Climate-Related Risks and Central Banks’ Collateral Policy

Traditional approaches to financial risks are backward looking and rely on standard probability laws. They are therefore unsuited to the complexity and uncertainty associated with climate-related risks.As an alternative to these methods, this Banque de France paper explores a potential alignment approach measuring the consistency of a portfolio of financial assets with specific climate targets (normally 1.5°C or 2°C of global temperature rise).

The paper finds that aligning portfolios with climate targets is already practiced by impact investors and the methodologies involved are relatively simple and transparent. Although the metrics produced by this approach are not directly financial, this is not a problem in the case of a central bank’s collateral framework. This is because the risk assessment is system-wide and the measure of alignment of the assets accepted as collateral is a relevant indicator of how exposed those assets are to transition risks.

The paper proposes that central banks reduce climate risk uncertainty by relying on a “climate-hedging portfolio approach” aiming for aggregate alignment of the collateral pools pledged by counterparties with a given climate target. As the composition of the collateral pools would remain in the hands of the counterparties, this approach avoids the perception that the central bank is involved with political decisions and maintains the strict principle of market neutrality.

Following a literature review on climate-related risks in relation to the collateral frameworks of central banks, the paper goes on to examine the challenges of accurately assessing such risk, as well as the climate-hedging portfolio approach. These methodologies and criteria are then applied to the Eurosystem’s collateral framework and the results discussed.

The paper concludes with an argument for the alignment of collateral pools as both consistent with market neutrality and more appropriate to the radical uncertainty of climate change.