Back in the 1970s Paul Erhlich, author of The Population Bomb and an early advocate for recognizing limits to growth, entered into a famous bet with Julian Simon, one of the world's most faithfully pro-growth economists. The bet was styled as a "put your money where your mouth is" challenge. Would the prices of five commodity metals--copper, tungsten, chrome, nickel, tin--increase in the next ten years, as would be expected if what environmentalists were saying about limits to growth were true? Or would prices decrease, as technological innovation led to substitutes and efficiencies? Reluctantly, Ehrlich accepted the bet. And as we all know, he lost.
The well-publicized outcome emboldened those who dismissed environmentalist cautions as overblown worrying, for it reinforced (and encouraged the general public's passive acceptance of) the idea that infinite growth on a finite planet is possible because there are no limits to what human ingenuity can accomplish.
The problem is, the game--the economic system in which the prices are expressed--was rigged.
Here's a brief account of the bet from a website titled Environmental Economics. The account also covers some additional work by a pair of economists who purported to show that the results obtained for the ten years of the bet could be extended to the entire twentieth century.
September 13, 2005
The Simon-Ehrlich Bet Revisited
In the 1970s, Paul Ehrlich, a neo-Malthusian biologist, argued that excessive consumption was producing shortages of natural resources. Economist Julian Simon responded that, for the foreseeable future, the operations of the market and of human ingenuity would be so successful that not only would shortages be avoided, but prices of natural resources could actually be expected to fall.
Simon bet that the price of any set of raw materials would be lower ten years from now than it was today. Ehrlich and his supporters took up the challenge and, in October 1980, chose five metals: chrome, copper, nickel, tin, and tungsten.
Simon won the bet as, by October 1990, the composite price index of these five metals had fallen by more than 40 percent. Simon was humble in victory, however, noting that because metal prices are volatile, he could quite easily have lost. His theory was that raw material prices would trend down in the long-run, not necessarily in a period as short as ten years.
Now, David McClintick and Ross B. Emmett have taken the Simon-Ehrlich bet one step further, by testing Simon’s long-run hypothesis. In "Betting on The Wealth Of Nature" (PERC Reports, at www.perc.org/publications/percreports/sept2005/betting.php), they have tracked a composite index of the prices of Ehrlich’s five metals from 1900 to 2000.
They find, first, that this index was approximately 50 percent lower at the end of the century than at the beginning. However, they also find that the index was, as Simon suspected, highly volatile. Generally speaking, from 1900 to 1920 it was falling (but with a spike in the late 1910s); from 1920 to the late 1970s it was rising (with a major "bump" in the early 1950s); and from 1980 to 2000 it was declining (with a spike in the mid-1980s).
So, both of Simon’s hypotheses were confirmed: (i) human ingenuity can overcome the effects of increasing demand; and (ii) that ingenuity will take time to manifest itself.
This seems like bad news for those of us interested in promoting ecologically sustainable economies based on the idea that there are ecological limits to industrial development. If there are environmental limits to growth, shouldn't they start manifesting themselves as scarcity of raw materials? Shouldn't the prices have gone up?
In a word, yes: and they probably would have, if other powerful economic forces hadn't suppressed them. McClintick and Emmett are guilty of a fallacy called "hasty conclusion." They don't examine or deal with alternative hypotheses. They don't take account of these other forces.
Let's ask: what assumptions and methodologies underlie the determination of price in this survey of the price of commodity metals in the twentieth century? The report given above doesn't say, but we can presume that the price being tracked was denominated in constant dollars. (Any price denominated in inflated dollars would have shown a remarkable increase.) But consider: since many ores and metals come from areas that do not use dollars as their official currency, we face the difficulty of allowing for exchange rates. Since for much of the twentieth century the U.S. dollar functioned as a reserve currency in other national economies, the U.S. had the luxury of printing dollars that would never be used as claims against our capacity to produce wealth. (Foreigners preferred to hold their dollars as a medium of exchange, rather than cashing them in by buying U.S. exports.) This kept dollars artificially valuable during this time--and it also kept the dollar price of commodity metals artificially low.
This is one way that the fix was in for much of the twentieth century. But there's a more difficult problem, one that's hard to see from within the framework of Neoclassical economic theory, which presumes that physical law--the laws of thermodynamics--don't apply to economic activity. You can begin to see this by exploring this question: what has been the energy cost of those metals throughout the century?
For commodity metals (or any material extracted from nature as a resource) dollar cost is not the relevant measure of cost. Dollar cost isn't absolute, objective, "real". An absolute, objective, real measure of cost of such natural resources would account for their caloric or thermodynamic cost. Dollar cost doesn't even begin to approximate it. Why? Because as long as there is reserve production capacity within the oil-production system, the price of a barrel of oil is set in the short, near, and long terms not by geological or ecological fact but by human decisions about production.
For more on how the production of oil has been regulated in order to stabilize price, see the history of the Texas Railway Commission. Later, after U.S. oil hit its Hubbert's Peak, OPEC took over and did market demand prorationing for the world market. MDP keeps oil prices relatively stable in the face of exogenous market effects--shifts in supply caused by new discoveries, technical glitches, hurricanes, wars, etc., and shifts in demand caused by technological change and predictable seasonal variations. (You can read the origin story of market demand prorationing in Texas here).
Market Demand Prorationing requires that there be a producer large enough to serve as a balancer, increasing or decreasing production in order that the market as a whole clears at an acceptable price. In the pre-1971, pre-Hubbert's peak days, this was Texas; post 1971, it has been Saudi Arabia. When producers no longer have reserve production capacity--when the valves throughout the system are open as far as they will go--there is no accordian in the system, no capacity to expand, and the price of oil will then be determined by the intersection of an inelastic supply with a demand curve that is also extremely inelastic in the short and near term. When demand shifts (as it does with the seasons, with every cold snap or heat wave, with every "let's-get-in-the-car-and-go-somewhere" holiday), the price bounces. With totally inelastic supply, and no capacity for market demand prorationing, we can expect to see the price of oil become extremely volatile. Which it has.
Energy has an energy cost--its Energy Return on Investment or EROI ratio--and that caloric cost then becomes part of the cost of any and everything whose production involves use of energy. Caloric cost does not track well with monetary cost. Caloric cost is set by physical realities (for instance, as a matter of physics it takes a certain amount of energy to lift a barrel of oil out of a well that has lost its dissolved-gas drive) and by the state of existing technology (which can invent new methods to use energy more effectively). The latter can be improved upon by human ingenuity; the former cannot. Monetary cost, in contrast, is pure social convention. Even the element of monetary cost that reflects the cost of energy inputs is a social convention, since the price of oil reflects decisions that are political, not strictly economic. The only rational, objective measure of the long-term cost of the matter and energy that enter into our economy is its energy cost. All else is social construct, which can be made to go up or down as we please.
Then: to say that the cost of metals has not increased but decreased over a century may say nothing more than that during that century fossil fuels were pumped out of the ground at a rate that kept the costs of energy low enough that increasing technological efficiencies could show up as a reduction in the market price of those metals. But: what happens to the long-range cost performance of those resources if you subtract this hidden subsidy from cheap oil?
Here's a research challenge for some enterprising environmental or ecological economists: how, in fact, did the energy cost of this market basket of metals fare over that same period? I think that an appeal to data will show that it has gone up. I expect that increasing efficiencies in the discovery, mining, transport, and refining of ores has been more than outweighed by the necessity of moving from rich, easily exploited ores to ores that are less concentrated, buried deeper, and farther from markets. But there's no need to guess. Data can demonstrate whether this presumption is true or false. It's a shame that Ehrlich didn't specify that the price of the metals would be denominated not in dollars but in units of energy embodied in the finished product.
To pump a finite, valuable resource faster in order to keep its price low may prove to have been a foolish strategy. Whether it will so prove--whether the low price of energy (the price that is deeply coded into the price of that market basked of metals that Ehrlich and Simon wagered over) was in the objective long-range economic interests of homo oeconomicus--depends on whether humans will discover another source of energy with an EROI as low as oil. The laws of thermodynamics would have one be skeptical of this possibility.
To trust that we will find a substitute for oil, with something like oil's enormously advantageous EROI, is not a rational belief--it can't be proved or disproved--but an article of faith. Thus does theology enter into what otherwise could be mistaken for an objective, data-based answer to a fairly simple question about prices. And if economics is going to be a science, rather than a priesthood expounding the dogmas of infinite economic growth, then economists ought to spurn the influence of faith whenever it crops up in their work. As it does here, in Simon's supposed victory over Ehrlich.