The Arbitrary 'Social Cost of Carbon'

This month, the U.S. Supreme Court blocked the centerpiece of President Obama's climate-change agenda from taking effect, pending judicial review. Responding to the Court's decision, an Environmental Protection Agency spokesperson said, "you can't stay climate change and you can't stay climate action."

One of EPA's main justifications for "climate action" is an estimate they call the "social cost of carbon." While the social cost of carbon is an interesting intellectual exercise, the estimates vary wildly and are too arbitrary to use for cost-benefit analyses — especially those involving expensive and quite possibly illegal climate-change regulations.

Social-cost-of-carbon estimates are analytical tools. Like all analytical tools, they can help inform policy decisions if they are realistic and reliable. For example, if policymakers knew the economic impact on future society caused by human-induced climate change, they could decide how much we should be willing to pay today to avoid or minimize those future damages.

Perhaps the best-known model for calculating the social cost of carbon is Professor William Nordhaus' Dynamic Integrated Climate-Economy model (DICE). As Nordhaus describes, his model "integrates in an end-to-end fashion the economics, carbon cycle, climate science, and impacts in a highly aggregated model that allows a weighing of the costs and benefits of taking steps to slow greenhouse warming."

Obviously, this is a very sophisticated model. But does it map to the real world? It turns out we don't know if Nordhaus' model describes the real world, because it is based mainly on assumptions, not scientific facts.

Don't take my word for it. MIT Professor Robert Pindyck believes we should have a carbon tax, but he also says that even "calling these models 'close to useless' is generous."

He continues, "when it comes to the damage function [the costs of warming], we know virtually nothing — there is no theory and no data that we can draw from." As a result the model makers "simply make up arbitrary functional forms." Pindyck believes the models "can fool policy-makers into thinking that the forecasts the models generate have some kind of scientific legitimacy."

One way we can see how arbitrary these estimates are is to look at how widely they vary. In 2010, the Obama administration claimed the social cost of carbon was $21 per metric ton, but a mere three year later, they claimed it was actually 76 percent higher — $37 per ton. In 2015, professors at the University of Stanford claimed the social cost of carbon was actually $220 a ton.

As research by Robert Murphy of the Institute of Energy Research shows, you can get any answer you want by fiddling with the model parameters.

Using such an arbitrary theory to justify a massive new regulatory regime breeds hubris. In the case of EPA and its carbon rule, the social cost of carbon makes bureaucrats confident enough to dismiss a crushing Supreme Court decision as essentially meaningless ("you can't stay climate action").

While it may be a neat academic exercise, the metric is just too arbitrary to be used to evaluate any policies in the real world. As much as environmentalists might try, they can't stay the facts.

William O'Keefe serves on the board of the CO2 Coalition. The CO2 Coalition is a non-profit organization that works to educate thought leaders, policy makers, and the public about the important contributions made by carbon dioxide and fossil fuels to our lives and the economy.