What We Publish

Country-level social cost of carbon

Katharine Ricke, Laurent Drouet, Ken Caldeira, and Massimo Tavoni · Nature Climate Change 8, 895-900 · 2018

Key finding. The central estimate of the global social cost of carbon is a median of 417 US dollars per tonne of carbon dioxide (66 percent confidence interval 177 to 805 dollars), of which India alone bears about 21 percent, at 86 dollars per tonne — the largest share of any country, ahead of the United States at 48 dollars and Saudi Arabia at 47 dollars.

Panel a is a pair of world maps shading each country's social cost of carbon in US dollars per tonne of carbon dioxide, from below minus 10 through to 50-100, with India, the United States, Saudi Arabia and Brazil in the darkest bands and much of the high latitudes negative or near zero. Panel b is a dot plot of country-level social cost of carbon for ten selected countries across five socioeconomic and emissions scenarios and four damage-function specifications; India spans the widest range, reaching beyond 400 dollars.
The global social cost of carbon is an average over a highly unequal distribution. India, the United States and Saudi Arabia carry the largest country-level shares (panel a), and the spread across scenarios and damage functions (panel b) is widest for exactly the country bearing most of the cost. Figure 2 from Ricke et al. (2018), Nature Climate Change 8, 895-900. Reproduced under author reuse rights. Extracted from the published PDF and resized for web display.

What question did this research address?

The social cost of carbon is the economic damage caused by emitting one additional tonne of carbon dioxide. It is a global number by construction, because a tonne emitted anywhere warms everywhere — but the damage it causes lands in particular countries, and the global total says nothing about where.

Knowing the country-level distribution matters for two separate reasons. It shows who bears the cost of climate change, and it shows what each country stands to gain from reducing emissions on its own — which is what determines whether unilateral action is in a country's narrow self-interest.

This work asked what the social cost of carbon looks like when it is decomposed by country, and how robust that decomposition is to the many contested assumptions that go into calculating it.

What did we find?

The central specification gives a median global social cost of carbon of 417 US dollars per tonne of carbon dioxide, with a 66 percent confidence interval of 177 to 805 dollars. That is far above the figures then used in United States rule-making, which ranged from 12 to 62 dollars per tonne depending on the discount rate.

The country-level costs are highly unequal. India's share is the largest at 86 dollars per tonne (49 to 157), some 21 percent of the global total, followed by the United States at 48 dollars (1 to 118) and Saudi Arabia at 47 dollars (27 to 86). Brazil, China, and the United Arab Emirates follow at about 24 dollars per tonne each.

The absolute numbers depend heavily on contested assumptions — the discount rate, the socio-economic scenario, and the form of the damage function — but the ranking of countries does not. The same countries incur large fractions of the global cost across specifications.

The method deliberately avoids reduced-form aggregate modelling. Country-level climate projections were taken directly from gridded climate model output, and country-level damage relationships directly from empirical macroeconomic analyses, rather than being inferred from a globally aggregated model.

Why does it matter?

A country that bears a fifth of global climate damage has a substantial unilateral economic incentive to reduce emissions, quite apart from any international agreement. Quantifying that incentive country by country changes how the negotiation problem looks.

The estimated global cost is several times the values embedded in regulatory practice at the time, which means benefit-cost tests for emissions-reducing policy were being run against a number far below this study's central estimate.

Because the country ranking is robust while the absolute level is not, the distributional result is the more usable one — it survives disagreement about discounting that the headline number does not.

Citation

Katharine Ricke, Laurent Drouet, Ken Caldeira, and Massimo Tavoni (2018). Country-level social cost of carbon. Nature Climate Change 8, 895-900.

Related