The Social Cost of Carbon in Regulatory Analysis

Main Content

A workshop co-sponsored by the Nicholas Institute for Environmental Policy Solutions and the Center for Law, Economics, and Public Policy
Duke University, May 21, 2014
Location: Duke Law School

How should the U.S. government continue to update and revise the social cost of carbon in the future?

The social cost of carbon (SCC) is currently used in a variety of regulatory impact analyses throughout the U.S. government (USG) to value the climate change impacts of various regulations.  In 2013, the U.S. government released an update to its previous, 2010 estimate.  This estimate, based on updates to the models used by the government in its 2010 analysis, raised the estimates by 40 percent and provoked a variety of questions and challenges mainly related to the process of establishing the SCC. 

This one-day workshop will examine the various questions the government faces when making use of the scientific literature on the SCC in order to create and maintain an estimate for regulatory purposes.  The purpose is to examine whether a consensus exists among the workshop participants on recommendations for improving the SCC process in the future.

Duke Law School, May 21, 2014

Agenda and charge questions


Coffee and breakfast


Overview and background


How should the social cost of carbon be measured?


The current government SCC specifies (a) five population, economic, and emission forecasts; (b) a distribution for  “climate sensitivity” (the temperature change from doubling CO2 concentrations); and (c) three constant, annual discount rates.  These input specifications were agreed within the USG through an interagency process, and then provided to modelers to produce SCC estimates.  The primary output of the exercise are four schedules of estimates over time.  The first three use each of the different discount rates, but otherwise average over both the above specifications and other model-specific uncertainty; the fourth is based on a 95th percentile using a 3 percent (the middle value) discount rate.

Is this the right way for the government to specify scenarios and present results?  Does the SCC constructed this way represent the appropriate concept for government regulatory impact analyses?  Does it create problems for models (e.g., inconsistency between discount rates and economic growth rates)?  What are other alternatives?


Initial remarks:  Rich Richels, Electric Power Research Institute and Ken Gillingham, Yale University.


Moderator:  Brian Murray, Duke University




What are the biggest challenges in modeling and valuing specific climate change impacts, and how are / can / should models and their valuations be validated?

The current government SCC is based on an average of three integrated assessment models (IAMs):  FUND, DICE, and PAGE.  Each of these models attempts to incorporate state of the art information on relations between emissions and climate change, climate change and various ecological and human impacts, and valuation of those impacts in monetary terms.

What are the most challenging and/or controversial elements of the IAM effort?  How are models currently peer-reviewed and/or validated?  Are there alternatives to the current peer-review / validation process; what are possible advantages and disadvantages?  Where could additional research usefully focus in the future? 


Initial remarks:  David Anthoff, University of California-Berkeley and Stephanie Waldhoff, Pacific Northwest National Lab


Moderator:  Billy Pizer, Duke University




How should the SCC address impacts across jurisdictions?


Most regulatory impact analysis focuses on impacts within the U.S., while the SCC emphasizes global estimates.  SCC estimates generally assume that a dollar impact is equivalent across countries.

Is the current SCC focus on global estimates appropriate and justified?  Should effects be valued the same regardless of whether they are in the U.S. or not?


Initial remarks:  Arden Rowell, University of Illinois


Moderator:  Matt Adler, Duke University




How should the government update estimates over time?

The 2013 revision to the SCC was based on more recent versions of the original models used in the 2010 analysis.  The updates were largely based on the existence of newer, peer-reviewed publications.

What should the basis be for updates of the SCC in the future?  How / should new models enter the calculation?  What is the role, if any, for an external advisory committee?  What agency(s) should lead the process, and should the value continue to be consistent across agencies? 


Initial remarks:  Joe Aldy, Harvard University; Maureen Cropper, University of Maryland; Michael Greenstone, MIT


Moderator:  Jonathan Wiener, Duke University


Concluding Remarks


Moderator:  Billy Pizer, Duke University




Registration is closed.


On Wednesday, May 21st, there will be parking for the conference in the visitor lot across the street on Science Drive, between NC 751 and Whitford Drive. Additional parking will be available in Parking Garage 4.  Additional parking information can be found here.



Victoria Zellefrow