AUTHOR=Aguais Scott D. , Forest Laurence R. 

TITLE=Climate-change scenarios require volatility effects to imply substantial credit losses: shocks drive credit risk not changes in economic trends

JOURNAL=Frontiers in Climate

VOLUME=Volume 5 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/climate/articles/10.3389/fclim.2023.1127479

DOI=10.3389/fclim.2023.1127479

ISSN=2624-9553

ABSTRACT=Climate related credit risks from a Macro-Prudential point of view for banking  system viability is a key discussion point driving climate stress test R&D.  Led by regulators and the NGFS, early modelling approaches utilize smooth, top-down scenarios coupled with carbon emissions data to assess future climate related credit losses for individual firms.  While the NGFS approach is in its infancy, industry feedback has identified a number of concerns with using top-down scenarios that don’t reflect the potential for a broader range of more extreme future climate impacts. Additionally, the use of empirical models of credit risk with dedicated industry and region models could improve on the current use of top-down distributed aggregate measures.

In contrast to the NGFS approach, credit risks are generally not driven by smooth macro-economic trends but by unexpected economic shocks that precipitate higher volatility and systematic deviations from average trends as seen in the three most recent recessions. Therefore any assessment of future climate induced credit risks must assess systematic volatility not just trends in economic variables. 

This paper explores future climate induced credit risk and credit risk volatility under three different empirical assessments.  To conduct these empirical estimates of climate driven credit risks, we utilize a well-known, multi-factor credit portfolio model implemented in the Z-Risk Engine. In each case credit losses up through 2050 are assessed using a benchmark US C&I credit portfolio and the C&I loss index published by the FRB. 

The first assessment of climate related credit losses compares smooth NGFS climate scenarios with a CCAR severely adverse scenario to demonstrate that volatility not trends drives credit risk. For the second and third climate credit risk assessments we use NGFS GMT projections to estimate volatility effects on climate induced credit losses. These assessments use both industry and region factor model and macro-economic factor simulations.  These three empirical assessments respond to key industry concerns and provide an alternative, foundation for assessing climate driven credit risks, highlighting the role of systematic credit volatility as compared to the NGFS approach focused on macro-economic trends.