Government rulemakers looking to decide how much money to spend to avoid adding more greenhouse gases to the atmosphere need a good estimate of what a warming climate will cost in social damages, for example through more extreme weather events.
That point makes the “social cost of carbon” one of the most critically important metrics underlying regulation of climate pollutants. An estimate of the dollar costs of each ton of carbon pollution caused by climate change, the social cost of carbon guides federal agencies that are required to consider the costs and benefits of proposed regulations. Federal agencies so far have used the social cost of carbon while writing regulations with more than $1 trillion in economic benefits.
In 2010, a governmental interagency working group in the Obama administration established the first federal social cost of carbon estimate of $45 per ton of carbon dioxide pollution. In 2017, newly inaugurated President Donald Trump quickly disbanded the interagency group by executive order, and within months his EPA slashed the metric to between $1 and $6. The latest research by an independent team of scientists concludes that the social cost of carbon should actually start at about $100 to $200 per ton of carbon dioxide pollution in 2020, increasing to nearly $600 by 2100.
Should presumptive Democratic nominee Joe Biden win the presidency in the November election, his federal agency appointees will undoubtedly set about revising the social cost of carbon to reflect the up-to-date climate science and economics research. The revised social cost of carbon will in turn justify more stringent federal climate regulations. A Donald Trump second term would instead result in another four years of underestimated climate impact costs and continued delays in efforts to curb carbon pollution.
A history of attacks
Since its inception, the social cost of carbon has been a target of those opposing climate regulations, including many Republican office holders in Washington, D.C. The neutered social cost of carbon estimate has now been used to justify weakening three major climate regulations: undoing the Clean Power Plan, freezing vehicle fuel efficiency standards, and, in July 2020, setting airplane greenhouse gas standards to levels matching those the industry already has already met.
In December 2017, congressional Democrats asked the Government Accountability Office to examine the Trump EPA’s new method for calculating the social cost of carbon. The GAO published its report in June 2020.
GAO confirmed that the Trump EPA slashed the social cost of carbon by implementing two dubious choices recommended by House Republicans in early 2017. The first was to consider only domestic, rather than global, climate damage costs. The vast majority of experts, including the National Academy of Sciences, agree that approach is inappropriate.
Carbon released by combustion of fossil fuels mixes with other gases throughout the atmosphere. Carbon pollution from the U.S. and associated damages don’t stay in the U.S. Rather, the climate impacts spread around the world, just as carbon pollution from other countries has consequences in America. Because all economies are interconnected through global trade, America’s economy is also damaged when other economies suffer. Considering only domestic climate damages lowered the social cost of carbon by a factor of seven, down to about $6 per ton of carbon pollution.
The second approach taken was related to the discount rate, which accounts for the fact that money saved and invested accrues interest and is thus more valuable than money spent today. The Trump EPA used discount rates of 3% and 7%, the latter of which suggests that we shouldn’t spend more than 3 cents today to avoid a dollar of damage 50 years from now. The 7% discount rate lowered the social cost of carbon to $1 per ton of carbon dioxide pollution.
The Trump EPA justified these two choices by citing a guidance document from the Office of Management and Budget called Circular A-4. But the GAO concluded that EPA had ignored inconvenient parts of that guidance, which states, “Where you choose to evaluate a regulation that is likely to have effects beyond the borders of the United States, these effects should be reported separately,” and “Special ethical considerations arise when comparing benefits and costs across generations … Future citizens who are affected by such choices cannot take part in making them, and today’s society must act with some consideration of their interest.”
The Trump EPA considered neither climate costs outside American borders nor the interests of future generations that would be harmed, economically and otherwise, by unchecked climate change.
What is the actual social cost of carbon?
To calculate the social cost of carbon, both the interagency working group and Trump EPA relied on three climate-economics models (called “integrated assessment models”). The most prominent of these is the Dynamic Integrated Climate-Economy (DICE) model, for which its creator William Nordhaus was awarded the Nobel Prize in Economic Sciences. On the same day Nordhaus won the Nobel Prize – October 7, 2018 – the IPCC published its Special Report on Global Warming of 1.5°C.
This was an odd coincidence. On the one hand, the IPCC concluded that even a global warming of 2 degrees Celsius (3.6°F) above pre-industrial temperatures would cause significantly more severe climate change consequences than 1.5 degrees Celsius (2.7°F), such as more extreme weather, severe coral reef mortality, disappearing Arctic sea ice, and so on. In contrast, Nordhaus’ DICE model indicated that the economically optimal climate pathway would result in 3.5 degrees Celsius (6.3°F) warming by 2100, which the IPCC has concluded would result in extremely severe consequences, such as 40–70% of all species being at risk of extinction, glacial retreats threatening water supplies for millions of people, sea-level rise inundating coastlines, and more. If this outcome is not actually optimal, it suggests the federal social cost of carbon estimate of $45 per ton based on DICE and similar models is too low, not too high, as the Trump EPA claims.
Newly published research consistent with Paris accord targets
Authors of a new study in the journal Nature Climate Change attempted to reconcile this disagreement by updating the climate science and economics assumptions in the DICE model. On the scientific side, the authors used recent research to update how much carbon is absorbed by nature and by how much the increased concentrations of greenhouse gases in the atmosphere affect temperatures. They found that with these updates, “the optimal temperature change by 2100 drops by 0.5°C compared to the original DICE calibration, to just below 3°C (5.4°F) by the end of this century.”
The researchers then updated what’s known as the model “damage function,” which translates global temperature changes to economic damages. Recent research has shown that DICE’s original damage function significantly underestimated the impact of climate change on economic activity and growth. Revising the damage function accordingly reduced the optimal temperature change another 0.8 degrees Celsius, to 2.2 degrees Celsius (4°F) above pre-industrial levels in 2100.
The Nature study also revised the discount rate used in DICE. This is a difficult issue involving how much we value saving vs. spending money today to slow climate change – an inherently subjective question. The researchers tackled this problem by using a 2018 study published in the American Economic Journal that surveyed 173 experts on the long-term social discount rate, whereas DICE previously relied on one person’s opinion (Nordhaus, who preferred a relatively high 4% discount rate). The lower resulting discount rate places “greater weight on the wellbeing of future generations than does Nordhaus’ calibration, leading to more stringent climate policies,” the authors wrote. It also lowers the optimal temperature change to 1.7–2.0 degrees Celsius (3.1–3.6°F) above pre-industrial levels by 2100.
When addressing greenhouse gases beyond carbon dioxide, the resulting optimal temperature change is even lower yet, at 1.5-1.8 degrees Celsius (2.7-3.2°F). Including uncertainties, the study authors concluded that the optimal economic climate pathway likely would result in a 1.2-2.2 degrees Celsius global warming by 2100 (2-4°F above pre-industrial temperatures). In 76% of the Nature study’s model runs, temperatures stayed below the Paris climate agreement’s target of 2°C (3.6°F) by 2100 in the optimal pathway.
In short, when incorporating the latest climate science and economics research, the optimal economic climate pathway is consistent with the Paris climate accord target of keeping global warming to less than 2°C, and ideally closer to 1.5°C above pre-industrial temperatures.
The researchers additionally evaluated what their updated model inputs would mean for the social cost of carbon. They concluded that it would translate to a social cost of carbon starting at about $100 to $200 per ton of carbon dioxide pollution in 2020, increasing to nearly $600 by 2100. As expected, this is substantially higher than the previous federal social cost of carbon estimate of $45 per ton, and vastly higher than the Trump EPA’s estimate of $1 to $6 per ton. It’s also consistent with several recent studies finding that the social cost of carbon is above $100 per ton of carbon dioxide pollution – about 100 times higher than the Trump EPA’s estimate.