I was reading a policy proposal that referenced a social cost of carbon analysis to justify new regulations. The number they used seemed incredibly high, and it got me wondering—how do economists even put a dollar value on future climate damage, and why do the estimates vary so wildly between different studies?
Social cost of carbon is the present value of the damages from one extra ton of CO2 emitted today. Economists estimate it by looping climate projections into economic impacts and discounting future harms back to the present. The exact figures swing a lot because of what you include, whether you measure globally or domestically, and what discount rate you apply. Historically, the Obama administration used a global SCC around $43/ton; the Trump administration focused on domestic effects, around $3–5/ton; and the Biden era has toyed with higher figures when broader damages and inflation are included. citeturn0search0turn0search3
Most of the variation comes from the models that generate SCC values, known as integrated assessment models like DICE, RICE, and PAGE. They couple a climate module with an economic module and use a damage function to translate climate changes into dollars. Different choices about climate sensitivity, damage functions, and the reachable time horizon can flip the numbers a lot. citeturn1academia13turn0search3
As of January 2025, the Interagency Working Group on the Social Cost of Greenhouse Gases was disbanded, and official guidance relying on the IWG’s numbers was withdrawn. That means there isn’t a single federal SCC figure guiding all analyses anymore; agencies may still use the same methodology but under different oversight, or rely on new guidelines. citeturn0search2
There's also the tail risk argument: if you worry about rare but catastrophic outcomes, some economists argue the SCC should be much higher; Weitzman and others have written about fat tails and how discounting interacts with risk. citeturn1search0
So, in practice: treat SCC as a range, not a single number; compare low/high discount rates (2.5% to 5%), domestic vs global, and include a sensitivity analysis; also check if the underlying model includes or excludes catastrophic damages. citeturn0search3
If you want, tell me the policy context and region, and I can help you map a quick sensitivity plan to sanity-check the numbers.