Inclusion of carbon pricing into stress testing for the Austrian banking sector

Abstract This paper introduces two approaches for integrating climate-related financial risks into bank stress testing by translating carbon price projections into parameters for assessing climate risk. The first approach employs a linear model to translate sector-level probability of default changes, derived from external models, to the business unit level. The second approach directly incorporates the carbon price and emission intensity at the level of the economic category. It captures the interaction between carbon pricing, specific emissions profiles for NACE Rev.2 categories, and default risk, allowing for the estimation of vulnerabilities within bank portfolios and providing a basis for ESG considerations. Furthermore, we introduce a ’sigmoid’ scenario for carbon price development, bridging the gap between orderly and disorderly transitions as well as facilitating the exploration of non-linear carbon price dynamics using economic parameters. Applying these approaches to a sample portfolio reveals differences in probability of default estimates, underscoring the significance of methodological choice.