By Greg Ip
As a growing number of global leaders pledged to achieve net-zero carbon emissions ahead of this week’s climate conference in Glasgow, an awkward question hovered in the background: Is it worth it? That is as much an economic as a scientific question, and the economics of climate are far less settled than the science.
Until recently, economic models, including one developed by the world’s best known climate economist, Nobel laureate William Nordhaus, concluded climate change would hurt economic growth, but not by enough to justify the steep cost of holding the rise in global temperatures to below 2 degrees celsius or achieving net zero emissions.
But some economists now argue those models use outdated data, put too low a value on future well-being and don’t properly price society’s desire for insurance against catastrophes. Their research is yielding higher costs from emissions, which doesn’t prove net zero is cost effective but suggests that prospect is more likely than once thought.
The White House is consulting these economists as it develops a new estimate of the social cost of carbon: That is the value, in today’s dollars, of all the future harm caused by a metric ton of carbon dioxide released now. A higher social cost of carbon would justify more stringent emissions regulations, and could thus influence how President Biden fulfills any commitments he makes in Glasgow.
Michael Greenstone, a University of Chicago economist who led the Obama administration’s calculation of the social cost of carbon, said this is no longer good enough. Mr. Nordhaus’ estimate of 2% GDP loss is “not grounded in empirical evidence,” he said. “Nor does it capture and reflect the heterogeneity, the differences in climate impacts around the world.”
He said that since 2010, 438 empirical studies of climate impact have been released, none of which are reflected in established models. Mr. Greenstone co-founded the Climate Impact Lab to put climate economics on an empirical foundation. Its findings break important new ground. For example, they show the two biggest sources of damage from climate are premature death and decreased labor productivity due to extreme heat, neither of which figure prominently in established models.
In the other direction, CIL tempers the harm from extreme weather and rising sea levels by assuming people adapt, such as by building levees. This, however, varies considerably by region: CIL finds Norway will spend less on winter heat while Nigeria will spend 20 times more on electricity for more air conditioning. But a swath of developing populations won’t even be able to afford air conditioning and will have to endure the consequences. By dividing the world into 25,000 regions, CIL identifies localized risks that less granular models like Dice miss.
CIL’s preliminary findings put the social cost of carbon at more than $100, with a 2% discount rate. Just as people buy insurance against catastrophes like their house burning down, CIL finds evidence they would pay to avoid the small probability of climate catastrophes, such as a six-degree rise in temperatures. This insurance premium boosts the social cost of carbon above $200.
Neither RFF nor CIL set out to put a higher price on climate damage. That their findings do just that matters for the net zero debate. The International Monetary Fund estimates if the whole world put an average $75 price on carbon (such as through a carbon tax) by 2030, it would be on the path toward holding the temperature increase to below 2 degrees. The 38-nation Organization for Economic Cooperation and Development suggests $140 would achieve net zero. These numbers used to be far above the expected damage from higher temperatures, and thus hard to justify, economically. As the social cost of carbon gets revised up, they start to look more economically sound.