A recent NBER working paper authored by Adrien Bilal of Harvard and Diego Känzig of Northwestern suggests the “social cost of carbon” is much higher than previous estimates.
If Bilal and Känzig’s new estimate of over $1,000 per metric ton is correct—and your correspondent believes it is likely closer to being right than that of a recent Nobel prize winning economist’s—carbon credits traded on Europe’s ETS market are undervalued by an eye-popping 1347 percent.
Sadly for both society and would-be hedge fund managers, SCC is not related to the price of carbon traded on cap-and-trade markets, so the undervaluation of carbon on the EU market is unexploitable by intelligent investors. This article explains the difference between carbon price and carbon cost and why the SCC is so important to our civilization continued survival.
Carbon dioxide is a chemical compound, so one molecule of CO2 is indistinguishable from another. This physical fungibility does not apply to the price paid on the carbon markets, where the traded value varies wildly depending on the market.
There are essentially two approaches to price carbon: sticks and carrots.
In the category of “sticks” we have cap-and-trade systems, where regulators create mandates related to greenhouse gas emissions for large, carbon-intensive industries such as electric utilities. If a utility emits more CO2 than its mandate, it must pay a per-ton fee for each ton of excess emissions. Closely related in approach are the so-called voluntary markets, where companies not subject to regulatory mandates nonetheless choose to pay for emissions above a target level of their own choosing.
Using carbon prices on these “stick” markets to get a sense of the perceived value of carbon by economic participants leaves one deeply confused.
Taking first the example of compliance markets, one metric ton of carbon dioxide equivalents was trading for $44.00 in California, €66.85 ($72.97) in the EU, and $6.43 in South Korean when I pulled data for this article.
The price of a ton of CO2 on the voluntary markets is a fraction of that paid on the cheapest compliance market; an Aviation Industry Offset was trading for $0.39 per metric ton on the same day carbon was trading for $72.97 in Europe’s compliance market.
Some carbon prices are set not by market participants but by government fiat—they represent the “carrots” in our carrot and stick world.
The U.S. government offers tax rebates to companies capturing carbon amounting to $85.00 per ton of carbon permanently sequestered; capturing a ton of carbon and using it to do something else (usually enhanced oil recovery) is worth $60.00 per ton. Under the Inflation Reduction Act, these rebates are payable in cash if the company’s operations are otherwise loss-making (i.e., they wouldn’t normally be required to pay taxes).
What is the difference between various market prices for carbon and the SCC? Unfortunately for humanity, traded and subsidized prices for carbon are fantasies that have no basis in physical reality. The molecule carbon dioxide has very little intrinsic value, so carbon market prices essentially reflect political policy timing (compliance markets and tax rebates) and corporate virtue signaling (voluntary markets).
The social cost of carbon, on the other hand, represents the windshield that the bug of our civilization will hit if we continue our present planet-wide experiment testing how far we can throw off the natural carbon cycle.
Or to use more staid language, the SCC is economists’ best guess at the future damage to society created by releasing another ton of CO2 equivalent into the atmosphere today, with the dollar value of future damage discounted to the present.
William Nordhaus, the Sterling Professor of Economics & Professor in the School of the Environment at Yale University built a vaunted academic career on his research into the economics of climate change which was rewarded by his winning of the (fake) Nobel Prize for Economics in 2018.
Dr. Nordhaus published a paper in 2017 in which he uses his DICE model and estimated that the SCC in 2010 was $31 per metric ton. Factoring in inflation, Nordhaus’ calculation implies that the social cost of carbon at present is around $47.
In his 2018 Nobel lecture, Nordhaus opines that the optimal level of global warming was just over four degrees Celsius (see page 6 of the slides that accompany his lecture at the above link). If this is Bill Nordhaus’ best guess on optimal temperature rise, my best guess is that Bill Nordhaus has not talked to any farmers or physical scientists.
As we get over two degrees of warming, we start to bump into tipping points that would make large-scale extinctions of marine and terrestrial life more likely and wreak havoc with global food systems. A sensible social cost of carbon should reflect the damage to our society would incur from these disruptions.
Neither Nordhaus’ $47 per ton estimate nor the $73 per ton price on the EU ETS market nor the $85 per ton 45Q tax credit come anywhere near accurately reflecting the damage of such civilization-scale chaos.
While a silver back gorilla of the economics world like Nordhaus cares nothing about what your correspondent thinks, I believe his SCC estimate is ridiculous and his DICE model is a travesty. I have publicly called DICE “a weapon of mass destruction” in a previous article due to the complacence it has bred in the political classes.
Why does Nordhaus get the SCC so wrong?
One of the pillars of Nordhaus’ DICE model is an assumption that any drop in economic output caused by global warming can be extrapolated linearly from data that relates local temperatures and local output.
Nordhaus compares average temperatures for individual states—New York and Florida, for example—and reasons that because Florida’s GDP is slightly lower than New York’s, the relationship of increasing temperatures on GDP is weak. Nordhaus extrapolates the weak negative relationship between local temperatures and GDPs to apply to the impact of warming on the entire planet.
Nordhaus’ conclusion is that society should wait for a long time before bothering to take climate action because the costs are so small and are so distant in time.
Bilal and Känzig sensibly reject Nordhaus’ association between local temperatures and state-level GDP. Instead, they build a dataset of 173 countries over 120 years that attempts to tie global changes in temperature to global economic output.
The conclusion they reach when analyzing this dataset is that each extra ton of CO2 emitted into the atmosphere carries with it a real cost to civilization of $1,056 per metric ton. They also contend that were the earth not to have already experienced fifty years of warming, our society would have been 37% more well off than it is today.
Is the model developed by Bilal and Känzig right? No. Their model suffers from weaknesses inherent in statistical forecasting related to the number of observations and explanatory power.
Also, like Nordhaus, Bilal and Känzig are attempting to use data collected during in one “attractor state” of a nonlinear dynamical system to predict outcomes in that system under a completely different attractor state. I am not a statistician, but I cannot see how this is not a misapplication of statistical reasoning.
Is the model developed by Bilal and Känzig a lot closer to being right than Nordhaus’ Nobel Prize winning attempt? Certainly so.
A die-off of half of all animals and plants on earth—a scenario the IPCC believes likely with four degrees Celsius of warming—will destroy human civilization as we know it. This is one case in which a seemingly academic difference between cost and price holds enormous consequences.
Intelligent investors take note.
