The effect of learning on climate policy under fat-tailed risk

In Chang Hwang, Frédéric Reynès, Richard S.J. Tol

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

This paper investigates the effect of learning on climate policy under fat tailed risk about climate change. We construct an endogenous learning model with fat-tailed uncertainty about the equilibrium climate sensitivity. We find that a decision maker with a possibility of learning lowers efforts to reduce carbon emissions relative to the no-learning case. The larger the tail effect, the larger the counteracting learning effect because learning reduces the marginal benefit of emissions control compared to the case where there is no learning. The optimal decisions (summarized by the carbon tax level) in the learning case are less sensitive to the true value of the uncertain variable than the decisions in the uncertainty case. Learning lets uncertainty converge to the true value of the state in the sense that the variance approaches zero as information accumulates.

Original languageEnglish
Pages (from-to)1-18
Number of pages18
JournalResource and Energy Economics
Volume48
DOIs
Publication statusPublished - 2017

Fingerprint

Climate policy
Uncertainty
Decision maker
Carbon tax
Climate
Climate change
Learning effect
Learning model
Carbon emissions

Keywords

  • Bayesian learning
  • Climate policy
  • Dynamic programming
  • Fat tailed risk
  • Integrated assessment

Cite this

@article{53096fc9057a493ebc027400159b4c4a,
title = "The effect of learning on climate policy under fat-tailed risk",
abstract = "This paper investigates the effect of learning on climate policy under fat tailed risk about climate change. We construct an endogenous learning model with fat-tailed uncertainty about the equilibrium climate sensitivity. We find that a decision maker with a possibility of learning lowers efforts to reduce carbon emissions relative to the no-learning case. The larger the tail effect, the larger the counteracting learning effect because learning reduces the marginal benefit of emissions control compared to the case where there is no learning. The optimal decisions (summarized by the carbon tax level) in the learning case are less sensitive to the true value of the uncertain variable than the decisions in the uncertainty case. Learning lets uncertainty converge to the true value of the state in the sense that the variance approaches zero as information accumulates.",
keywords = "Bayesian learning, Climate policy, Dynamic programming, Fat tailed risk, Integrated assessment",
author = "Hwang, {In Chang} and Fr{\'e}d{\'e}ric Reyn{\`e}s and Tol, {Richard S.J.}",
year = "2017",
doi = "10.1016/j.reseneeco.2017.01.001",
language = "English",
volume = "48",
pages = "1--18",
journal = "Resource and Energy Economics",
issn = "0928-7655",
publisher = "Elsevier",

}

The effect of learning on climate policy under fat-tailed risk. / Hwang, In Chang; Reynès, Frédéric; Tol, Richard S.J.

In: Resource and Energy Economics, Vol. 48, 2017, p. 1-18.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - The effect of learning on climate policy under fat-tailed risk

AU - Hwang, In Chang

AU - Reynès, Frédéric

AU - Tol, Richard S.J.

PY - 2017

Y1 - 2017

N2 - This paper investigates the effect of learning on climate policy under fat tailed risk about climate change. We construct an endogenous learning model with fat-tailed uncertainty about the equilibrium climate sensitivity. We find that a decision maker with a possibility of learning lowers efforts to reduce carbon emissions relative to the no-learning case. The larger the tail effect, the larger the counteracting learning effect because learning reduces the marginal benefit of emissions control compared to the case where there is no learning. The optimal decisions (summarized by the carbon tax level) in the learning case are less sensitive to the true value of the uncertain variable than the decisions in the uncertainty case. Learning lets uncertainty converge to the true value of the state in the sense that the variance approaches zero as information accumulates.

AB - This paper investigates the effect of learning on climate policy under fat tailed risk about climate change. We construct an endogenous learning model with fat-tailed uncertainty about the equilibrium climate sensitivity. We find that a decision maker with a possibility of learning lowers efforts to reduce carbon emissions relative to the no-learning case. The larger the tail effect, the larger the counteracting learning effect because learning reduces the marginal benefit of emissions control compared to the case where there is no learning. The optimal decisions (summarized by the carbon tax level) in the learning case are less sensitive to the true value of the uncertain variable than the decisions in the uncertainty case. Learning lets uncertainty converge to the true value of the state in the sense that the variance approaches zero as information accumulates.

KW - Bayesian learning

KW - Climate policy

KW - Dynamic programming

KW - Fat tailed risk

KW - Integrated assessment

UR - http://www.scopus.com/inward/record.url?scp=85010310647&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85010310647&partnerID=8YFLogxK

U2 - 10.1016/j.reseneeco.2017.01.001

DO - 10.1016/j.reseneeco.2017.01.001

M3 - Article

VL - 48

SP - 1

EP - 18

JO - Resource and Energy Economics

JF - Resource and Energy Economics

SN - 0928-7655

ER -