Evaluating the Social Cost of Carbon: Implications for Policy
Introduction
Under President Biden’s administration, the Environmental Protection Agency (EPA) revived the concept of the Social Cost of Carbon (SCC), a metric designed to estimate the economic impact of carbon dioxide (CO2) emissions. To calculate the SCC, the administration utilized three Integrated Assessment Models (IAMs): the Data-driven Spatial Climate Impact Model (DSCIM), the Greenhouse Gas Impact Value Estimator (GIVE), and the Howard and Sterner meta-analysis models. These models have significant implications for regulatory decisions affecting various everyday products, from automobiles to household appliances.
Understanding Integrated Assessment Models and the Social Cost of Carbon
Integrated Assessment Models (IAMs) serve to quantify the SCC by analyzing the interconnectedness of the economy, society, and environment. They utilize “damage functions” to project economic losses attributed to increased temperatures, thereby facilitating cost-benefit analyses. Monte Carlo simulations introduce uncertainty by running the model multiple times with varied inputs to generate a spectrum of possible SCC outcomes. However, the high sensitivity of IAMs to key assumptions raises concerns about their efficacy and reliability.
During the Obama administration, the SCC was pivotal in climate policy, with estimates ranging from $26 to $95 per metric ton of CO2 emissions for 2050. The Trump administration disbanded the Interagency Working Group that developed these estimates and shifted the focus to domestic rather than global effects. The Biden administration subsequently reinstated the SCC framework with the introduction of new models, including DSCIM.
The DSCIM Model in Focus
Originating from the University of Chicago’s Climate Impact Lab, the DSCIM assesses the SCC based on damages across five categories: health, energy, labor productivity, agriculture, and coastal regions. The model generates outcomes through a four-component structure consisting of projections for socioeconomic growth and greenhouse gas emissions, climate modeling, damage functions, and discounting.
Monte Carlo simulations—executed over 10,000 iterations—enable the model to account for uncertainties in variables such as equilibrium climate sensitivity and economic indicators. However, the underlying assumptions of DSCIM warrant scrutiny. This article examines key areas of sensitivity: discount rates, time horizons, and climate sensitivity.
Discount Rates
The SCC relies heavily on discounting, a financial concept vital for comparing future climate costs and benefits. Different discount rates significantly alter the present value of future CO2 reduction benefits. The Biden administration’s updated policy implements a central discount rate of 2%, alongside the consideration of declining rates to alleviate long-term uncertainties. The OMB estimated the SCC under various rates, producing mean SCC estimates for 2030 that illustrate considerable variability based on the chosen discount rate.
Re-evaluating the Time Horizon
The DSCIM projects damages extending nearly 300 years into the future, a timeframe filled with uncertainties. Adjusting this horizon to 150 years yields significantly reduced SCC estimates, indicating that duration plays a critical role in economic assessments related to climate impacts.
Equilibrium Climate Sensitivity
Equilibrium climate sensitivity (ECS) distributions, which gauge temperature response to CO2 doubling, are central to SCC calculations. Although the EPA’s mean ECS of 3.18°C aligns with major climate assessments, empirical data suggests existing models may overestimate warming. Different ECS distributions yield vastly divergent SCC estimates, further highlighting the importance of choosing the right parameters.
The Concept of Negative SCC
Interestingly, the DSCIM also acknowledges scenarios where CO2 emissions might yield positive environmental effects, such as enhanced agricultural yields. Current findings suggest a non-negligible probability for a negative SCC—where the benefits of emissions may outweigh the costs—under certain ECS conditions. This raises questions regarding the rationale behind carbon taxation.
Policy Recommendations and Conclusion
The sensitivity of the SCC models to user-defined parameters suggests that regulatory decisions can be manipulated based on selected assumptions. This reality highlights the necessity for greater transparency and reproducibility in regulatory analyses, particularly regarding the SCC.
- Federal agencies should be mandated to maintain and openly share the full suite of models and codes used in deriving the SCC, enabling independent scrutiny and accountability.
- Legislative measures should be pursued to prevent potential reinstatement of the SCC framework by future administrations.
In conclusion, while models like DSCIM aim to quantify the economic impacts of climate change, their inherent sensitivity to assumptions presents challenges for sound policymaking. A more robust and transparent approach is essential for fair and effective regulatory decision-making.
