The Social Cost of Carbon Initiative
Researchers in this initiative are leading a team of distinguished economists and scientists to improve the science behind estimates of the social cost of carbon—the means by which the US federal government, state governments, and foreign governments account for climate change in their actions—through a process that ensures the highest levels of scientific quality and transparency and builds the scientific foundation for future estimates.
This page provides a detailed description of current research within RFF's Social Cost of Carbon Initiative. For a refresher on what the Social Cost Carbon is, please see our SCC explainer.
Work within RFF’s Social Cost of Carbon Initiative falls into three broad categories:
- Creating an open source computing platform for Integrated Assessment Modeling
- Providing scientific updates across each of the four core steps of estimating the social cost of carbon
- Facilitating more informed policy making world-wide
Read below for more information on RFF's research in each of these areas.
Creating an open source computing platform for Integrated Assessment Modeling
A core recommendation from the National Academies of Sciences (NAS) report for improving the Social Cost of Carbon was to create an open source computing platform to improve the transparency of and foster collaboration among the climate impacts modeling community. RFF is working with a team of researchers at UC Berkeley lead by David Anthoff to continue developing Mimi.jl, a software package for creating, running, and performing analyses on Integrated Assessment Models. Mimi.jl provides an easy-to-use interface for defining components and building models in a modularized, transparent way. It is implemented in the Julia programming language, which is computationally fast while maintaining accessible syntax and conventions for novice programmers.
There is also a user forum that is actively monitored to answer any questions from users, which can be found at forum.mimiframework.org
Providing scientific updates across each of the four core steps of estimating the social cost of carbon
Building a new set of long-run projections of Economic Growth, Population, and Emissions
For each of the following variables, RFF is working to produce central projections out to the year 2300, as well as complete probability distributions around the central projections to characterize their uncertainty.
- Economic Growth: RFF is conducting formal expert elicitations of leading growth economists to generate long-run projections of economic growth. The results from these elicitations will be used in concert with new research from Ulrich Müller and Mark Watson (Princeton University) and James Stock (Harvard Kennedy School), who have refined a foundational statistical methodology for generating long-run projections of economic growth at the country level.
- Population: RFF is collaborating with Adrian Raftery and Hana Ševčíková (University of Washington) to generate extended, country-level population projections. Raftery and Ševčíková’s Bayesian methodology forms the foundation of the United Nations’s official population projections.
- Emissions: To calculate future trajectories of CO2 emissions, RFF is creating probability distributions of future energy intensity of the economy (energy/GDP) and carbon intensity of energy (CO2/energy).
- The Müller-Stock-Watson team that are modeling long-run economic growth are also using the same methodology to generate projections of energy per unit of GDP.
- RFF researchers are collaborating with Valentina Bosetti of the RFF-CMCC European Institute on Economics and the Environment to generate a range of projections of emissions per unit of energy based upon the formal elicitation of experts.
Updating the climate model
RFF will be using an updated simplified climate model such as FAIR or SNEASY to model the temperature changes and other earth system responses under different emissions trajectories. Both of these models have been implemented on the Mimi.jl platform and are responsive to the recommendations of the NAS for updating the climate model used in SCC estimation. Because the Integrated Assessment Model will be run hundreds of thousands of times in Monte Carlo Simulations in order to span the uncertainty of key input parameters, it is critical that the climate model runs quickly, while still satisfying key diagnostics of accuracy in comparison to more complex general circulation models.
Building new damage functions from the best available literature
RFF is working to produce a new set of updated damage functions for key sectors of society that will be impacted by climate change based on the current state of the science in the published academic literature. The NAS noted that many of the damage functions in the integrated assessment models used to generate the current US federal estimates of the SCC do not reflect the explosion of climate impacts research that has occurred in recent years, so incorporating such updates is a critical step in improving the estimates overall.
RFF is implementing a Ramsey-like discounting framework to better model the uncertainty of future discount rates as related to the uncertainty in economic growth. President and CEO Richard Newell and fellows Brian Prest and Billy Pizer are developing a new method for estimating the relevant discounting parameters—the pure rate of time preference and the elasticity of the marginal utility of consumption—from evidence on the long-run term structure of interest rates and the projected distributions of future economic growth rates.
Facilitating more informed policy making worldwide
RFF is engaged with decisionmakers in various levels of government and the private sector to provide technical assistance related to the Social Cost of Carbon. Among the groups we have consulted with are the US Climate Alliance, the California Air Resources Board, the New York State Energy Research and Development Authority, and Environment Canada.