BIS study: Carbon uncertainty key for sovereign bond portfolios
A new Bank for International Settlements (BIS) working paper introduces a novel framework for constructing sovereign bond portfolios that integrates financial and environmental considerations. It models carbon returns as random variables to capture future emissions uncertainty.
Modeling carbon as a random variable
The paper introduces "carbon returns" for sovereign bonds, defining them as the negative change in carbon emissions, analogous to financial returns.
This novel approach treats carbon footprints as random variables, explicitly accounting for future uncertainty, unlike existing methods that rely on historical data or single deterministic scenarios.
This shift avoids simply favoring current low emitters and instead rewards actual reductions, enhancing robustness.
To construct these decarbonized portfolios, the authors adopt a Hierarchical Risk Parity (HRP) algorithm.
This method balances financial and carbon risks, offering a robust and diversified approach that is resilient to estimation errors and avoids corner solutions, making it well-suited for this application.
The growing imperative for green portfolios
Institutional investors, including reserve managers, increasingly integrate climate risk into their portfolios, driven by both risk implications (physical and transition risks) and the potential to influence climate outcomes.
Given the central role of government bonds, reducing their carbon footprints is a pressing concern.
Existing literature often uses constrained optimization, prioritizing financial returns while incorporating carbon budgets.
However, a key limitation is treating carbon footprints as deterministic values, either historical or tied to a single scenario.
This fails to account for inherent uncertainty, akin to asset allocation without considering return variance, potentially leading to suboptimal outcomes.
Beyond greenwashing metrics
This framework offers a significant methodological leap, moving beyond simplistic, backward-looking carbon metrics to embrace the inherent uncertainty of climate transition.
By modeling carbon as a random variable, it provides a more robust and forward-looking tool for investors genuinely committed to decarbonization.
While complex, its adoption could lead to more impactful and financially sound sustainable investment strategies.