BCBS issues discussion paper on climate scenario analysis

The Basel Committee on Banking Supervision (BCBS) issued a discussion paper on the use of climate scenario analysis to strengthen the management and supervision of climate-related financial risks. The feedback period on the paper ends on July 15, 2024. Based on the feedback received through this consultation and building on the work underway by authorities such as the Financial Stability Board (FSB) and the Network for Greening the Financial System (NGFS), the Basel Committee will consider complementary work in line with its mandate to strengthen the related regulation, supervision, and practices of banks worldwide.

The discussion paper outlines the objectives of a climate scenario analysis exercise in the Basel framework and differentiates between the climate stress testing and climate scenario analysis. The paper notes that climate scenario analysis is a forward-looking tool to assess resilience of business models and strategies of banks. This tool uses a range of plausible climate-related pathways for determining the impact of climate-related risk drivers on the overall risk profile. However, differences in the scope, features, and approaches of climate scenario exercises across jurisdictions limit the harmonization of supervisory expectations and the comparability of results. In seeking to advance common practices for climate scenario analysis, the Basel Committee has provisionally identified key features of climate scenario analysis exercises that would be relevant for all climate scenario analysis exercises conducted by banks or supervisors:

• Motivation: The exercise should have clear and well-defined objectives. These objectives should drive the development of frameworks, be adequately documented, and be clearly communicated to all relevant stakeholders.
• Comprehensiveness: The analysis should encompass all relevant factors, considering the nature and composition of the institution or type of exposures in focus.
• Plausibility: The scenarios should be realistic and believable. Scenario assumptions and their rationale should reflect the most recently available information from credible experts and should be supported by rigorous analytics. Modeling assumptions should consider how historical relationships may plausibly shift to reflect structural shifts in economic and climate conditions analyzed by the scenarios.
• Coherence: The scenario design, risk analysis, modeling, and overall exercise design should be internally consistent.
• Transparency: The scenarios should be clear and understandable to users while maintaining transparency of assumptions and limitations.
• Tractability: The scenario suites should be adaptable to changing circumstances and easy to reproduce. Shocks from climate-related financial risks should be mapped to economic and financial variables used by banks and supervisors in a reproducible manner, and historical series of the variables should be provided where appropriate
• Proportionality: The complexity of the climate scenario analysis should be aligned with the significance of the risks for the specific institution.
   

These initial considerations are intended to foster a robust dialog on the design, implementation and use of climate scenario analysis by banks and supervisors and to channel industry and supervisory efforts toward collective progress on these exercises. The usage-specific considerations may include the degree of standardization, time horizons, severity of scenarios, baseline selection, granularity, balance sheet assumptions, and analytical frameworks. Standardization can include, among other things, providing on a common set of climate risk scenarios; utilizing scenarios from specific scenario builders; requiring the use of specific scenario variables; introducing required data elements; selecting particular asset classes, sectors, or geographies; specifying modeling techniques; or requesting specific data outputs. The Basel Committee is seeking broad feedback on these considerations from a range of stakeholders, including the identification of features that have not yet been noted in this discussion paper.