In response to my recent post about the EPA’s little-known social cost of carbon (SCC) value and its importance in setting the stringency of U.S. emission reductions measures, one reader posted this comment:

… It seems to me that the social cost of carbon is whatever it is. The issue is how close EPA gets their estimate to the true value…

Perhaps, but the EPA’s current efforts at calculating the SCC – with its strong bias towards the lower end of estimations from the climate economics literature – is unlikely to arrive at that putative “true value.”

Here’s another way to calculate the SCC: We could ask ourselves, in general terms, what payment would we accept in exchange for our permission to allow greenhouse gas emissions to continue to grow, and to accept a drastically changed climate and all the social and economic damages that would come with it?

Before you answer, here’s something to consider: If we allow emissions to continue, we know that the climatic changes will be profound and the damages serious, but we don’t really know how profound and how serious.

This is what climate scientists and economists call the “problem of uncertainty.” We have a good idea of what the most likely damages will be, and even a good idea about their lower bound (best case). But the upper bound (worst case) is almost impossible to imagine, let alone quantify, with any confidence. In other words, it’s a big gamble: What would you accept in exchange for my assurance that the damages probably will be difficult but not devastating?

If your answer is some variation on “not for all the money in the world!” then for you (and for me) the SCC is infinite.

The SCC answers the question: What’s the damage done by one more ton of CO2 emitted into the atmosphere? Or, what benefit would make it worth it to you to allow the damage caused by that one additional ton? An infinite SCC tells us that future damages are so great that any additional emissions are simply unacceptable.

For anyone out there who likes to think graphically: The idea presented here is that part of the SCC curve is vertical. For a somewhat technical discussion of these issues see my critique of the now-defunct British method of calculating the SCC.