Somebody Owns a Lot of Copper
Let me tell you about copper. Copper is a metal that lives in the earth, and sometimes people pull it out of the earth and make it into pennies or whatever. There is a lot of it. To give you some idea, Freeport-McMoRan, a big copper mining company, owns about 50 million tons of it, though most of that is buried in the ground and takes some effort to get out. Freeport took about 1.9 million tons of copper out of the ground last year; around the world, miners got out a total of around 18 million tons.
Here is a story about copper:
A single buyer has snapped up more than half the copper held in London Metal Exchange warehouses, giving it control over a crucial source of supply and raising concerns among traders about the potential for higher prices.
On several occasions in the last month, this buyer held as much as 90% of the world’s copper stored in LME-licensed warehouses, equal to about 140,000 tons, or enough to make the copper parts of the Statue of Liberty more than 1,700 times.
The Wall Street Journal reports that the single buyer is rumored to be Red Kite Group, "a London hedge-fund manager that focuses on metals trading." Whoever it is seems to have held more than 50 percent of LME stocks for most of the last four months. This is unusual, and sort of creepy. When one person owns 50 or 80 or 90 percent of the supply of a commodity, people tend to throw around words like "cornering the market," and worry about higher prices and manipulation.
But then there is that number. Red Kite, or whoever, owns about 140,000 tons of copper. That's a lot of copper compared to the amount of copper in the Statue of Liberty, or in the LME-licensed warehouses, but it's not a lot of copper compared to the amount of copper in the world. It's a little under three days' worth of worldwide copper production. The world produces one Statue of Liberty worth of copper every two and a half minutes. Having one or two thousand Statues of Liberty lying around is not, in that sense, a particularly big deal.
Or put it another way: Red Kite, or whoever, has a little more than one-quarter of one percent as much copper as Freeport-McMoRan. Of course Freeport's copper is mostly in the ground, and Red Kite's is in LME warehouses. "And it's probably easier to get copper out of an LME warehouse than it is to get it out of the ground," you might say, but it's not clear that you'd be right. A famous fact about LME warehouses is that it's really really hard to get metal out of them. In particular, that's the foundation of a famous story about aluminum: that Goldman Sachs and its commodity-trader buddies manipulated the price of aluminum by keeping it in warehouses and not letting it leave except for brief sightseeing tours around Detroit. There were long waits to get your aluminum out, so no one could get any aluminum, so Goldman and its cronies could manipulate the price of aluminum higher.
A judge recently rejected this story, and there were a bunch of problems with it. One problem is that the price of aluminum mostly went lower instead while the warehouses were doing their thing. Meanwhile, here's the price of copper over the last four months:
That is not, all in all, the chart of a commodity that has been cornered. There are some upward moves, especially the big one at the beginning and another one at the end, but the trend for most of the last four months has been down. You don't buy up all the copper in the world and then push its price down.
But Red Kite, or whoever, doesn't own all the copper in the world. It owns a tiny fraction of one percent of the copper in the world. Now, it owns 90 (or 50, or whatever) percent of the copper in LME warehouses. But that is an arbitrary quantity that sounds more important than it is. The LME warehouse system, as we discussed in the aluminum context, is set up to allow people to trade standardized abstract amounts of metal for financial purposes. If you want to hedge or speculate on metal prices, you trade standardized abstract LME contracts underpinned by metal in the warehouses. If you want to make things out of metal, you buy actual metal to be delivered to your factory.
The abstract-metal and actual-metal markets are different markets, but the boundaries between them are porous. When the highest use for a metal is abstraction, a lot of it goes into LME warehouses. When the highest use for it is pennies or beer cans or Statues of Liberty, it comes out of warehouses. From the Journal:
Accumulating such a dominant position became easier in June because the amount of metal under the exchange’s watch had plummeted, as had prices. The warehouses have held less than 160,000 tons of copper since mid-June, compared with more than 360,000 tons at the start of the year. Some analysts say copper production is running behind demand, forcing some users to draw on stockpiles in LME-licensed warehouses.
Loosely speaking, the problem of aluminum was that it was in deep contango: Prices for immediate delivery were low, prices for future delivery were high, and so buying aluminum and chucking it in a warehouse to deliver later was profitable. So people did, and the warehouses got pretty jammed up, and other people who wanted aluminum for immediate use found it all a bit unsporting.
Copper now has the reverse problem. It's in backwardation -- prices for immediate delivery are higher than prices for future delivery -- so there is a tendency to take it out of warehouses and sell it now rather than later. Why pay rent to keep metal in an LME warehouse while it declines in value? Why not sell it now, make more money and save on rent? So the amount in warehouses goes down, and the percentage represented by any fixed amount of copper goes up. If you own some copper in a warehouse, and you keep it there, and everyone else takes theirs out, then you now own all of the copper in the warehouse, though that does not by itself make you an evil market-cornering genius.
That doesn't mean that you couldn't do some evil market-cornering stuff by owning all of the copper in the LME warehouse system: The LME does set the benchmark price, and while it "says that it has mechanisms in place to prevent market squeezes," others quoted in the Journal article express some nervousness about what the big copper owner could get up to. But there are at least some limits: If someone managed to double the copper price on the LME, people would buy copper elsewhere. Or they would move their own copper into the LME warehouses, making the cornering not that effective.
The LME warehouse system is an interesting abstract representation of a commodity market, but you can get into trouble if you confuse it with the actual commodity market. One example of the trouble: Goldman and its cronies were accused of manipulating aluminum prices up by putting too much aluminum in LME warehouses. The worries about copper -- that it could be cornered, pushing prices up -- stem from there being too little copper in those warehouses. Both of those things can't be true.
I'll use metric tons throughout. Page 2 of Freeport-McMoRan's 10-K lists 111.2 billion pounds of "recoverable proven and probable mineral reserves" of copper. Google will convert that to 50.4 million metric tons. Converting proven and probable reserves to actual amounts of copper is a more metaphysical problem; I have just unscientifically asserted that Freeport "owns" that much.
Obviously I would advocate for measurement of copper in SoLs. More broadly, if there's a journalistic need for whimsical references for quantities, why not just use the whimsical references as units of measurement? "JPMorgan reported net income of three Olympic-sized swimming pools full of $100 bills, beating analysts' expectations of 2.5 swimming pools, on revenue of 17 swimming pools." Units are arbitrary.
The usual disclosure: I used to work at Goldman and still own a little restricted stock. I also own some aluminum and copper, both in my 401(k) plans and also in, like, I have some cans of Coke and aluminum foil and pennies and TV cables and stuff.
Because, basically, the trade had a tendency to flatten the contango: If contango causes people to buy spot and sell forward, that will depress forward prices and raise spot prices. If you are a spot buyer for manufacturing use, you might object to the higher spot prices.
The Journal again:
Some traders say the concentration of so much copper under one firm’s control is already driving up prices. It costs about $72 more per ton to buy copper for delivery today than for delivery in three months. Others say copper is more expensive because miners aren’t meeting global demand.
I see LMCADY LME <Comdty> (spot LME copper) at $6,762 a ton, versus $6,690 for LMCADS03 LME <Comdty> (3-month forward LME copper).
Possibly the reverse! Though anyone trading copper is smarter about copper than I am, so I hesitate to make fun. The Journal says that "The metal’s owner could be wagering that global copper supplies will tighten, causing prices to shoot up, analysts say," and, yes, a good reason to own copper is if you think copper prices will go up.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Matthew S Levine at email@example.com
To contact the editor on this story:
Zara Kessler at firstname.lastname@example.org