The East Troublesome fire, Colorado’s second largest ever, was sweeping across Grand County forests on Oct. 25, 2020, part of a record-breaking wildfire season that forced evacuations, destroyed homes and took lives.
Half a world away, heavy rains were flooding the Bagerhat District in Bangladesh, in one of the worst monsoon seasons on record. And three days later, record-tying Hurricane Zeta slammed into the Gulf Coast, breaching Louisiana levees and causing storm-surge flooding in Mississippi and Alabama.
Such “natural disasters,” scientific studies find, are being bolstered in frequency and ferocity by the buildup of human-caused greenhouse gases in the atmosphere, and now there are growing efforts — in Washington and at the Colorado statehouse — to put a price on those gases for the economic damage they wreak.
The Biden administration and Colorado legislators are turning to a calculation called the “social cost of carbon” to measure the economic harm of putting one additional ton of carbon dioxide, the prime greenhouse gas, into the air.
The dollar value of the social cost of carbon — Colorado is using a minimum of $46 a ton — is, however, not without controversy.
“We really can’t pin down the social cost of carbon,” said Robert Pindyck, a Massachusetts Institute of Technology economist.
Thirteen Repubican attorneys general have sued to block the Biden administration’s effort to upgrade the models used by the federal government to calculate carbon costs, arguing the figure is “inherently speculative.”
Borrowing a phrase, and a bit of hyperbole, from James Madison’s “Federalist Papers,” they called the exercise “the very definition of tyranny.”
Attorneys general from 10 other Republican states have also filed a similar lawsuit, calling the social cost of carbon “the most important number you’ve never heard of.”
“This is where science butts up against political reality and necessity,” said Gernot Wagner, a New York University economist. “The question is, is a number better than no number? No number means there are no costs of carbon, but we know there is a cost.”
The 2020 Colorado fire season tallied $226 million in firefighting costs — there were nine fires, including the Cameron Peak fire, the largest in the state’s history — and $1 billion in insured property damage.
Hurricane Zeta, a record-matching sixth hurricane to make landfall in a season, inflicted $4.3 billion in damages across nine states, and the monsoons destroyed 1.3 million homes in Bangladesh.
The social cost of carbon is already embedded in a 2019 Colorado law regulating utilities, and there are two bills working their way through the legislature this year that would employ it to manage natural gas energy efficiency programs and cut the state’s greenhouse gas emissions.
The 2019 law, Senate Bill 236, directed the Colorado Public Utilities Commission to use a social cost of carbon in evaluating all existing electric generation and in the approval of the plan by Xcel Energy, the state’s largest electricity provider, for closing plants and adding new generation.
This legislative session, Senate Bill 200 would place firm caps on the state’s greenhouse emissions and give the Air Quality Control Commission the ability to use the social cost of carbon in any rulemaking affecting greenhouse gas emissions. Gov. Jared Polis has said he would veto the bill because of the emission caps.
A second bill, House Bill 1238, directs the PUC to use a social cost of carbon and a social cost of methane in evaluating energy efficiency and demand management programs for utilities, like Xcel and Atmos Energy, selling natural gas to homes and businesses.
“We really needed to have a planning assumption on the cost of carbon in investing taxpayer or rate payer dollars,” said Sen. Chris Hansen, a Denver Democrat and cosponsor of House Bill 1238.
Hansen said because there isn’t a carbon market price — some European countries do have markets where carbon credits are bought and sold — “this is the next best thing.”
In the first Colorado use of the social cost of carbon, Xcel Energy was required to include it in the resource plan for new and existing generation and transmission.
Under its preferred plan, Xcel proposes closing all its coal-fired power plants by the end of 2029, except the Comanche 3 unit, in Pueblo, which it wants to run until 2039. Environmental groups are pushing for Comanche 3 to be closed by 2029.
In its calculations, without factoring in the cost of carbon, closing Comanche 3 early will be $103 milion more expensive than Xcel’s preferred plan. But once the cost of carbon is added, closing Comanche 3 creates a $267 million savings, the company said in its filing.
“When you include the social cost of carbon there is no economically rational basis for running coal units,” Matt Gerhart, a Sierra Club attorney, said.
Jonathan Adelman, an Xcel vice president, said “it is really easy to cherry pick one number out of a complex filing. The cost of carbon is a planning tool, not something customers pay.”
Closing Comanche 3 early would require an additional 1,000 megawatts of new “dispatchable” generation, most likely natural gas-fired turbines, and would definitely have an impact on customer bills, Adelman said.
Keeping Comanche 3 running on a limited basis “buys us time for new technology to emerge,” he said. “We are trying to strike a balance.”
Seven other states — from Maine to California — are also using the social cost of carbon in utility sector policy and evaluation.
A serious push to cut greenhouse gases, Adelman said, will have to expand the use of the social cost of carbon beyond utilities to other economic sectors, especially transportation, now the largest source of carbon emissions in the U.S. and Colorado.
“There is an evolution here, a broader focus,” he said, “as we think about decarbonizing the broader economy.”
How the math works
But where does this number, this social cost of carbon come from? Is it, as NYU’s Wagner says, based on the “best available science,” or is it “arbitrary and capricious” as the Republican attorneys general contend?
It was another group of state attorneys general — Democrats this time — that helped create the social cost of carbon through a lawsuit that argued the administration of President George W. Bush failed to take into account carbon costs in issuing auto emission standards.
In 2007, the Supreme Court ruled that the federal Environmental Protection Agency had the authority and responsibility to regulate greenhouse gases. Bush-era federal agencies began applying various carbon cost measures in their planning and cost-benefit analyses.
President Barack Obama created the Interagency Working Group to develop a model to calculate a uniform cost of carbon and two other potent greenhouse gases — methane and nitrous oxide — to be used across the federal government.
The group was also charged with monitoring and improving the calculation, which initially came to a midrange figure of $42 a ton in 2007 dollars. This is the figure upon which Colorado carbon cost is based.
The Trump administration disbanded the working group and by tweaking the calculation reduced the cost of carbon to between $1 and $7.
One adjustment was to consider only domestic losses rather than global ones as had previously been done. So, yes to the East Troublesome Fire and Hurricane Zeta, and no to the Bangladeshi monsoons.
On his first day in office, President Joe Biden reestablished the working group and gave it a year to come up with an improved model for calculating the cost and a new dollar figure. In the interim the administration adjusted the Obama carbon cost for inflation to $51 a ton.
The social cost of methane, which is 30 times more effective than CO2 in trapping the Sun’s heat, was set at $1,500 a ton, and nitrous oxide, 200 to 300 times more potent, at $18,000 a ton.
“An accurate social cost is essential for agencies to accurately determine the social benefits of reducing greenhouse gas emissions when conducting cost-benefit analyses of regulatory and other actions,” Biden said in an executive order.
Based on research since the Obama days and initial positions taken by the Biden administration, it appears the price of a ton of carbon will rise. This is of more than passing interest to Colorado for legislation has tied the state’s cost of carbon number to the federal one.
The model that the working group must overhaul is a bit of a Rube Goldberg affair that starts with projections of future greenhouse gas emissions. Those emissions are fed into a model that forecasts the impact of those gases on climate — sea level rise, storms, droughts, wildfires and such. Then a second calculates the economic damage from those climate events and finally a financial assessment is made of how to value those losses into a single number.
“All of the pieces of the social cost of carbon are independent until you’ve fitted them together to know the number,” said Kevin Rennert, director of the social cost of carbon initiative at the Washington, D.C.-based think tank Resources for the Future.
Linking all the bits together is a tricky exercise and one that is filled with assumptions and decisions the modelers must make.
MIT’s Pindyck warned that models sometimes give the impression we know more than we do and can confer a “veneer of scientific legitimacy” to policymaking.
Some parts of the social-cost-of-carbon train are, however, more sturdy than others.
It is clear that heat-trapping greenhouse gases are building-up in the atmosphere, reaching 419 parts per million of carbon equivalent last month, according to the National Oceanic and Atmospheric Administration. That is the highest it has been in 3.6 million years.
Carbon dioxide makes up 75% of the greenhouse gases, with methane accounting for 16% and nitrous oxide 6%. The remaining 2% comes from fluorinated gases, such as HFCs, commonly used as refrigerants.
Annual greenhouse gases are projected by the federal Energy Information Administration to rise about 25% through 2050 to an emission rate of just under 44 billion metric tons a year.
So for the social cost of carbon, researchers just have to add a set amount of greenhouse gases into the models that generate forecasts of climate change. These so-called Earth-system models have over the last two decades become more accurate, detailed and with a finer resolution so that they are able to indicate regional and local impacts.
Among the things they have been able to help show is that climate change was a contributing factor in more than half of the 20 million acres of Western forests that burned over a 36-year period; that warming Atlantic Ocean sea surface temperatures are increasing storm activity as hurricanes become more powerful; and that warmer temperatures lead to heavier monsoon rains, with each 1 degree Centigrade increase adding 5% more rainfall.
And so, climate change likely had a hand in the East Troublesome Fire, Hurricane Zeta and the Bangladeshi floods.
But it’s complicated
Calculating the damages from climate change, however, is where the exercise hits a rough patch. The models, which date from the 1990s, use very simplified relationships and generalized assumptions to get to a dollar figure.
There are, however, efforts underway to develop better damage estimates.
The Climate Impact Lab, a group of more than 30 researchers from half a dozen universities and institutes, is trying to develop empirical data.
In a study of the impact of climate change on mortality and gross domestic product as the world warms, the lab collected mortality data from 40 countries, accounting for 57% of the world’s population.
The study found that there would be winners and losers. Oslo, Norway, where weather-related deaths are due to cold weather, would be a winner, while Accra, Ghana, where mortality is heat-related, would be a loser.
The study calculated a social cost of carbon for mortality at a median of $17 a ton when a moderate amount of emissions is added to the model and $36.60 if a high level of emissions is added.
A comprehensive 2017 study of the impacts of climate change on agriculture concluded that instead of $2.70-a-ton benefit (a carbon-enriched atmosphere helps plants grow) each additional ton of CO2 creates $8.50 a ton loss, as higher temperatures reduced yields.
Just those two elements, in a moderate emission scenario, would account for more than half the $46 a ton social cost of carbon Colorado is using, and that is before adding costs for energy, labor productivity, forestry and other economic activities.
And then there are “ecosystem services,” such as the mountain forests and streams, which serve as a reservoir and water system for Colorado. These ecosystem services don’t have a market value because they aren’t bought and sold, said Maximilian Auffhammer, a University of California Berkeley economist.
“Just because there isn’t a bald eagle McNugget doesn’t mean there isn’t a value to a bald eagle,” Auffhammer said.
The final element is the financial assessment that tallies up all the damages over time — as long as 100 years — and discounts them to present dollars. The discount rate is the key element in setting the figure: the lower the rate, the higher the cost of carbon.
The Obama administration used a 3% discount rate, reflecting going interest rates at the time and got its $42-a-ton average figure. The Trump administration used 7% and got the cost down to as low as $1 a ton.
The Biden administration has indicated that due to continued low interest rates it might use 2.5%, but to do that it will have to establish a clear method for setting the rate.
“This is a heavy lift. This is not an easy number to calculate,” Auffhammer said. “But in every sector we look at the damage and costs go up. The number is likely too low and not too high.”
“We are probably going to be wrong. But does that mean we should ignore the issue, that’s the wrong answer,” Auffhammer said.
Even Pindyck sees the need for a number. “We think we can describe the range,” he said. “We have to pick a number somewhere in the middle of the range, that is all you can do.”
Still, Sen. Faith Winter, a Westminster Democrat and a cosponsor of Senate Bill 200, said that it was essential to have the social cost of carbon in the bill and to have the AQCC use it in their deliberations.
“The social cost of carbon is a measure of the eventual costs to my constituents,” Winter said. “There are real social costs from greenhouse gases.”