The Cost of Carbon Revisited

In 2010, an inter-agency task force provided a series of estimates of the “social cost of carbon” to guide government cost-benefit analyses. The estimates vary with the discount rate and the date.  For instance, using a 5% discount rate, it would be worth spending hardly anything — only $4.70 — to eliminate a ton of CO2 in 2015.  On the other hand, with a 3% discount rate, the amount rises to $23.80, and at 2.5%, it rises again to $38.40.  An alternative estimate, intended to give more weight to possibility of more extreme impacts, was $72.80 (at a 3% discount rate).

The government has now updated the estimates.  Under the new estimate, using a 5% discount rate, it would be now worth spending over twice as much — $12 instead of  $4.70 — to eliminate a ton of CO2 in 2015.  With a 3% discount rate, the amount rises to $38 (instead of $23.80), and at 2.5%, it rises again to $58 (instead of $38.40).  The alternative estimate, giving more weight to possible extreme impacts, is now $109 instead of $72.80 (at a 3% discount rate).  The estimates also rise quickly over time — by 2050 they more than double.

The estimates still leave a lot to be desired.  They are derived from averaging the results of three very different models. The reason for using those three is that they happen to be in use already by economists.  There’s no effort to decide which models are better or to create a better model.  The new estimates are simply based on updates to the models by their creators, not any new research by the government.  Any cost-benefit analysis of climate change is going to have great uncertainties, but the federal government really should be able to do better than this.  It’s like making dinner by mixing a can of chili, a can of cream of mushroom, and a can of peaches, just because they happen to be what you grab quickly from the shelf.

, ,

Reader Comments

6 Replies to “The Cost of Carbon Revisited”

  1. Dan,

    I respectfully disagree for two reasons. First. PAGE, DICE, and FUND are not chili, mushroom, and peaches. They are attempts by three of the very best economic modelling groups to create the best integrated assessment model possible using relatively similar approaches. These groups work collaboratively with each other through the EMF – all of them actively critique each other’s models and incorporate improvements based on these critiques. This effort is analogous to what the climate modelling community does with its model intercomparison work which is one of the key products of IPCC WG1. Noone has a problem when people compare three models of climate that have different specifications for cloud physics – and there is a lot of benefit in looking at the places where models agree. Why not also take this approach with economic modeling of climate change?

    Second, after reading through the new SCC document, I’m actually pretty impressed by the overall transparency and rigor of what the interagency working group has done. Is it perfect? No. But is it much better than the usual ways in which costs of pollution are done in CBA – yes, and in a number of respects. (1) transparency, (2) use of Monte Carlo simulation to assess model sensitivity and distribution of impacts, (3) use of multiple models (ie cost estimates) and multiple scenarios (forecasts of future behavior), (4) use of models that are publicly available – anybody can go get these models off the web and try to reproduce the results or modify the assumptions used.

    Is it the best of all possible worlds – perhaps not – there is always room to improve. But my view is that the SCC work is a lot better than what passes for CBA in other environmental areas or elsewhere in the administrative state.

    Best,
    MW

    1. Mike,

      It seems to me that the government’s effort in this area does not compare well with the depth of analysis that EPA usually devotes to scientific and economic issues. It would be very surprising to find an EPA risk assessment that simply averaged all the reported studies without any attempt to consider their assumptions, strengths, and weaknesses. In addition, the current IAM models strike me as incredibly crude — the damage functions, in particular, are very weakly grounded and probably meaningless for higher temperature levels. For instance, some models assign very low values to costs that get very high values in other models. Climate models are better grounded, but no one thinks you should make policy based on the arithmetic means of whatever models happen to exist anywhere in the world.

      Dan

  2. Dan,

    I respectfully disagree for two reasons. First. PAGE, DICE, and FUND are not chili, mushroom, and peaches. They are attempts by three of the very best economic modelling groups to create the best integrated assessment model possible using relatively similar approaches. These groups work collaboratively with each other through the EMF – all of them actively critique each other’s models and incorporate improvements based on these critiques. This effort is analogous to what the climate modelling community does with its model intercomparison work which is one of the key products of IPCC WG1. Noone has a problem when people compare three models of climate that have different specifications for cloud physics – and there is a lot of benefit in looking at the places where models agree. Why not also take this approach with economic modeling of climate change?

    Second, after reading through the new SCC document, I’m actually pretty impressed by the overall transparency and rigor of what the interagency working group has done. Is it perfect? No. But is it much better than the usual ways in which costs of pollution are done in CBA – yes, and in a number of respects. (1) transparency, (2) use of Monte Carlo simulation to assess model sensitivity and distribution of impacts, (3) use of multiple models (ie cost estimates) and multiple scenarios (forecasts of future behavior), (4) use of models that are publicly available – anybody can go get these models off the web and try to reproduce the results or modify the assumptions used.

    Is it the best of all possible worlds – perhaps not – there is always room to improve. But my view is that the SCC work is a lot better than what passes for CBA in other environmental areas or elsewhere in the administrative state.

    Best,
    MW

    1. Mike,

      It seems to me that the government’s effort in this area does not compare well with the depth of analysis that EPA usually devotes to scientific and economic issues. It would be very surprising to find an EPA risk assessment that simply averaged all the reported studies without any attempt to consider their assumptions, strengths, and weaknesses. In addition, the current IAM models strike me as incredibly crude — the damage functions, in particular, are very weakly grounded and probably meaningless for higher temperature levels. For instance, some models assign very low values to costs that get very high values in other models. Climate models are better grounded, but no one thinks you should make policy based on the arithmetic means of whatever models happen to exist anywhere in the world.

      Dan

  3. Dan said;
    “…For instance, using a 5% discount rate, it would be worth spending hardly anything — only $4.70 — to eliminate a ton of CO2 in 2015…”

    Why would anyone spend $4.70 to eliminate a ton of CO2 if this could be done for less than $1.00 ? The cost of carbon dioxide mitigation could be substantially reduced with free markets, and global competition. It may be technically feasible to mitigate carbon dioxide for less than $0.10 per ton – this would be good news for those compelled to mitigate.

  4. Dan said;
    “…For instance, using a 5% discount rate, it would be worth spending hardly anything — only $4.70 — to eliminate a ton of CO2 in 2015…”

    Why would anyone spend $4.70 to eliminate a ton of CO2 if this could be done for less than $1.00 ? The cost of carbon dioxide mitigation could be substantially reduced with free markets, and global competition. It may be technically feasible to mitigate carbon dioxide for less than $0.10 per ton – this would be good news for those compelled to mitigate.

Comments are closed.

About Dan

Dan Farber

Dan Farber has written and taught on environmental and constitutional law as well as about contracts, jurisprudence and legislation. Currently at Berkeley Law, he has al…

READ more

POSTS BY Dan