Gold's strong rebound upleg this year has been driven by big gold-futures buying. After abandoning gold last year, American futures speculators are returning to the yellow metal in droves. These capital inflows are a very bullish harbinger, as major futures buying is the primary fuel for young gold uplegs before investors return to take the baton. And this big gold-futures buying is likely less than half done!
From a pure fundamental supply-and-demand standpoint, gold's crushing losses last year were solely attributable to record gold-ETF selling by stock traders. The World Gold Council's comprehensive 2013 data showed that global gold-ETF outflows from epic share selling was actually a third greater than the total worldwide drop in gold demand! Without those extreme gold-ETF liquidations, gold wouldn't have plunged.
Thankfully, stock traders are just starting to buy gold-ETF shares again, resulting in capital inflows from the stock markets to gold for the first time in over a year. This critical mean reversion of investor interest in gold has barely even begun. So far, the flagship American (NYSEARCA:GLD) gold ETF has only recovered 1/25th of its bullion hemorrhaged in 13 months ending in January! Investors are driving this new gold-ETF holdings recovery.
But a major secondary factor in gold suffering its worst loss in a third of a century last year was record futures selling. In the first half of 2013, American futures speculators dumped gold at blistering sustained rates. Provocatively as soon as their outsized selling peaked mid-year, gold prices stabilized even though the heavy gold-ETF liquidations continued. Futures trading dominates global gold-price action!
There are multiple reasons for this. While gold trades universally in physical form, the actual prices vary slightly. The American gold-futures market provides one centralized price quotation that the rest of the markets can cue off. Actual gold bullion is costly and cumbersome to trade, but futures allow instant leveraged gold-price exposure to large hedgers and speculators. And gold futures have been around for decades.
American gold futures started trading in late 1974, when gold ownership finally became legal again for Americans after being banned for four decades by a Democratic president. Meanwhile GLD wasn't born until late 2004, three decades after U.S. gold futures started trading. So from a real-time-price and trader-sentiment perspective, American gold futures remain the only game in town. They truly are the gold price.
So just as extreme gold-futures selling slaughtered the gold price in the first half of 2013, heavy gold-futures buying is lifting it this year. The implications of this critical shift are very bullish. Based on multi-year averages, this gold-futures buying is likely only half done at best. As futures buying continues to push gold higher, more and more investors will be enticed back to strengthen and amplify gold's new upleg.
It's important to remember that futures are a zero-sum game. Every futures contract has one trader on the long side and another on the short side. The former is betting the underlying price will rise, and the latter that it will fall. Every dollar won by the winner is a direct dollar loss for the loser. Because of this core structure, the total number of longs and shorts outstanding in gold futures are always perfectly equal.
But there are two distinct groups of futures traders, hedgers and speculators. Hedgers actually produce or consume the underlying commodity, so they simply use futures to lock in their future selling or buying prices to minimize market risks on their businesses. But speculators trade futures solely in the hunt for profits, they have no commercial dealings in gold. Their highly-variable buying and selling drives the gold price.
Every week the main U.S. futures regulator releases a great report called the Commitments of Traders that breaks down the futures positions held by both hedgers and speculators. The charts in this essay are built from that CoT data, revealing how American futures speculators are betting on gold. And they have been buying it aggressively, which is why the gold price has surged so nicely in the past few months.
This chart may look complex, but it's quite simple. The green line shows the number of gold contracts that American futures speculators hold the long side of on a weekly basis. These are leveraged bets the gold price is going to rise, so the higher this metric the more bullish traders collectively are on gold. And the red line shows their bets on the short side, where higher numbers mean they are more bearish as a herd.
In order to grasp the implications of the big gold-futures buying this year, understanding the context of the big gold-futures selling last year is essential. Gold plunged 26.4% in the first half of last year, in three distinct selloffs that all had major futures-selling components. Last February, it all started when gold fell on a futures bear raid while most Asian traders were away for week-long Lunar New Year celebrations.
American speculators triggered this 6.7% 2-week decline by aggressively selling short gold futures. They effectively borrowed gold from other traders, sold it, and then hoped to buy it back cheaper later to repay their debt after its price had fallen. Speculators' total short-side bets on gold surged about 50k contracts in that time! This is truly a vast amount, as each futures contract controls 100 troy ounces of gold.
The equivalent of 5m ounces of gold hitting the markets in a couple of weeks, or 155.5 metric tons, was brutal. By that point in 2013, the total gold-bullion outflows from differential GLD-share selling was just 51.6t over 7 weeks. This pushed gold down near critical multi-year support at $1550, setting it up for April's shocking panic-like plunge. Once that technical line in the sand crumbled, all hell broke loose.
As $1550 failed in mid-April, gold plummeted 13.8% in just 2 trading days! Gold hadn't seen anything remotely close to that for three decades, it was crazy. That critical-support break triggered stop losses on speculators' long gold-futures contracts, so they were forced to liquidate. This sparked margin calls on other traders, spawning a vicious circle of selling. Unfortunately the weekly CoT data masks this anomaly.
The CoT reports are current to each Tuesday's close. Gold's panic-like plummet in mid-April happened on a Friday and Monday, right in the middle of a CoT week. While many traders were getting stopped out of long contracts, many other traders were buying them aggressively since gold's selloff was so extreme. So despite the minor weekly CoT changes, there was massive volume and churn within that week.
That event was so scary that it galvanized futures speculators into a hyper-bearish outlook. Just like at all extremes, they assumed that anomaly was the start of a new trend that would persist for some time. This led them to continue dumping gold futures relentlessly, making their bet a self-fulfilling prophecy. Between late April just after that plummet and early July, speculators fled gold futures at an unprecedented rate.
You can see this on the chart, the falling green line showing long positions being sold while the rising red one shows short bets growing. In futures trading, the price impact of selling an existing long position and selling to create a new short position is identical. The shorting accelerated as gold plunged again in June after Ben Bernanke laid out the Fed's best-case timeline for slowing its QE3 debt monetizations.
2013 was as far from a normal year in gold as you can get. Not only was it gold's worst year in nearly a third of a century, the second quarter was gold's worst quarter in an astounding 93 years! Epic gold selloffs like we witnessed last April and June simply don't happen, they are exceedingly rare. So there was no doubt that both futures speculators' extremely-bearish psychology and resulting bets weren't normal.
As I was trying to figure out just how wildly outlying all this was in the middle of last year, I needed some baseline for normal gold markets. I decided to simply look at their post-stock-panic years before 2013, the 2009-to-2012 era, for that comparison. As the next chart shows in a little bit, both speculators' total long and short contracts held in gold futures had radically different averages over that secular span.
The total deviation of both speculators' gold-futures longs and shorts from their 2009-to-2012 averages is represented by the yellow line above. By early July this critical metric had ballooned to over 204k contracts. This meant American futures traders had sold the equivalent of 20.4m ounces of gold that they would normally hold, or 634.8 metric tons! This dwarfed GLD's year-to-date liquidation of 411.1t.
The sheer magnitude of this first-half-of-2013 gold-futures selling defies belief. The World Gold Council reports all the world's mines supplied 2969t of gold in all of 2013. Since gold is produced at a constant rate, halving that yields 1484t in the first half. So American speculators' futures selling alone was so great in that span that it was like a 43% boost in mined supply! No wonder gold wilted under such an onslaught.
But the great thing about futures and markets in general is extremes never last. Eventually after anything is sold too long, bearishness peaks and the anomalous selling burns itself out. So there was no doubt that American speculators would have to start buying gold futures again soon to reverse these hyper-bearish bets. On the short side in particular, it was mandatory. Those record shorts had to be covered!
Despite their sophistication, gold-futures traders are human just like the rest of us. They too suffer from groupthink and herd mentality, getting too greedy and bullish when prices are already too high and too scared and bearish when prices are already too low. Historically, the aggregate speculators' gold-futures positions are actually strong contrarian indicators. Their low longs and high shorts predicted an imminent reversal.
Indeed in July and August the speculators started aggressively covering their shorts, buying about 75k contracts or 233.3t of gold. This drove a sharp 18.2% 2-month gold rally, but unfortunately it fizzled out. Major new uplegs are always born with widespread short covering, as speculators buying to close their shorts at profits are often the only buyers around near extreme lows at peak despair. But their buying is finite.
An upleg can't continue unless the upside price action initially sparked by short covering leads to enough bullish psychology to bring in other buyers. First futures speculators need to start adding new long-side bets, and then investors must gradually return to take over the capital-inflows lead. While there were encouraging signs of both gold-futures buying and gold-ETF buying, it soon ran out of steam.
So futures speculators resumed shorting gold with a vengeance in November, as they continued to whittle down their long-side bets. By early December, the total deviation of spec long and short contracts from their 2009-to-2012 averages was back up near 201k! But just as this extreme anomaly proved unsustainable in early July, it was no more so in early December. Futures selling was simply exhausted.
Provocatively for most of 2013, futures speculators feared nothing more than the Fed slowing its QE3 money printing to buy bonds. But when the rumor became fact and the QE3 taper arrived by surprise in December, gold only slumped to modest new lows. The American futures speculators didn't add to their high short positions, and they actually started buying longs again! Thus gold started to reverse higher.
In January and February the speculators' short covering accelerated, they have bought back over 62k contracts (the equivalent of 194.5t of gold) since early December. Once again this major short covering has birthed what is likely to grow into a major upleg. But even more encouraging, they have also started to buy on the long side in a major way for the first time since last year's carnage. This is a super-bullish omen!
Futures contracts have expiration dates, so speculators legally have to buy to cover their shorts to effectively repay their borrowed gold in a matter of months after selling it short. But they have no similar obligation to buy on the long side. So new long-side buying reflects a genuine shift in their collective sentiment away from the extreme bearishness that crushed gold in early 2013. And it feeds on itself.
The more futures contracts speculators buy, the higher the gold price rallies. This brings in even more buyers, both in the futures realm initially and later in the far-more-important investment realm. It also puts tremendous pressure on the remaining speculators with short positions to buy back their bleeding bets to stem their mounting losses. And incredibly, this highly-likely futures buying is only half over!
Once again that yellow data series shows the total deviation in speculators' gold-futures long and short contracts from their respective 2009-to-2012 averages in normal markets. This deviation peaked at 204.1k contracts in early July, and again at 200.8k in early December. All this gold that was sold had to be repurchased, driving gold higher, to return to market normalcy. As of the latest CoT report, it is still at 102.0k!
Gold has run 15.0% higher in the past several months or so almost solely on American futures buying. While stock-market capital has just started returning to gold via GLD, that is just 23.1t so far compared to 317.6t of futures buying. And these futures speculators still need to buy another 102.0k contracts, or 317.2t more gold, merely to mean revert to their secular-average levels of bets in normal market conditions!
This final chart extends this same CoT data back to 2008, to show those critical long-term averages. And the word average is key. It's certainly not like speculators' long-side gold-futures bets have to hit new record highs, or their short-side ones have to fall to zero. Between 2009 and 2012, before 2013's craziness, speculators averaged 288.5k long gold contracts and 65.4k short ones on a weekly basis per the CoTs.
As of the latest CoT report, speculators' longs were only back up to 208.6k contracts. This is still 79.9k, or the equivalent of 248.6t of gold, below their long-term post-panic average. By mid-December 2013, these bullish bets on gold had fallen to their lowest level in 5 years, since the also extreme and short-lived anomaly of 2008's stock panic. Just take a look at what speculator longs did after that crazy event!
They rocketed dramatically higher over the subsequent year or so, catapulting the gold price 54% higher and paving the way psychologically for investors to return en masse. This is all but certain to happen again after today's extreme, as that's just the way mean reversions out of extremes work universally in the financial markets. Once buying after extreme selling starts, it takes a long time for it to run its course.
There is less gold-buying fuel left on the short side, with speculators total shorts now at 87.4k contracts. This is only 22.1k or 68.6t above their pre-2013 post-panic average. Still, that certainly isn't a trivial amount of gold. Most short covering happens early on, before long-side buying. Back in early July, speculators' total shorts hit 178.9k contracts! That was their highest level in at least 14.5 years, if not ever.
Another 102k contracts of gold-futures buying, as much as has already happened, is bullish enough for gold. But one of the greatest things about mean reversions is they seldom merely return to averages after hitting extremes. They nearly always overshoot to the opposite side! Like a pendulum pulled too far to one side, the momentum built in the reversion is so strong that the pendulum can't just stop mid-arc.
So there is nearly a certainty that we are going to see way more than another 102k contracts or 317t of gold-futures buying by speculators. The odds are overwhelming that their total longs will not stop mid-arc at their 288.5k 2009-to-2012 average, but soar well beyond that up towards 375k or so like after the stock panic. And their shorts are likely to fall far below their 65.4k average, likely challenging 40k again.
Run these numbers, and we're looking at potentially 214k contracts of gold-futures buying or 665 metric tons in the next year or so! That much futures buying along with the investment buying the resulting gold upleg will create should easily push gold back up over $1800 again, it not much higher. Mean reversions out of extremes are the most powerful trends in all the markets, incredible profit opportunities.
The bottom line is big gold-futures buying has fueled this year's strong gold rally. And it wasn't just short covering like last summer, but the first major new long-side buying since last year's carnage as well. This greatly increases the odds that we are witnessing the birth of a major new mean-reversion upleg in gold. And incredibly, even to mean revert to average levels this futures buying is only half done so far.
But mean reversions out of market extremes never simply stop at averages, they overshoot in an often dramatic fashion. So there is likely a lot more futures gold buying coming than merely a return to normalcy would suggest. As this continues to relentlessly push gold higher, more and more investors with their far-larger pools of capital will return to take the baton. And a massive upleg will be the ultimate result.
Adam Hamilton, CPA
March 14, 2014
Copyright 2000 - 2014 Zeal LLC (ZealLLC.com)
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: I own extensive gold, silver, and precious-metals stock positions.