An Integrated Credit/Climate Scenario Approach Combining Firm-Level Climate Sensitivity with Climate Volatility Add-Ons
Published|
Share|
This paper by CGFI Associate Fellow Dr Scott D. Aguais presents a detailed approach for developing climate risk stress test scenarios that applies firm-level climate physical-and transition-risk sensitivities and introduces credit risk shocks into climate risk modelling.
For firm-level climate effects, the framework applies the recent ECB climate risk approach assessing business climate sensitivity. To assess credit risk shocks and potentially rising future climate volatility, firm-level climate effects are integrated with the Z-Risk Engine credit factor models that assess the systematic component of unexpected credit risk shocks. The main research question therefore focuses on ways to develop climate risk scenarios that fully reflect the complexity of credit risk, and which provide multiple channels for future climate change to influence credit risk portfolios.
The paper presents the detailed approach and technical details and applies the integrated framework to a roughly £140 billion UK/European business credit portfolio to assess credit/climate scenario effects on various credit measures, including business default probabilities, expected credit losses and low-probability ‘tail’ credit losses. The paper also outlines our ongoing credit/climate research focused on detailed calibration of the integrated approach.
Acknowledgments
This Z-Risk Engine Research Paper has been made possible by close collaboration with the UK Centre for Greening Finance and Investment (CGFI), based on Scott’s affiliation with CGFI as an Associate Research Fellow working on integration of climate considerations into state-of-art credit risk assessment. The paper does not necessarily represent the views of CGFI or the CGFI consortium members.
We would like to thank Dr Gireesh Shrimali, Head of Transition Finance Research at the Oxford Sustainable Finance Group, University of Oxford, for directly supporting this collaboration and for his overall guidance for this paper, in particular, via reviews of the research proposal, research design, and paper drafts. We also thank our Associate Fellow colleagues at CGFI; in particular, Dr Chris Cormack and Dr David Wilkinson, for their helpful comments and feedback – which helped us balance academic rigor with practical insights – during a thorough and insightful review process.
Finally, we want to thank all the participants in an internal CGFI Workshop on March 6, 2024, where
we presented our paper, for their helpful comments.
