Copper: A Macro Perspective

|
Includes: CU, JJCTF
by: Macro and Cheese

As a global macro manager, I don't trade the "wiggles" of copper prices. Instead, I take tactical positions that last anywhere from weeks to months. Because these positions are longer-term, I keep them small, and monitor them regularly to determine whether they are on track or not. Position sizes are modest, so I don't have to worry that events like a Peruvian miner's strike or a port closure will force me out of my position. I can wait them out.

Copper is a great instrument for macro trading because it is so abundant, and so tied to the economy. It can't easily be squeezed (though at times it is), and is generally faithful to the economic cycle. Volatility is reasonable (about the same as the stock market averages), and as far as markets go, relatively simple.

Here's what I think I need to know about copper: There are 400 pounds in a house, 50 pounds in a car, 40 pounds in an air conditioning unit, and another 30 pounds or so in household goods (dishwashers, refrigerators, washer-dryers, etc.). About 40% of all copper consumed is used in construction, and the balance, more or less, is in consumer durables. So if I am going to trade copper, I have to watch construction and durable goods.

That is, China's construction and durable goods. Americans are of course still using copper, but with our construction glut and crowded roads with 2.3 cars per household, we are not the people who set copper's price at the margin. That distinction goes to the Chinese. Our consumption of copper is relatively stable; Chinese consumption is not. For example, China is putting three times more cars on the road today than they were 5 years ago (14.5 million vs 4.3 million annually), whereas annual US auto sales have actually declined over that period, from 16.5 million to 13.5 million. That last Chinese car to roll off the assembly line is what moves the price.

I've used the example of automobiles deliberately, and this chart shows why (this and subsequent charts courtesy of Bloomberg):

Copper composite

This is a noisy chart, but please bear with me, because it has an important message. The blue and white candlestick chart is copper; the yellow line is US housing starts; the green line is the Shanghai Property Index, a proxy for Chinese real estate activity. As you can see, all three collapsed into early 2009 along with the world economy. Note that the price of copper recovered, but the green and yellow housing indicators have not.

The purple line represents Chinese monthly auto sales. With Chinese growth averaging 10% annually, auto sales have been strong, and it seems clear that this is a key factor affecting the price of copper. Auto sales are of course tied to the Chinese economy. As far as copper goes, if you are in the bull Jim (Rogers) camp you are bullish on copper, and if you are in the bear Jim (Chanos) camp you are bearish.

But if I'm going to take a position in copper I need to do better than that. There's a lot more information we can consider to hone our opinion. Let's look at Chinese inventories:

Copper, inventories

The green line below the blue and white chart of copper represents the copper inventory of the Shanghai Futures Exchange. As you can see, inventories have been rising rapidly, and in the past this movement has been associated with declines in the price of copper.

As we saw earlier, household appliances together with air conditioners account for even more copper than an automobile, so we should not overlook this sector in our decisionmaking process. Here is copper, again in blue and white, overlaid with the stock price of a major Chinese appliance manufacturer Qingdao Haier (a proxy for Chinese appliance consumption):

Qingdao Haier

As we see, the prices of the two assets have followed each other quite closely over the past three years. Starting at the end of the third quarter of last year, however, the price of Qingdao stock has not confirmed the strength of copper. The breakdown of this relationship is not conclusive, but it shows up in the price of other Chinese companies linked to copper consumption such as Jiangling Motors.

Next, we can examine the shape of the "copper curve." As you are no doubt aware, the various maturities of futures contracts do not trade at the same price (or very rarely), and commodity traders talk of "contango" (the months farther out are higher in price) or "backwardation" (the nearer months are more expensive). When the nearest month declines in price relative to the "back" months, this can be seen as copper supply looking for storage. In other words, there's too much of it.

Copper, curve

The red line represents the May 2012 contract price minus the December 2012 price. Because the December copper contract is higher in price than the May contract, this value is currently negative. Note that the curve generally follows the copper market rather closely. This graph therefore gives us two clues: (1) copper is plentiful based on the contango, and (2) the price of the metal is diverging from its normal relationship to the calendar spread.

Finally, here is a graph of copper overlaid with an indicator designed to mimic its price:

Copper, indicator

This indicator, shown in red, is a composite of the stock prices of more than ten international companies linked to copper in one or more fundamental ways. As we saw with Qingdao, the value of the indicator has begun to diverge from the price of the commodity. This evidence is not conclusive, of course, but is flashing a yellow light that copper may experience some headwinds.

I am rather bearish on China, for many of the well-known reasons that have been making the rounds in the press these days. But setting that view aside, on the strength of the evidence we've seen--the lack of price support from the real estate sector, rising Chinese inventories, flagging consumer durable stock prices, the shape of the copper futures calendar--I would have a negative outlook for this commodity.

I am short copper and plan to add to the position if (and only if) the price continues to decline from its recent peak set on January 27.

Disclosure: I am short copper futures and may modify this position at any time.