my attempt to be objective, up until now there is no outward sign of the extraordinarily large gold deliveries reflected in the price. Yes, we did jump yesterday, but if that jump was related to deliveries, then I missed it, because all the metals jumped in price. The real telltale price sign of a delivery squeeze is in the spread differentials involving the June contract, something I monitor closely. The spread differences between the June contract and the key rollover month, August, have been remarkably stable and, if anything, reflect more ample supply than tightness. That can change in a heartbeat and if it does, it will be noticed and acted upon by the same market participants that tested the shorts as described above. The sharks on the COMEX can smell the blood of someone in trouble as well as the ocean variety. I don't know how the heck they are doing it, but the issuers of physical gold have been able to manage this delivery with no apparent stress to date.
Somewhat counterintuitively, the very heavy stopping by JPMorgan might also suggest the gold market is not about to explode in the very near future. After all, how hard would it be for even the dimmest of market regulators to connect the dots between the extraordinarily large deliveries taken by JPM, in its own trading account and an immediate surge in the price of gold? And to those who suggest that JPMorgan may be stopping all this gold, like it has done in silver, on behalf of a client (the favorite being China), that offers no great disagreement with my basic premise; it just involves a slightly different motive. And if I were an outspoken political candidate, I would imply a dose of treason in JPM's actions, if it were acting on behalf of a foreigner.
I also can't help from comparing what JPMorgan is doing in gold to what it has done in silver. This week, JPMorgan has stopped almost 520,000 oz of gold, worth nearly $650 million. By coincidence, that dollar amount comes to roughly 40 million oz of silver or close to the amount JPMorgan has taken in COMEX silver deliveries over the past 15 months. In other words, it took JPMorgan 15 months to buy as much silver, in dollar terms, as it took in delivery of gold this past week alone. The real reason JPMorgan never bought more than the 1500 contract spot limit in COMEX silver was not because it couldn't evade the position limit legally (as it did in gold), but because there wasn't enough physical silver available.
Since I've been so wordy this far, let me mention briefly that the physical turnover or movement into and out from the COMEX-approved silver warehouses in the four day workweek was just over 3 million oz and total inventories rose 0.4 million oz to 154 million oz. There was a big switch of 3.5 million oz from registered to eligible, but until I think that really matters, I'll just leave it at that. COMEX gold warehouse inventory movements and category changes were much more interesting, but basically covered in the delivery discussion above. There was a 1.5 million oz deposit in the big silver ETF, SLV, but compared to the dollar amount of gold deposited into GLD, the deposit in SLV doesn't look terribly significant.
Onto the changes in this week's Commitments of Traders (COT) Report, where I missed on my predictions by an embarrassingly wide margin, despite trying to qualify those predictions as craftily as possible. Where I predicted a 40,000 contract reduction in the gold headline number of the total commercial net short position, the report indicated a reduction of 11,200 contracts. And where I guessed a 5000 to 10,000 contract headline reduction in silver, there was only a 2300 contract reduction (although I did come closer on the managed money category in silver).
As I hope you know, I don't usually miss by that much and as much as it doesn't matter much what my predictions turn out to be, it is possible that the accounting for this report may have been the culprit since the Monday holiday featured very heavy trading and price selloffs which were carried over and combined into the Tuesday cutoff data. I have mentioned in the past that sometimes heavy trading into the Tuesday cutoff doesn't get completely picked up timewise in the report. This week, the Monday government holiday might have complicated things further. Please don't think I'm making excuses, but if all the data wasn't reported on a timely basis and is carried over to the next report, that makes predictions for the next report even chancier. In other words, yesterday's heavy technical fund buying and commercial selling might get obscured in the next COT report.
In COMEX gold futures, the total commercial net short position was reduced by 11,200 contracts to 214,000 contracts. The prior COT report featured a greater reduction than expected, this one a much smaller reduction. Over the two weeks, some 76,000 commercial contracts were bought and around 63,000 managed money contracts were sold on as much as a $75 decline in the price of gold, which included a decisive penetration of the 50 day moving average and a slight penetration of the 100 day moving average. The 200 day moving average was never seriously threatened by a downward penetration. Since I had a minimum expectation of 100,000 to 150,000 contracts of commercial buying to occur before we could even think of a price bottom in COT terms, 76,000 contracts wouldn't appear to satisfy my expectations. Then again, there's nothing magical or sacrosanct about my expectations, as my COT predictions just reflected.
By commercial category, the big 4 actually added nearly 5100 contracts of new shorts, while the raptors bought 8500 contracts and swung to an expected net long position, by a small 1500 contracts. The big 5 thru 8 rounded things off by buying back 7800 short contracts. As off as I was in my overall expectations, I never get encouraged when the big 4 add to shorts in either gold or silver.
On the sell side in gold (apart from the big 4, the managed money traders sold less than a net 8000 contracts, including the sale of 13,020 long contracts and the buyback of 5170 short contracts (which was the real puzzler). Based upon previous expectations it's hard to call the gold report as anything but disappointing in that more technical fund contracts weren't sold. And with yesterday's new technical fund buying and commercial selling, in COT terms, the gold market structure isn't as good as I hoped. The June delivery process has me on the edge of my seat, the COT structure decidedly less so. That doesn't mean we can't rally further in gold or have yesterday's rally turn into something very serious, as suggested by the performance of the mining shares, but I'm reporting on COT considerations here.
In COMEX silver futures, the commercials reduced their total net short position by 2300 contracts, to 76,000 contracts total. I was hoping we would be down by at least 20,000 total commercial net short contracts over the past two weeks, but the reduction came to less than 14,000 contracts. By commercial category, the raptors (the smaller commercials apart from the eight biggest traders) accounted for the entire commercial short reduction in adding 2300 contracts to a net long position now amounting to 16,500 contracts. The big 4 added around a 100 contracts of new shorts, while the big 5 thru 8 bought back around the same small amount. Hopefully, this Friday's Bank Participation Report for June may shed some light on JPM's silver net short position and updated prospects for my double cross premise. For now I'll leave JPM's short position pegged at 24,000 contra
On the sell side in silver, the managed money traders sold a bit over 4000 contracts net, including 6,264 contracts of long liquidation (in my predictive range and yes, I'm scraping for some redemption), combined with the buyback of 2164 short contracts. As was the case in gold, the buyback of managed money shorts, while not large, was a head scratcher on a reporting week featuring the salami slicing of new price lows.
Where to from here? After being out close to $1.5 billion in unrealized combined losses at $1295 in gold and near $18 in silver, to being ahead a similar amount at recent lows, I would roughly estimate the commercials as having booked several hundred million dollars in realized gains on closed out positions and close to that amount in still open gains. While that means Friday's rally erased much of the commercials' unrealized open profits and the technical funds open losses, there would appear to be little question that the commercials reloaded short positions at the day's highs, while the technical funds reestablished long positions at those same high prices (after selling and booking losses in the prior two weeks at lower prices). This has been the essence of gold and silver price movement. Thus, in strictly futures market positioning terms, the commercials would seem to be in control.
Complicating matters, perhaps greatly, are the goings on in GLD and the June COMEX gold deliveries, where absolutely astounding amounts of gold are changing hands, at least on an historical comparison basis. After all, we are talking about hundreds of millions and billions of dollars' worth of gold. Of course, in terms of annual gold mine production (100 million oz) and the several billions of physical ounces of gold existing in the world, the hundreds of thousands or millions of ounces of gold involved recently in GLD and the COMEX June delivery process perhaps don't appear quite as significant.
Further complicating matters is the role of JPMorgan, which in addition to having accumulated 500 million ounces of physical silver over the past five years, has clearly emerged as the unprecedented largest taker of the June gold deliveries and, as I have speculated previously, is behind the 2 million oz of gold deposited into GLD during May and the 400,000 oz deposited this week. Certainly, the timing sequence would appear to make sense because to someone with unlimited buying power purchasing as much silver as one could afford would take time and patience, while buying billions of dollars' worth of gold could be accomplished on a much shorter timeframe. Such an entity, should it desire to buy both, would buy silver first, then add gold.