commodities cycles

commodity rhythms


The co-owner of one of the smaller investment companies I’ve worked for was a farmer.  He made me realize that there are no long cycles for most agricultural commodities.  If prices for a particular crop are high, farmers will plant more–usually a lot more–the following season.  That virtually guarantees that prices will either level out, or more likely fall.  The opposite happens–supply falls, and prices subsequently go up–if prices are currently low.

Considering that many crops have two or three growing seasons in a year, price adjustment comes swiftly.


Metals mining, especially base metals mining, is just the opposite.  Mines tend to be gigantic projects, costing billions of dollars and designed to last 20 years or more.  Most of that money is spent up front:  for the mine itself, for all the drilling machines and other earth moving equipment, for the ore processing plants, for the roads or rails to tap into a country’s established transport infrastructure, and maybe even for new sources of electric power.

Because the optimal project size is “humongous,” mines tend to spew out very large amounts of output when they open.

Because–unless you’re very unlucky–the running costs are low relative to the initial investment, projects seldom shut down once they’re up and running.  They normally don’t even consider doing so unless the output price falls below out-of-pocket extraction costs.  And even then a mine may not shut down.  Miners always identify pockets of especially rich ore that they set aside for a rainy day.  So the first response to weak pricing it to turn to these high-grade areas in order to keep going–and spewing even more price-depressing output on the market.

In addition, some emerging countries run their mines to create employment and get foreign exchange.  Because whether they make money or not is a secondary concern, such mines almost never shut down.

The result of all this is a supply/demand dynamic somewhat like the farm one I sketched out above.  When times are good and metals prices are high, miners generally spend their cash developing new mines.  This creates periodic overcapacity when supply outstrips global industrial demand as all the new mines open at once.  But, unlike the case with soybeans or corn, excess capacity doesn’t disappear come winter.  Instead, it can stay for a decade.  What cures the oversupply is the eventual expansion of the world economy to the point where it can use all the raw materials being produced.

an example

I was a starting-out analyst when a supply-demand imbalance sent base metals prices skyrocketing in 1980.  I remember copper briefly hitting around $1.40 a pound and bringing previously loss-making capacity back onstream.  The price almost immediately fell back.  It took nine years for demand to expand to the point where it absorbed all available supply–and for the price to regain that 1980 high ground.

Another wave of new capacity pushed the price back down in the mid-1990s, where it stayed again until sharply climbing demand from China absorbed all the new output.  The price began to rise again in 2003.

For most metals, this pattern of feast and famine is common.  It’s not alone.  Chemicals and shipbuilding are the same way.  The common threads are:  commodity industry; long-lived assets with most of the capital in up front; capacity additions coming in large chunks.

Try to find a copper chart that goes back to the 1980s.  It isn’t that easy–suggesting to me that commodities traders aren’t as up on their history as they should be.

investment significance

I think that for base metals, and maybe for gold as well, we’re deep in the end-game transition from fat years to lean.  It has less to do with the state of demand in China than the state of supply among mining companies.  If I’m correct, time–and the accompanying gradual world economic expansion–is the only cure.