How Do Environmental Economists Know that the Social Cost of Carbon is $42 a Ton?

The National Academy of Sciences has released a brand new report that you can download here that has a definitive feel on stating the academic environmental economics’ community consensus on the social cost of carbon.  In this blog post, I will clearly state that I do not believe this number nor do I believe that such a number can be estimated.  It is a moving target that changes every day and it may be actually declining over time.  My recent NBER Paper offers one induced innovation theory for why this can happen.

 Here is the citation;

Suggested citation: National Academies of Sciences, Engineering, and Medicine. 2017. 37 Valuing Climate Damages: Updating Estimation of the Social Cost of Carbon Dioxide. 38 Washington, DC: The National Academies Press. doi: 10.17226/24651.

I will directly quote from the Executive Summary and then make some points.

DIRECT QUOTE

The social cost of carbon (SC-CO2) is an economic metric intended to provide a 4 comprehensive estimate of the net damages—that is, the monetized value of the net 5 impacts, both negative and positive—from the global climate change that results from a 6 small (1-metric ton) increase in carbon-dioxide (CO2) emissions. Under Executive Orders 7 regarding regulatory impact analysis and as required by a court ruling, the U.S. 8 government has since 2008 used estimates of the SC-CO2 in federal rulemakings to value 9 the costs and benefits associated with changes in CO2 emissions. In 2010, the Interagency 10 Working Group on the Social Cost of Greenhouse Gases (IWG) developed a 11 methodology for estimating the SC-CO2 across a range of assumptions about future 12 socioeconomic and physical earth systems. 13 The IWG asked the National Academies of Sciences, Engineering, and Medicine 14 to examine potential approaches, along with their relative merits and challenges, for a 15 comprehensive update to the current methodology. The task was to ensure that the SC- 16 CO2 estimates reflect the best available science, focusing on issues related to the choice 17 of models and damage functions, climate science modeling assumptions, socioeconomic 18 and emissions scenarios, presentation of uncertainty, and discounting. 19 Integrated assessment models (IAMs) are currently used by the IWG to estimate 20 the economic consequences of CO2 emissions. The IAMs define baseline emission 21 trajectories by projecting future economic growth, population, and technological change. 22 In these IAMs, a 1-metric ton increase in CO2 emissions is added to the baseline 23 emissions trajectory. This emissions increase is translated into an increase in
atmospheric CO2 concentrations, which results in an increase in global average 25 temperature. This temperature change, as well as changes in other relevant variables, 26 including CO2 concentrations and income, is translated (either explicitly or implicitly) to 27 physical impacts and monetized damages. These damages include, but are not limited to, 28 market damages, such as changes in net agricultural productivity, energy use, and 29 property damage from increased flood risk, as well as nonmarket damages, such as those 30 to human health and to the services that natural ecosystems provide to society. Because 31 most of the warming caused by an emission of CO2 into the atmosphere persists for well 32 over a millennium, changes in CO2 emissions today may affect economic outcomes for 33 centuries to come. Streams of monetized damages over time are converted into present 34 value terms by discounting. The present value of damages reflects society’s willingness 35 to trade value in the future for value today. 36 The IWG methodology combines tens of thousands of SC-CO2 results obtained 37 from running three IAMs using five different socioeconomic and emissions projections, a 38 common distribution of equilibrium climate sensitivity (a parameter that characterizes the 39 relationship between CO2 concentrations and long-term global average temperature 40 change), and distributions for other parameters. These results yield three distributions of 41 SC-CO2 values for three different discount rates, from which the IWG calculated an 42 average value for each discount rate. The IWG’s current estimate of the SC-CO2 in the 43 year 2020 for a 3.0 percent discount rate is $42 per metric ton of CO2 emissions in 2007 44 U.S. dollars. If, for example, a particular regulation was projected to reduce CO2 45 emissions by 1 million metric tons in 2020, the estimate of the value of its CO2 emissions 46 benefits in 2020 for this SC-CO2 would be $42 million dollars.

1.  So,  Robert Pindyck’s warning about not relying on CGE models have been ignored and the NAS has double down.   Here is a quote from Dr. Pindyck

Climate Change Policy: What Do the Models Tell Us?

Robert S. Pindyck

Very little. A plethora of integrated assessment models (IAMs) have been constructed and used to estimate the social cost of carbon (SCC) and evaluate alternative abatement policies. These models have crucial flaws that make them close to useless as tools for policy analysis: certain inputs (e.g., the discount rate) are arbitrary, but have huge effects on the SCC estimates the models produce; the models’ descriptions of the impact of climate change are completely ad hoc, with no theoretical or empirical foundation; and the models can tell us nothing about the most important driver of the SCC, the possibility of a catastrophic climate outcome. IAM-based analyses of climate policy create a perception of knowledge and precision, but that perception is illusory and misleading. ( JEL C51, Q54, Q

2. The word adaptation appears 68 times in this 400 page document but it is never quantified.  This entire document is subject to the Lucas Critique again.   The Lucas Critique from 1976 argues that economic agents reoptimize as the “Rules of the Game change”.  As Mother Nature changes the climate conditions, risk averse forward looking people will seek out “higher ground” and safety.  This is adaptation.

As I explain in this paper, here is the core issue;  The mathematical modelers who built the core model used in the NAS study must take a strong stand in the year 2017 that they know the following functions;

A.  What will the world’s population and per-capita income be each year in the future?
B.  How population and per-capita income together determine aggregate greenhouse gas emissions and thus future atmospheric carbon dioxide concentrations.
C.  how rising carbon dioxide concentrations affect the population’s well being and per-capita income.

DO you see that this is an impossible prediction exercise?  Yet, the modelers press ahead under the assumption that they can model each of these pieces.  The confidence intervals on each of these equations would be enormous.

This last function (called “C” above) is the “damage function”.  As I have argued here, here and here, the spatial economy is constantly reorganizing driving this slope of   the damage with respect to CO2 concentration function closer and closer to zero.  This is the algebra of adaptation.

To provide one example;  this peer reviewed study documents that deaths from heat waves are falling sharply over the 20th century in the U.S as air conditioning diffuses widely. This is adaptation and this is missing in all of the calculations listed above because the flow of technology cannot be predicted ex-ante.

Barreca A, Clay K, Deschenes O, Greenstone M, Shapiro JS. Adapting to climate change: The remarkable decline in the US temperature-mortality relationship over the twentieth century. Journal of Political Economy. 2016 Feb 1;124(1):105-59.

I am a pinch sad that my Climatopolis work isn’t cited.    The world is moving to cities and this migration boosts income and helps to shield the urbanites from the “social costs of carbon”.  The mathematical models discussed in this long report do not allow for detailed spatial disaggregation to allow capital and labor to move to “higher ground” such micro adjustment margins are at the heart of my optimism that the social cost of carbon is actually declining over time.

Read my Climatopolis Revisited.

Now, I do agree with the report’s authors on one important point.  Holding constant the global atmospheric carbon dioxide level, we grow better at adapting to this challenge each day.  But, at the same time that our adaptation resilience rises, the stock of GHG emissions increase and this means that the adaptation challenge grows more challenging.

By relying on “1970s style” pre-Lucas Critique IAM models,  this report doesn’t feel like it embodies frontier academic macroeconomic logic.  It would interest me if any academic macro economists reviewed the piece or would be comfortable with its claims.

None of the following reviewers  are academic macro-economists.

We thank the following individuals for their participation in the review of this report: 182 Hadi Dowlatabadi, Institute for Resources Environment and Sustainability, University of 183 British Columbia; 184 James (Jae) Edmonds, Joint Global Change Research Institute, Pacific Northwest 185 National Laboratory; Copyright © National Academy of Sciences. All rights reserved. Valuing Climate Damages: Updating Estimation of the Social Cost of Carbon Dioxide ix Prepublication Copy Uncorrected Proofs 186 Karen Fisher-Vanden, Environmental and Resource Economics, The Pennsylvania State 187 University; 188 Michael Greenstone, Energy Policy Institute at Chicago and Department of Economics, 189 University of Chicago; 190 Anthony C. Janetos, The Frederick S. Pardee Center for the Study of the Longer-Range 191 Future, Boston University; 192 Peter B. Kelemen, Department of Earth and Environmental Sciences, Columbia 193 University, and Lamont-Doherty Earth Observatory; 194 Bryan K. Mignone, Corporate Strategic Research, ExxonMobil Research and 195 Engineering Company; 196 Richard H. Moss, Joint Global Change Research Institute, University of Maryland; 197 Elisabeth Moyer, Department of Geophysical Sciences, University of Chicago; 198 Richard L. Revesz, New York University School of Law (emeritus); 199 David A. Weisbach, Law School and Computation Institute, University of Chicago, and 200 Argonne National Laboratories; 201 Jonathan B. Wiener, Law, Environmental Policy, and Public Policy Law School, 202 Nicholas School of the Environment, and Sanford School of Public Policy, Duke 203 University; and 204 Gary W. Yohe, Economics and Environmental Studies, Wesleyan University.