« Back to Results

Theoretical Advances in Climate Change Economics

Paper Session

Friday, Jan. 3, 2025 2:30 PM - 4:30 PM (PST)

Parc 55, Divisadero
Hosted By: Association of Environmental and Resource Economists
  • Chair: Christian Traeger, University of Oslo

Considering Robustness to Deep Uncertainties Drives More Rapid Emissions Reductions

Lisa Rennels
,
University of California-Berkeley
Frank Errickson
,
Princeton University
David Smith
,
U.S. Environmental Protection Agency
Bryan Parthum
,
U.S. Environmental Protection Agency
Klaus Keller
,
Dartmouth College

Abstract

Evaluating the economic impacts of climate change is crucial to inform climate policy. One typical approach to assessing mitigation policy options uses integrated climate-economy models to analyze tradeoffs between the costs of reducing greenhouse gas emissions and the benefits of avoiding climate damages. The deep uncertainty characterizing these models poses challenges for policymakers. We address this challenge using a robust decision-making framework to evaluate mitigation policy. We show that a shift from a decision framework that maximizes expected outcomes to one that is averse to regret supports precaution in the face of uncertainty and faster emissions cuts than currently implemented. Uncertainties about socioeconomic trajectories and the magnitude and functional form of climate damages create asymmetric consequences from delayed or weak mitigation policy.

Dynamic Climate Adaptation

Christopher Costello
,
University of California-Santa Barbara
Sam Collie
,
Natural Capital Consulting

Abstract

Managing natural resources like fisheries, forests, and ecosystems requires making decisions that have dynamic consequences. Climate change, defined as a shift in the distribution of weather experienced over time, is unequivocally dynamic and affects natural resource dynamics as it unfolds. How, then, should rational agents adapt to climate change in a dynamic setting? We provide a constructive theory that formalizes the role of adaptation, and its value, in dynamic decision environments. We show that policy functions derived to be optimal in dynamic settings without climate change are remarkably robust to climate change, so further adapting to climate change delivers only modest benefits. We illustrate this theory with a global dataset of fisheries where we find that perfect adaptation to climate change increases the value of fisheries only marginally (typically <1%) relative to optimized management without climate change. This surprising result arises despite the fact that climate change is having substantial impacts on fisheries productivity.

Structure, Shocks, and Speed: Learning’s Impact on Optimal Climate Policy

Christian Traeger
,
University of Oslo
Svenn Jensen
,
Oslo MET

Abstract

We investigate the impact of learning on the formation of optimal economic policy, with a particular emphasis on climate policy. Dynamic economic models dealing with uncertainty inherently rely on assumptions about how agents anticipate and adapt to new information. We find that seemingly similar 'no learning' approaches can result in widely divergent risk premiums applied to policy decisions. Specifically, our study focuses on the uncertain factor of climate sensitivity - the degree to which our planet's temperature responds to the accumulation of greenhouse gases over medium to long-term periods. We carefully distinguish three components of uncertainty: natural temperature variability, measurement error, and subjective uncertainty governing unknown model parameters. Even though learning reduces uncertainty over time, our analysis using a version of Nordhaus’ Dynamic Integrated Model of Climate and the Economy (DICE) framework reveals a paradox: accelerated Bayesian learning can, in fact, increase the risk premium on the optimal carbon tax. We provide an analytic formula for optimal carbon pricing under anticipated Bayesian learning, and we offer an in-depth discussion of the mechanisms through which uncertainty and learning interplay to influence policy formulation. We illustrate how the speed of information acquisition affects policy risk premiums through different channels. Our findings are quantified within a DICE-based recursive stochastic dynamic programming model, providing insights that challenge traditional notions of learning and policy-making in the context of environmental economics.

Tiered Climate Clubs: Global Abatement Without Global Agreement

Terrance Iverson
,
Colorado State University

Abstract

The paper proposes a novel policy structure with the potential to effectively reduce global carbon emissions without the need for broad multilateral cooperation. It extends Nordhaus’s (2015) climate club by adding a second tier. Countries in the second-tier price carbon at a fixed fraction (the “match rate”) of the average carbon price adopted within the first tier, or face tariffs. Tier-one countries abate more since doing so induces matching abatement elsewhere. The analytical section derives closed-form expressions for the optimal carbon price and global abatement. It shows how a TCC navigates the major challenges that arise when carbon abatement is adopted by less than the Grand Coalition. The quantitative section studies coalition stability in a quantitative model that resembles C-DICE, the coalition-formation model used in Nordhaus (2015). The unique stable coalition consists of the US, EU, and UK. It optimally sets the first-tier carbon price near the global Social Cost of Carbon, achieving global abatement just over half the efficient level. The agreement could be strengthened to achieve the efficient level of global abatement if the US-EU-UK coalition could increase the match rate to 67 percent.

Discussant(s)
Terrance Iverson
,
Colorado State University
Renato Molina
,
University of Miami
David Kelly
,
University of Miami
Larry Karp
,
University of California-Berkeley
JEL Classifications
  • Q5 - Environmental Economics
  • D8 - Information, Knowledge, and Uncertainty