Commodity prices in the (very) long run

BACK in the late 1960s and early 1970s, rapid worldwide population growth and soaring commodity prices gave rise to fears that humans were outgrowing their planet's resource capacity. Some worried that crisis and Malthusian collapse was imminent. Among these pessimists was one Paul Ehrlich, a biologist who warned that population increase had gotten dangerously out of hand. Mr Ehrlich's writings generated scepticism in some quarters, however. Economist Julian Simon famously disagreed with Mr Ehrlich's view and argued instead that rising commodity prices would lead markets to respond, through efficiency, substitution, and supply increases. In 1980, he entered into a bet with Mr Ehrlich: that the price of a basket of five commodities (chromium, copper, nickel, tin, and tungsten) would be lower in a decade's time, in 1990. Mr Simon easily won his bet, striking a blow for the view that over the long run commodity prices effectively trigger market responses, thereby preventing Malthusian catastrophes.

Still, the view that a fundamental scarcity may generate commodity price spikes—and economic damage—beyond our capacity to respond is alive and well, fueled by a new era of dear commodities. It seems possible, some reckon, that Mr Simon just got lucky with the timing.

He may have, according to an interesting new NBER paper examining commodity prices over the very long run, from about 1850 on. David Jacks has assembled real commodity price data for 30 commodities, spanning animal products, energy products, and industrial and precious metals. He identifies three price trends corresponding to three time horizons: long-run trends, medium-run "supercycles", and short-run booms and busts.

Short-run booms and busts make for compelling financial journalism and can have nasty effects on the economies of commodity exporters. But it is the medium-run supercycles, which generally span a few decades, that seem to do most to shape our perceptions of "long-run" commodity price trends. Mr Jacks writes that these episodes seem to correspond to periods of rapid industrialisation and growth, producing an upswing in prices as soaring demand faces supply constraints, followed by a downswing as slowing growth meets expanding supply. Mr Jacks identifies major supercycle starting points in the 1890s, 1930s, and 1960s; followed by peaks in the 1910s, 1950s, and 1970s (around the time Messrs Ehrlich and Simon were preparing the bet); and endpoints in the 1930s, 1960s, and 1990s (the period at which Mr Simon was declared the winner). Mr Jacks notes the possible beginning of a new broad supercycle in the late 1990s but reckons its still too early to tell. Most of the commodities he tracks have returned to their very-long-run trends from below-trend points in the 1990s, and about half are now above trend. But based on historical patterns, he suggests that prices may be close to their supercycle peak.