Understanding and analyzing climate risk has recently become a critical goal for banks — driven in part by regulatory mandates and market pressures. In response, US banks have now started to get going on the task. In the last year, some have made substantial progress in developing climate scenario analysis (CSA) capabilities. Still, key gaps must be filled to attain the quality of analytics that would enable banks to use the analytics beyond regulatory compliance. In our view, banks can benefit by deepening and accelerating their work on the topic.

In this report, we examine the current state of the industry and provide perspectives on US financial institutions’ climate scenario analysis capabilities based on a survey conducted by the Risk Management Association and Oliver Wyman in Q4 2022. The key takeaways include:

The potential of climate scenario analysis is just beginning to be tapped

Use cases such as strategic planning (35% of respondents employ CSA in this process), credit assessment (26%), and client engagement (14%) are not yet common among surveyed banks. When well developed and effectively deployed, CSA can generate useful insights for decision-making across these use cases.

Many banks have not yet deployed the analytics needed to successfully complete the Federal Reserve’s CSA pilot

Of the focus areas of the pilot, banks as a group lag most in the ability to assess transition risk in commercial real estate (39% of respondents can do so), largely due to data challenges. The banks a size smaller than the six in the pilot should begin to prepare capabilities, as broader regulatory exercises are expected to follow.

Overall, survey results revealed banks’ CSA methodologies are evolving, they sometimes miss key risk drivers or are based on overly simplified approaches, and there is still work to be done in strengthening capabilities. Some challenges to advancing the use of CSA across portfolios include the following:

Transition Risk for C&I: Sector-specific approaches provide more differentiated counterparty -specific answers. However, only 22% of those surveyed use this approach for at least part of their portfolio. Simplified “emissions-based” or top-down approaches are more popular but may not adequately capture the transition dynamics of high-risk sectors such as energy and power generation.

Transition Risk in Real Estate: In the US, this has not been widely modeled due to data challenges. Energy consumption and efficiency data is not easily available across the board outside of certain urban areas such as New York City that have requirements to report this type of data. However, there are approaches to estimate proxy data and perform meaningful transition risk analysis.

Physical Risk in Real Estate: Banks are typically reliant on specialized vendors for the data to perform property-level physical risk analysis. Most banks have contracted such vendors to provide physical risk data for Commercial Real Estate (85%) and for Residential Real Estate (71%). However, offerings vary widely in scope, quality of underlying models, and output metrics. Understanding vendor capabilities and appropriate “fit-for-purpose” application of the data is critical.

Physical Risk in Real Estate. Banks are missing out on several key physical risk drivers in their current modeling. Only 37% of banks currently model how higher future insurance premiums or other climate-related costs of ownership could affect the market price of real estate.

Why does it matter?

Deepening and accelerating the development of select CSA capabilities can position banks well for any upcoming climate risk-related regulatory requirements. More importantly, it allows them to better understand their climate risks, make strategic business decisions, and draw conclusions for credit underwriting.

This report was written in collaboration with the Risk Management Association (RMA). The Oliver Wyman authors of this piece includes Lukas Widera, Ilya Khaykin, and Alban Pyanet.