Copper: SPX Usurps Stockpiles
From boring industrial metal to commodities superstar, copper's wild bull market has been full of excitement. Its huge 475% run from its 2003 low to 2008 high made investors fortunes. Its subsequent panic-driven 69% plunge was catastrophic. And then copper's resilient 183% post-panic recovery to its recent April high has been breathtaking.
Futures traders, hedge-fund managers, and stock investors have endured extreme highs and lows across this bull. And quite often they're left scratching their heads wondering what's going to happen next. But as any seasoned copper trader well knows, this base metal's only certainty is acute volatility. When it comes to analysis, forecasting, or your best educated guess as to copper's future direction, be prepared to circumstantially adapt.
From a strategic perspective, copper's secular bull is driven by foundational economic fundamentals that will keep it alive for many more years to come. But things get a little trickier when we look at copper's tactical price movements. Naturally technical, sentimental, and shorter-term fundamental indicators serve as logical price drivers. But as we've seen one should not rely on any single indicator.
One such indicator that had been very reliable for us over the years was London Metal Exchange (LME) stockpile levels. As the world's premier non-ferrous metals exchange, the LME must hold physical inventory for when its contracts are redeemed. And thanks to the LME's ability to track the collective inventory of its global network of warehouses on a real-time basis, we have a unique fundamental read on aboveground copper supplies.
Now most of the world's mined copper is shipped directly to the end user after it is refined, leaving the volume that flows through these LME warehouses insignificant as far as the global copper trade goes. But these stockpile levels do play a vital role in the balance of the copper market. If end users aren't able to get enough of the metal from their regular supply channels, these LME stockpiles serve to supplement this excess demand.
LME stockpiles offer the largest source of accessible aboveground copper supply. And with the metal flowing in and out of the warehouses on a daily basis, these stockpile levels fluctuate accordingly. Well with copper demand outpacing supply for an extended period of time, naturally these stockpile levels are going to decrease on balance.
And when they fell to alarmingly-low levels where equivalent global consumption was measured in mere days, these stockpile levels became a lot more relevant as to where copper was being priced. As a result a striking inverse relationship developed between stockpile levels and prices. When stockpiles fell, prices would rise. And when stockpiles mounted a build, prices would fall.
As you can see in this chart, the inverse relationship between copper stockpiles and prices had played out incredibly well from 2006 to 2008. LME stockpile levels would fluctuate between about 100k to 200k tonnes in a tight horizontal trend channel. At support a big risk premium was placed on copper, driving its prices higher. While at resistance traders weren't as nervous about supply, which allowed copper prices to fall.
This relationship was apparent not only visually, but mathematically. From Q2 2006 to the end of 2008, stockpiles and prices had an inverse correlation of -0.852 with an r-square of 73%. This means 73% of the daily behavior of copper's price could be explained by the daily movement of LME stockpile levels. This is a very strong inverse correlation, and it was very actionable and reliable on the trading front.
But thanks to the first stock panic in 101 years, the integrity of this relationship was nuked. Panic selling drove copper prices so low that its inevitable bounce would happen regardless of interim fundamentals. When copper finally hit its December 2008 panic low of $1.28, its lowest price in over 4 years, it had actually fallen below the cost of production for many of its miners. Well as expected, copper's pop to the upside recovered most of its losses. But the stockpile-to-price inverse relationship would have trouble reconnecting.
As a result of the panic LME copper stockpiles mounted a huge build, soaring to 550k tonnes. But this build didn't come as too much of a surprise considering nearly all economic activity temporarily ground to a halt. Now in a normal environment a build of this magnitude would obliterate prices. But with copper so deeply oversold it bucked the trend and mounted a justifiable rise.
Finally in February 2009 LME stockpiles decisively rounded the corner and shed some weight. And with copper prices rising on balance while stockpiles were falling, the inverse correlation was back on. But this return to normalcy was short-lived.
By July 2009 stockpiles changed directions again, mounting another build. If this fundamental relationship held true, copper should have at least consolidated, if not corrected. And prices were high enough to afford a pullback. Copper had already soared 78% off its bottom and was settling in nicely near $2.50. But prices kept on rising.
One year following the panic-build high, LME copper stockpiles had clawed back up to the 550k-tonne level. And with prices also rising on balance during this build, it became apparent we were in the midst of a huge fundamental disconnect.
By all appearances traders were completely disregarding this once-reliable fundamental read. And the math confirms the visual. From the very day of copper's panic low to current, stockpiles and prices are sporting a correlation of 0.200 with a trivial r-square of 4%. What this r-square value tells us is there is currently no correlation at all between stockpiles and prices.
So what on earth is happening with copper? Why has this stockpile-to-price inverse relationship seemingly disappeared? Are today's copper prices justified based on such high LME stockpile levels? Well the easiest of these questions to answer is the first. The horse that continues to pull copper's carriage is the S&P 500 index (SPX).
This visually-striking chart provides the inarguable answer to what's driving copper's price action. And the uncanny similarity between the SPX and copper was born out of the Great Stock Panic of 2008. It is the actions of the SPX that led to the panic selling of virtually all assets and markets around the world. And copper, along with most of these other assets, has been looking to the SPX to guide its post-panic recovery.
Most commodities and commodities stocks found their bottoms several months before the SPX hit its own in March 2009. But when the SPX took off from this bottom, its ensuing upleg gave confidence to copper and the rest of the group to mount recoveries of their own. And as you can see copper has emulated the SPX's every move. Like the SPX it enjoyed its own 10-month uptrend, and has also been dragged down with the SPX's two recent pullbacks.
Over the course of this tight uptrend that spanned most of 2009, copper more than doubled in price. But as this upleg matured virtually every indicator screamed for a copper correction. Copper was way overbought technically, greed was waxing extreme, and LME stockpiles were mounting a huge build. But copper would wait for the SPX to show its hand first before it made a decisive move to the downside.
For nearly a year the SPX had not undergone a meaningful pullback. In the early months of summer 2009 it did retreat 7.1%, but this loss was quickly erased as the melt-up forged ahead. By January 2010 the SPX, and copper, were way overdue for pullbacks. At the time my business partner Adam Hamilton wrote pivotal essays discussing the SPX's staggering levitation act and calling for a copper correction. And indeed this correction unfolded.
Also as Adam had predicted, this correction came on the heels of an SPX pullback. Over 14 trading days into early February, the SPX had shed 8.1%. And over this exact same span copper shed 15.0%. But this pullback was also short-lived. The SPX bounced off its year low and charged higher with amazing resiliency, and to no surprise copper followed suit. By April both the SPX and copper achieved new cyclical-bull highs.
Well with renewed euro fears these highs didn't last for long, and the SPX entered into another retreat that has again taken copper with it. And this ongoing correction is panning out to be the biggest yet. As of May 26 the SPX was down 12.3%, with copper off 16.0% from its April high. And there is no sign of copper stepping away from the domineering effect of the world's premier stock index.
It is apparent that with the ongoing uncertainty in the global economy copper traders aren't yet willing to price this metal on its own merits. As shown in the chart above copper is mirroring the SPX's every move. And with copper and the SPX sporting an r-square of 95% from the SPX's March 2009 low to current, the math confirms this super-tight correlation.
So for now traders must consider the fortunes of the SPX when gaming copper. But when copper eventually diverges from the SPX's grip, which it will, where should it be priced? If we go back to the historic stockpile-to-price inverse relationship, at current levels of 480k tonnes LME stockpiles tell us prices should go down. But stockpile levels aren't the only thing we need to consider when pricing copper.
First, this LME data was never meant to serve as a primary indicator. Yes, this stockpile-to-price relationship has been reliable in the past. But technicals and sentiment among other indicators must always be considered. Also, the fundamental read from this LME data might not tell the same aboveground-inventory story now as it did a few years ago.
It is important to remember that LME stockpiles fell to such low levels as a result of copper's huge economic imbalance. According to the International Copper Study Group (ICSG), from 2003 to 2007 copper's refined balance was negative 1500k tonnes. With less copper supplied than was being consumed, the aboveground stockpiles were pilfered to historic lows.
This stockpile draw not only went directly into products, but the end users were building up their own inventories in order to secure copper for future use. At the time there was no way of knowing how long this supply deficit would last, so it was prudent to stock up on the metal.
Well as a result of the stock panic and ongoing global recession, from 2008 to current the ICSG calculates a supply surplus of about 480k tonnes. Provocatively this happens to be the same amount currently stored in the LME's warehouses. Now this 2+ year surplus still represents a tight balance, and in no way does it challenge the secular nature of copper's bull. But it does allow some wiggle room from an inventory perspective.
With end users currently uncertain of their own sustainability, and now comfortable that copper will be available in the warehouses of the major metals exchanges if needed, they can afford to draw down the very same inventories that they stocked up years earlier, their own.
You see, considering the current state of the economy higher LME stockpile levels don't necessarily mean overall aboveground copper supplies are that much bigger than they were a few years ago. It might simply be the case where consumers are drawing down their own private inventories, choosing not to replenish in order to preserve capital. This will naturally lead to an LME build as long as mine production stays consistent.
The scarcity, or perception of scarcity, of an asset is what causes consumers to stock up. But when that sense of scarcity is eased, consumers are more than willing to allow someone else to pay the carrying costs. Unfortunately there is no way of knowing the collective inventory levels of businesses and governments. But I suspect this higher LME stockpile level is not as bearish as many think. If "private" inventories are appropriately managed based on market conditions, they're probably a lot lower on balance than they were a few years ago.
Ultimately I believe it will take a bit of time to restore the stockpile-to-price inverse relationship. But because of copper's newfound tight correlation with the SPX, it wouldn't be the best indicator now anyway. For this reason we can continue to look to the SPX for copper and copper-stock trading cues.
And considering the SPX's current oversold conditions, the door has been opened to exceptional buying opportunities. At Zeal we're taking advantage of these bargains by buying elite copper and materials stocks. In our weekly Zeal Speculator newsletter we've been layering in to these stocks and options for a couple weeks now. And in the upcoming June issue of our acclaimed monthly Zeal Intelligence newsletter, we'll be recommending several fresh new buys to our subscribers. Subscribe today to view these trades and also gain unique insights into today's wild and crazy markets!
The bottom line is copper's price action is still heavily influenced by the fortunes of the SPX. While this market-darling base metal still has rock-solid long-term fundamentals, the ongoing global economic uncertainty is clouding the vision of copper's market makers. Even the once-reliable LME stockpile-to-price inverse relationship has factored little into copper's price action.
I suspect stockpile reads will eventually revert to normality once the economy improves and the global growth story returns. And it is a certainty that copper will eventually escape from the shadow of the SPX. But in the meantime we can use the SPX as a guide for trading this metal and its mining stocks. And with the SPX, and thus copper, enduring sharp corrections lately, investors now have excellent entry points.