What part of the real price of carbon is due to uncertainty?
Ton S. van den Bremer and Frederick van der Ploeg tackle how to price carbon emissions in an uncertain world in their 2021 paper « The Risk-Adjusted Carbon Price ».
Using an analytical asset-pricing framework, they incorporate climate and economic risks into the Social Cost of Carbon (SCC) and derive a tractable formula for the “risk-adjusted” carbon price.
Their conclusions include:
- Accounting for uncertainty significantly raises the optimal carbon price: when future climate damages are uncertain, the SCC should include a risk premium on top of expected damages.
- In some scenarios, the risk-adjusted SCC more than triples the « deterministic » SCC. Plausible risk-adjusted SCC falls around $80/tCO2eq versus about ~$30/tCO2eq value without risk.
- A « prudence » effect (preferring safety for the future) and an « insurance » effect (mitigating climate shocks) both push the optimal SCC higher and tend to dominate in risk-averse societies.
- A « risk-exposure » effect emerges if climate damages mainly occur when the economy is strong, which can moderate the SCC and limit hedging against bad climate-economic outcomes.
- Economic volatility leads to a higher SCC, which interestingly links climate policy with financial market uncertainty.
- Uncertainty in climate sensitivity, or how much warming results from emissions, justifies a much more aggressive carbon price as a form of precautionary mitigation.
- If climate disasters tend to strike during economic downturns, then carbon pricing provides valuable insurance and should be set extra high to guard against compounded shocks.
The paper shows correlation patterns between climate and economy risk can shift the SCC, even though this consideration is largely absent in simpler models.
Bottom line: the more we fear extreme climate outcomes, the higher the price on carbon today. Carbon-intensive assets may face dramatic revaluation as this risk premium is recognised.
The authors capture all key uncertainties from a few parameters, which means unknown unknowns like unforeseen climate tipping points are not fully modelled and could push the optimal SCC even higher.