Multiple Fallacies of the Simon-Ehrlich Wager

 Five years ago, I described this very famous 1980 wager here. Wikipedia has a good, understated description of the wager.  

Ehrlich-Simon wager according to the Wall Street Journal, Jonathan V. Last, August 30, 2013, more less when I started analyzing this wager.

The fact that Ehrlich "lost" according to the omnipotent neoliberal economists was used broadly as a vindication of the power of economics in predicting what would happen in the physical world.  But were the economists mostly spouting nonsense, as is their long-standing tradition?

Natural copper.  Copper is one of the most important metals in the global economy that gravitates towards the "low carbon solutions" (whatever this term may mean to you). Source: Wikipedia.

In short, Simon challenged Ehrlich to choose any raw material he wanted and a date more than a year away, and he would wager on the inflation-adjusted prices decreasing as opposed to increasing. Ehrlich chose copper (Cu), chromium, nickel (Ni), tin (Sn), and tungsten. The bet was formalized on September 29, 1980, with September 29, 1990, as the payoff date.  

Since chromium is a stainless steel additive and tungtsen a rare earth metal, the consistent monthly price data for these two metals are missing in public domain. Consequently, I replaced chromium with zinc (Zn) and tungsten with silver (Ag). Zinc is used as a component of bronze and for metal plating to prevent corrosion. Silver is an essential metal used in electronics, solar cells, but also in coins, jewelry, etc.

Because I started to analyze this wager around September 2013, my starting date was September 29, 1982, not 1980. This is how far back the public domain price data went. Here I will update the wager and question its usefulness in the physical world and in empirical economics.

Let me begin with the "inflation adjusted prices." This term may be well defined when we focus on the USA with our global reserve currency and carefully orchestrated inflation index. As recent work of my brilliant PhD student, Philip Mitchell, shows however, other countries suffer from (a) their own oft erratic inflation rates and (b) the inevitable speculation related to the rate of exchange of their currencies to the US dollar (item (b) feeds into item (a)). Thus, for most countries on the planet, the US inflation adjusted index makes little sense. To put it simply, one day a unit of some currency may buy a different quantity of goods denominated in USD, than next day - while the US dollar price of that good remains constant. The last I checked, most people in the world live outside of the USA.

Let me now focus on inflation in the US. The "observed" inflation is a simple story by definition. The Central Bank here pretty much sets the inflation's pace by political considerations and by manipulating carefully interest rate and printing money. As we will point out in our upcoming paper, as of recent, this printing out has gotten rather out of hand.

One hundred dollars on September 29, 1982 is worth 310 dollars today. It is kind of scary, but quite predictable if you have dollars. The US Federal Bank manipulates the dollar inflation index to be more than 2% per year but not too much more, based on a basket of goods they define. Barring the post-2008 financial shenanigans, the US inflation rate has been almost constant at roughly 5%/year, and the plot is a straight line. 

Before I discuss metals, I should describe the role of crude oil in stimulating economic "growth".  I have talked about it for many years in this blog. Oil permeates so thoroughly the "tissue" of the societal superorganism that its use scales with "mass" or size of this superorganism (its GDP/capita) as a power law.  The exponent of this power law is similar to the scaling of metabolism at rest with body mass in living organisms.  Thus, cheap oil stimulates almost everything in an economy.  In particular, it makes mining, refining, finishing and distribution of all metals cheaper, provided that these metals retain a similar level of physical availability. In other words, if the mines are still functioning, ore grade does not deteriorate, plentiful water for mining is available and demand is stable, a metal's price should go down in tandem with oil price.

The West Texas Intermediate (WTI) price since September 1982 in dollars of the day and constant 1982 dollars. If your currency is pegged to US dollar or to oil (almost the same thing), oil has remained cheap during most of the ensuing four decades. The post October 2004 price increase is real.  This is when production rate of conventional oil plateaued. The 2008 spike had to do with the pre-market crash speculation. You can also see the first Iraqi war (Jan 1991) and the 9/11 price spikes.  The late 2014-2016 price crash was caused by too much oil chasing insufficient demand.  Finally, you can see the 2020 COVID dip in the price. The black curve shows you how inflation squeezed the "real" oil price down by a factor of three by 2021. The last statement is true only if you always have access to an abundant supply of US dollars to finance your oil purchases.

After this preamble, let's look now at the metal prices in dollars of the day, and see how they correlate with the WTI price. The portfolio prices have been above the oil price most of the time, and were well correlated with the oil price. After 2008, this correlation became almost prefect. Since you already know about the role crude oil plays in the human economy, you understand why. If  you buy metals on the world market, and your currency is not pegged to dollar, you likely pay attention only to the dollar of the day, and constant 1982 dollars mean little to you.  This is the first major fallacy of the Simon-Ehrlich wager.

Prices of the metal portfolio with the WTI price superimposed (the black curve).  We start from $200 of each of the five metals.  Their prices are stacked on top of one another, adding to $1000 at the start of the zinc (Zn) curve.  For comparison, we also show $1000 of WTI crude oil.  Since October 2004, there is an almost 1:1 correlation between the price dynamics of the portfolio and oil.  In dollars of the day, Ehrlich has been 80% correct.  One can argue that for most metal purchasers around the world, it is the metal price of the day that counts, not its discounted version.  

The second major fallacy stems from the abstract way economists think about material goods.  A tacit assumption in economics is that if price of something becomes too high this thing can be replaced with another one made from different materials and using a different technology.  But this way of thinking is incorrect when your task is to buy a specific metal for specific applications.  Copper wires, for example, could be replaced with aluminum  ones, but no one in their right mind would volunteer this replacement.  I know because my parent's house was built by Germans during WWII, when copper was unavailable for civilian uses.  So all the wiring in our house was made of aluminum.  When used, the wires would really heat up, because of their higher specific resistance that was not offset by the larger wire cross-sections.  Also aluminum flows under the pressure of screws in the outlets and in appliances, so most wire contacts would become loose  and would spark and/or overheat.  Bad news all around.  

So a better question to ask is this: How much of a metal or oil a given quantity of constant 1982 dollars would buy, if it bought 1 unit of each in September 1982?  Note that I do not believe in constant dollars as much as economists do, simply because most people in the world have to convert their currency of the day to US dollars of the day to buy a metal or oil.  The relationships among the world currencies and the US dollar are complex and mostly nonlinear.  A simple discount rate, shown in the first graph above, will not do!

Nevertheless, in constant 1982 dollars, Simon would win the wager for oil 80% of the time, for tin (Sn) 99%, and silver (Ag) 90% of the time.   Ehrlich would win for copper 63% of the time, nickel (Ni) 60%, and zinc (Zn) would be almost a break even.  Thus the simplistic pronunciations of Simon winning  are simply false, even in constant dollars.

Here is my olive branch to the economists.  The question I ask is this: How much of a metal and oil would the constant 1982 dollars buy, if they bought one unit of each in September 1982.

Now if the same question were asked in dollars of the day, as it would be in most countries, Ehrlich would win for copper 93%, nickel 94%, and zinc 95% of the time, while oil and silver would essentially break even. Simon would win for tin 61% of the time.

A follow up to the previous graph. A slightly different question I ask is this: How much of a metal and oil would dollars of the day buy, if they bought one unit of each in September 1982. Note the incredible impact of the stagnating production rate of conventional oil worldwide.  The world to the left of the vertical dotted line is different from that to the right. Some of the conventional pushback is summarized here.

If magical technologies, in which Simon believed religiously, were to supplant the physical Earth and mineral availability, and the "laws" of economics dominated the Laws of nature, then past October 2004, and in the dollars of the day, a purchasing manager would still be able to buy more of each metal and oil than in 1982, regardless of inflation. Please look at the chart above and check if you believe your own lying eyes or Dr. Simon. This is the third major fallacy that Simon brought into this (in)famous discussion of metal prices.

"The essence of wealth is the capacity to control the forces of nature, and the extent of wealth depends upon the level of technology and the ability to create new knowledge." J. Simon, "The Ultimate Resource", 2nd ed. Princeton University Press, 1998) - ISBN: 9780691003818.

In summary, the blanket statements issued by economists and journalists are too simplistic and do not reflect the nuances of this complex wager.  Dr. Paul Ehrlich has proven himself to be a deep thinking scientist, while Dr. Julian Simon remained a happy imbecile until his death. So much for the confrontation of science of the spherical physical world with the flat-earth, infinity-seeking economics.  But I already lamented this state of affairs eight years ago. There is little new I can add today.

Here is  - in my opinion - the most imbecilic comment Simon ever made: "We now have in our hands—really, in our libraries—the technology to feed, clothe, and supply energy to an ever-growing population for the next seven billion years. "The State of Humanity: Steadily Improving". Cato Institute Policy Report, www.cato.org, September/October 1995. 

A simple, conservative calculation shows that at the end of these 7 billion years there would be more humans than atoms in the Universe. I guess the intricacies of exponential growth and mass conservation eluded Dr. Simon.  Oh, and 7 billion years is almost twice the age of the Earth and 1/2 of the age of the Universe.