General equilibrium, anyone?

Paul Krugman doesn't think commodity prices are coming down. They did after the big spike in the '70s, but this time, he says, the supply is just not sufficient:

[T]his time may be different: concerns about what happens when an ever-growing world economy pushes up against the limits of a finite planet ring truer now than they did in the 1970s.

For one thing, I don’t expect growth in China to slow sharply anytime soon. That’s a big contrast with what happened in the 1970s, when growth in Japan and Europe, the emerging economies of the time, downshifted — and thereby took a lot of pressure off the world’s resources.

Meanwhile, resources are getting harder to find...Suppose that we really are running up against global limits. What does it mean?...[R]ich countries will face steady pressure on their economies from rising resource prices, making it harder to raise their standard of living. And some poor countries will find themselves living dangerously close to the edge — or over it.

Wait, run this by me again...rich countries will face pressure to stop growing, poor countries will be pushed over the edge...but China's growth won't slow? How does that work? Does China possess some magical property that will allow it to grow and grow and grow as resource costs skyrocket, while all other countries tumble into depression?

The answer is: No. When world growth outstrips resource availability, forces act to bring growth back in line with what is physically possible. The main force is "demand destruction" - economies crash, until they force themselves to use commodities more efficiently, and reconcile themselves to a permanently slower pace of growth. (Note: Matt Yglesias also picked up on this. So did The Economist, sort of.)

I'm surprised that Krugman doesn't realize this will inevitably happen. He's making a rookie mistake, applying "partial equilibrium" analysis - i.e., assuming that current trends will continue without affecting the forces that are driving those trends - instead of "general equilibrium," where everything feeds back into everything. As my professor Jim Hines says, "a lot of great economics papers have been written by people who saw a partial equilibrium result, and proved that it disappeared in general equilibrium."

No, "demand destruction" is not a fun way to restrain commodity prices. But, without infinite supplies of oil and food and metal, it's the only way. And commodity prices may not plunge, but they will have to plateau. As Herb Stein so famously declared, "Things that can't go on forever, don't."