The Tendency of Commodity Prices to Fall Over the Long Run

The Prebisch-Singer hypothesis states that real commodity prices have a tendency to fall over the long run. Harvey et al. have shown that the historical evidence from four centuries of commodity price data is consistent with the hypothesis. The most convincing explanation of the phenomena is that, with important exceptions, the income elasticity of primary commodities is less than 1. This means that incomes grow faster than demand for primary commodities, so that prices must fall to clear markets. This mechanism is subject to ecological constraints since if the ecological constraint bites, prices must rise to clear markets. So far, fundamental ecological limits have not trumped the tendency of commodity prices to fall over the long run. The tendency introduces a dynamic, systematic bias in favor of the core and against the periphery of the world economy since, by construction, nations or regions specializing in primary goods belong to the latter. In what follows, I’ll illustrate the tendency for a large number of commodity classes since 1850. All data is from David S. Jacks’s website.

Overall, the picture is that non-fuel commodity prices have fallen. Figure 1 shows the unweighted average of 37 non-fuel commodities. While there is a medium term cycle, the trend is clearly negative.


Figure 1. Non-fuel real commodity prices (1850-2010).

The main exception is energy prices, which have shown a tendency to rise over the long run. See Figure 2.


Figure 2. Real energy prices (1850-2010).

We should not expect the real price of gold (or other precious metals) to fall over the long run because it was, and still approximately is, the numeraire. Indeed, there is no long term trend in precious metal prices. See Figure 3.



Figure 3. Precious metal prices (1850-2010).

Figure 4 shows the price of beef and other animal products. We see that even though pork and hide prices have fallen over the long run, the price of beef (and lamb) have increased significantly. This is because the income elasticity of beef is greater than 1 since it is not yet a necessity for the bulk of global households. Indeed, as mass affluence spreads around the world, we should expect the demand for beef to grow faster than global income.


Figure 4. Animal product prices (1850-2010).

Let’s move to proper commodities starting with grains. Growing wheat, rice and corn is the mainstay of the world’s farmers. The real price of their product has been falling systematically over the past 150 years. See Figure 5.


Figure 5. Grain prices (1850-2010).

I know what you are thinking: These farmers ought to plant cash crops such as cotton. Not so fast. The prices of cash crops have also fallen just as much if not more than grains. Figure 6 shows non-food soft commodities.


Figure 6. Non-food soft commodity prices (1850-2010).

Figure 7 shows the systematic decline in the price of “drug foods” that played such an important role in the early modern world economy.



Figure 7. Drug food prices (1850-2010).

What about metals and minerals? Are commodities that are dug out lucky as a class? The evidence does not support that conclusion. Figure 8 shows metal prices. The data is noisier but the common trend component is clear.


Figure 8. Metal prices (1850-2010).

Figure 9 shows the prices of minerals such as iron ore and sulphur. Here the trend is even more manifest.


Figure 9. Mineral prices (1850-2010).

I hope to have convinced you of the tendency of commodity prices to fall over the long run. The finding raises the stakes for the politics of global inequality and the international division of labor. But trend analysis is merely the first step. For a full analysis of the role of commodities in the global condition, we have to look at how both the terms of trade between the town and the countryside (defined in terms of price levels) and the volatility of commodity prices affects commodity producers and informs their politics.