Response to consultation on Guidelines on ESG scenario analysis

Question 1: Do you have any comments on the interplay between these Guidelines and the Guidelines on the management of ESG risks?

• The EBA Guidelines on the management of ESG risks comprehensively set out the processes for identifying, measuring, managing, and overseeing ESG risks. These guidelines are useful for identifying the necessary elements to integrate ESG risks into the bank's risk management, considering the current practice and the further enhancement of the bank's risk management framework.
• On the other hand, the Guidelines on ESG scenario analysis cover the scenario analysis process in detail, providing specific methods for analysing both transition and physical risks, the relationship with existing scenario analyses, and recognising the challenges for future enhancement.
• These guidelines provide useful insights for bank’s risk management, and we appreciate the EBA’s effort.

Implementation timeline

• According to the Guidelines on the management of ESG risks, a transition plan must be created by 11 January 2026 (11 January 2027 for SNCI), and the same effective date applies to scenario analysis.
• However, for scenario analysis, the effective date should be set taking into account an appropriate analysis period after the finalisation of the guidelines. These guidelines will only be finalised later this year, leaving very little time for institutions to implement the necessary changes.
• We support a phased implementation approach for these guidelines or alternatively extending the  implementation deadline for at least one year. This is particularly important given the significant efforts required for the implementation of the Guidelines on the management of ESG risks. A phased approach would facilitate a smoother transition and allow institutions to build expertise incrementally, ensuring compliance without undue burden.
• We also encourage the EBA to consider the changes proposed in the Omnibus proposals regarding CSRD which will have an impact both on the EBA guidelines and on ESG risk management.

Question 2: Do you have comments on the proposed definition of scenario analysis and its various uses as presented in Figure 1?

• The definition of scenario analysis presented is based on the TCFD recommendations, and we support the proposed definition.
• We also agree with the uses of scenario analysis. Scenario analysis can be utilised for various purposes, primarily in strategic planning and risk management. Analysis aligned with the objectives provides opportunities to utilise scenario analysis for various usages. Since ESG is a broad topic, identifying the key areas concerning scenario analysis would be effective from a business/risk management perspective.
• The PDCA cycle of risk management presented in Figure 1 is sequentially described, making it easy to understand the flow from the adaptation of strategies and business models (P) to assessing results through engagement with counterparties (C), and the actions such as strategy revisions (A).
• Financial institutions might benefit from more illustrative examples (e.g., case studies) showing how to translate high-level scenarios into concrete action plans or performance indicators. Such examples would further clarify the practical uses of scenario analysis.
• Moreover, we welcome further clarification on the requirements and the use of scenario analysis within prudential transition planning under CSRD, especially in light of the Omnibus package.
• More broadly, given the acknowledged inherent uncertainties in CRA, we seek better clarity regarding the integration into strategy and business model planning.

Question 3: Do you have comments on the proposed distinction made between short-term scenario analysis (CST) and longer-term scenario analysis (CRA) as illustrated in Figure 3?

• Considering the integration with existing risk management frameworks such as ICAAP and ILAAP, it is reasonable to introduce analysis on a short-term time horizon. We do not have any objection to contrasting baseline and adverse but plausible scenarios, as in conventional stress testing frameworks.
• While we generally agree with the distinction showed in Figure 3 itself and such clear distinction made between CST and CRA is helpful for conceptual understanding, practically there seems to be considerable overlaps in the methodologies across both versions of scenario analysis. As stated in the guidelines, scenario analysis is required in both components of "Capital Adequacy" (CST) and "Business Model Analysis (BMA)" (CRA).
• In this regard, especially for smaller financial institutions with limited resources and capacities (and for subsidiaries of foreign financial institutions where the strategic direction of the parent company has a significant influence), the application of CRA should be carefully considered. Therefore, it would be helpful if further guidance is provided on proportionality in the application of CRA.
• We acknowledge that using a dynamic balance sheet in long-term scenario analysis presents several challenges to enhance effectiveness, such as eliminating arbitrariness, reflecting industry-wide restructuring (e.g., separation and integration of high-emission businesses), and changes in procurement structures (e.g., from business corporations to SPCs and funds). For conducting the dynamic balance sheet approach, further guidance/hypotheses to align the methodology among financial institutions would also be necessary.
• Moreover, as stated in Chapter 6, CRA is required to be conducted over several time horizons. Therefore, we assume that it is possible to conduct short- or medium-term analysis on business model resilience (not financial resilience). However, such analysis is not considered in the distinction proposed in Figure 3, resulting in an inconsistency. It would be preferable to align the figure with the former approach.
• We welcome EBA's acknowledgment that, for the assessment of transmission channels, more qualitative approaches based on expert judgment will be needed. To further improve the exercise, we encourage the EBA to provide clear guidance on the expected granularity of "material" risks (risk drivers and transmission channels). We also welcome more clarity and specific requirements on the level of severity (and probability of materialisation) of future pathways that are expected for CST and/or CRA.

Question 4: Do you have any comments on the interplay between these Guidelines and the Guidelines on institution’s stress testing?

• There are no particular objections regarding the interplay between the guidelines.
• However, for the integration into stress test models, it is necessary to understand the interaction between economic recessions, which were the focus of conventional stress testing, and the impacts of climate change. It is much more important to consider industry and company-specific characteristics compared to conventional stress tests. It should be noted that it is challenging to immediately incorporate climate change risks into stress tests. Additionally, duplicated analysis should be avoided.
• In the conventional stress testing (ICAAP, common stress testing) of a particular bank, macro stress is limited to indicators such as GDP growth rate and unemployment rate, and analysis is conducted using regression models between these macro indicators and financial indicators such as sales and profits. On the other hand, climate change related scenario analysis introduces various variables, including energy prices and carbon taxes, and uses methods that add these as stress factors to individual items in the P/L and B/S, resulting in a significantly different level of granularity in analysis.
• When analysing both and interacting them, it is necessary to consider a mechanism that allows for the comparison of financial impact (credit costs) as outputs, using some measurements to compare the intensity of stress between the two. Additionally, when using short-term scenario analysis for managing or disclosing capital and liquidity, it is necessary to ensure the comparability chronologically  and with peers in the industry. This requires not only standardisation of scenarios but also a certain level of standardisation of simulation model. Particularly regarding validation methods, combining macroeconomic and climate-specific variables may require more sophisticated validation frameworks. Additionally, since historical data for climate change is limited and our analysis relies on expert judgment, transparent documentation becomes especially important.
• We welcome the acknowledgment that ESG scenario analysis, while expected to be part of the ICAAP and ILAAP, is still at a nascent stage given significant model uncertainties and a lack of established standards for assessment.

Question 5: Do you have comments on the Climate Scenario Analysis framework as illustrated in Figure 4?

②Define the scope for exercise 

• As mentioned in our response to Question 3, we recognise that using a dynamic balance sheet in climate resilience analysis presents high hurdles. From the perspective of objectivity, the use of a static balance sheet should also be permitted.
• We also expect more clarity on the definition of scope for material exposures, for example, lending exposures versus other risk categories potentially subject to more immediate climate shocks (e.g., market and liquidity risk) that would require very specific modelling techniques and potentially requiring further research to identify risk drivers and transmission mechanisms.

③Set scenarios

• We consider that defining one central scenario for CRA with a set of alternative scenarios is in contrast to the current approach for CSA in the first place, which conduct analysis of scenarios in parallel to explore future climate pathways (e.g., Hot House World and Net-zero).
• However, based on the description in paragraph 39 on page 28, it is assumed that, for NGFS, NDCs scenario is conceptually more aligned with the baseline scenario than the Current Policy scenario. Paragraph 40 states, "The central scenario extends the baseline scenario by covering a much longer time horizon," which could also be interpretated that using the NDCs scenario as the central scenario may be more appropriate. At the same time, we acknowledge that some banks currently position the Current Policy scenario as what could be described as the central scenario in long-term scenario analysis (CRA) and conducts comparative analysis with several other scenarios in parallel. Given that many banks use NGFS scenarios, we request to provide examples of which scenarios should be adopted as the baseline scenario and the central scenario.
• We also would like to point out that scenario should be approved as part of the entire methodology rather than on a scenario-by-scenario basis.

⑥Assess the impacts

• CRA can be conducted quantitatively. A certain number of financial institutions are conducting analyses based on long-term scenarios using a bottom-up approach, which can be used to evaluate the resilience of financial institutions. As above-mentioned, for conducting the dynamic balance sheet approach, further guidance/ hypotheses to align the methodology among financial institutions would be welcomed.
• Also, we would welcome the EBA to specify which modelling approach is deemed adequate (e.g., bottom-up, top-down).

⑦Use the result

• We expect clarification on "management actions", in particular how CSA outcomes are to be used for "adjustment of financial terms, conditions, and/or pricing based on climate risk considerations". "Management action" in this context can be interpreted as granular risk mitigation, potentially requiring different modelling granularity compared to the conventional use-case of ICAAP, ILAAP and long-term strategy information.
• Given that CST and CRA involve various assumptions and significant uncertainties, it should be noted that there is considerable uncertainty regarding the validity of any "management actions" based on CST and CRA results.
• It is not only premature given the timing and likelihood of climate risks manifesting but it is also worth noting that it poses a significant risk of leading to substantial divestment.

Other

• As part of the common framework content for both CSA and CST, we suggest considering descriptions that assume scenarios where transition risk and physical risk interact (e.g., a carbon-intensive sector located in a flood-prone area). Such compound scenarios are also considered in the latest short-term scenarios by NGFS and addressing them might yield more realistic insights.

Question 6: While respecting the definitions provided in other parts of the regulation, is there any concept/s used in these guidelines that it would be useful to include in an annexed glossary?

• The glossary could include “compound risk,” “double materiality,” and “feedback loops,” as these terms increasingly appear in ESG related discussions and could be relevant in scenario analysis contexts.

Question 7: Do you have comments on section 4.1 Purpose and governance?

• We agree with the suggestion that the analysis could be used for the purpose of identifying business risks and opportunities, assessing vulnerabilities to transition and physical risks, and supporting the formulation of strategic and transition planning through resilience evaluation.
• Page 24 paragraph 19 states that scenario analysis should be substantiated and documented, and we have  no objection to this requirement. However, unlike short-term stress tests, long-term scenario analysis cannot be back-tested against actual values, which presents a challenge. The validity of the narratives and results therefore has to be evaluated by testers and users, including management. This makes it difficult to separate the "validators" from the "users of the results" as it is often assumed that both roles are performed by the same individuals.
• At present, due to the significant uncertainty surrounding ESG and the fact that CST and CRA are still in development in terms of data and methodologies, caution should be exercised in overinterpreting scenario analysis results or using them for decision-making and internal/external communications, as stated in page 16.

Question 8: Do you agree that the proposed proportionality approach is commensurate with both the maturity of the topic and the size, nature and complexity of the institution’ s activities?

• We believe that the approach of tailoring scenario analysis methods to the specific business and portfolio characteristics of each financial institution is highly appropriate.
• Considering that ESG scenario analysis is still in its early stage, it is effective to start analysis with the most material ESG risks based on the results of the materiality assessment, and to conduct stress testing and resilience analysis while balancing quantitative and qualitative approaches. We would appreciate EBA being transparent about when qualitative approaches are sufficient.
• We suggest that it should also be explicitly stated that the most material sectors, regions, and portfolios should also be focused on, based on the principle of proportionality.
• Regarding scientific understanding, it would be helpful to define the context or process for adopting a new common scientific understanding.
• We would also appreciate EBA incorporating recent developments on the potential amendment of CSRD with the sustainability omnibus and addressing potential delays or lack of data from CSRD disclosures of counterparties.

Question 9: Do you agree with the proposed references to organisations in paragraph 28? Would you suggest alternative or complementary references?

• While we agree with the proposal as referenced scenarios are advantageous in that they have comprehensive macroeconomic indicators, it is suggested that the IPCC, IEA, and SBTi be explicitly referenced since they are also widely recognised organisations. Scenarios like the IPCC will be useful for estimating physical risk.
• Additionally, in sector-specific analyses, the use of scenarios provided by respective industry associations should also be permitted.
• While we recognise that NGFS scenarios are credible, there seems to be differences in how financial institutions approach and use carbon price parameters in scenario analysis. At this point, it would be helpful to have a standard within the financial sector.

Question 10: Do you have additional comments on section 5.1 Setting climate scenarios?

• When setting scenarios, the purpose of use and the limitations of the analysis methods should be considered, and it is preferable not to use overly complex scenarios.
• We understand there are varying scenario source expectations, for example, use of external scientific scenarios versus modelling material risk drivers and transmission channels that may impact a bank's position. We encourage the EBA to clarify their expectations in using scenario analysis to inform strategy, since, on the one hand the guidelines state that "Scenarios are hypothetical constructs and not designed to deliver precise outcomes or forecasts," but also indicate that "use of scenarios to help define and adjust the institution’s strategy".
• With respect to paragraphs 29, 32, and 36, there are technical challenges. Specifically, for paragraphs 32 and 36, since there are no credible scenarios that incorporate the correlation between transition risks and physical risks or consider the impact across the entire value chain, we believe that quantitative analysis is difficult. Regarding these points, best practices should be shared, and it should be considered that addressing these paragraphs will take time.
• Regarding paragraph 39 (the consideration of additional macroeconomic shocks in adverse scenarios), while importance of compound risks is understandable, there is a concern that considering additional macroeconomic shocks may not necessarily be appropriate at a stage where the understanding of the characteristics of climate risks is not yet deepened. We consider that enhancing the assessment of climate risks is also an appropriate approach.
• As mentioned in our response to Question 6 above, it would be helpful to define specific examples of the baseline scenario in CST and the central scenario in CRA, such as NGFS scenarios.

Question 11: Do you have comments on the description of the climate transmission channels?

• While avoiding overlaps, it is very important to consider macroeconomic impacts and microeconomic impacts separately in order to appropriately assess the effects of climate risks.
• Although we welcome the detailed and comprehensive list of micro- and macroeconomic factors listed, these requirements are considered to be too specific and should be subject to materiality and applicability. Additionally, since it is not easy to fully understand climate-related transmission channels, it is preferable to focus on CST and CRA for risks recognised as material, rather than refining the understanding of these channels completely. While the guidelines acknowledge data challenges, this does not seem to be reflected in this list of transmission channels.
• Similarly, the counterparty-level approach and supply-chain consideration in paragraph 49 also neglects the vast uncertainties and lack of visibility on indirect ESG risk exposure at a counterparty level.
• While paragraphs 49 and 51 are useful when detailed information about the value chain and the regions where the business operates is available, usefulness is considered limited for globally operating firms. Therefore, it is recommended that the requirements be amended to apply only to exposures where they are highly useful, such as real estate-backed loans or firms operating in specific regions.
• However, in the physical risk analysis of asset finance (real estate, project finance) and mortgage loans, insurance coverage is undeniably important. We believe it is necessary to appropriately reflect how the provision of insurance for natural disasters by insurance companies (insurance amounts, premium rates) will change according to future scenarios. It is essential to consider a method to incorporate the results of insurance companies' scenario analysis into the bank's scenario analysis.
• As such, it should be noted that addressing these paragraphs will take time. Furthermore,  it would be desirable to share best practices to enhance analysis.

Question 12: Do you have comments on climate stress test (CST) tool and its use to test an institution’s financial resilience?

• We agree with the recommendation of the proposed phased approach. Since, however, it is difficult to immediately integrate it into the existing risk management framework, it is reasonable to initially conduct the analysis within a separate system.
• Setting appropriate stress levels is a challenge because we could not rely on historical data due to the nature of climate risks. We understand that NGFS is currently working on short-term scenarios, but as there are no globally recognised reference scenarios available at this point, there needs to be a clear definition of appropriate stress levels.

Question 13: Do you have comments on the Climate Resilience Analysis (CRA) tool and its use to challenge an institution’s business model resilience?

• For managing the financial stability and business continuity of financial institutions, CST is considered sufficient.
• As for CRA, it is closely related to the strategic planning of each financial institution (and in the case of subsidiaries of foreign financial institutions, the strategic direction of the parent company has a significant influence). Therefore, it should be operated more flexibly.
• Considering the operational burden on financial institutions, it is particularly desirable to align as much as possible with scenario analysis guidance under CSRD and ISSB.
• On page 35, the use of a dynamic balance sheet is recommended. However, as mentioned in our comments on Question 3, when developing a dynamic balance sheet, it is essential to avoid arbitrarily incorporating portfolio changes based on divestment from high-emission sectors. Additionally, analysis using a dynamic balance sheet that does not account for industry restructuring (e.g., separation and integration of high-emission businesses) or changes in funding structures (e.g., from business corporations to SPCs and funds) is not only insufficient in appropriately reflecting the changes in the economic environment and credit risk volume towards long-term decarbonisation but also potentially misleading.
• As the methodology for validation is limited, estimating a reasonable and feasible balance sheet and creditworthiness of individual debtors in sustainable procedures is indeed a challenging task. In addition, a methodology for assessing business model resilience has not been established.
• Given these points, it would also be preferable to start with qualitative analysis and limited initiatives, gradually advancing toward more sophisticated approaches.
• Also, the latest scientific assessments suggest that limiting global warming to 1.5°C might still technically be feasible, but it is becoming increasingly difficult to achieve and would require unprecedent action and behavioural and economic shifts. We believe that these objectives should be formulated by the EBA and how they connect to management perspective in terms of elevated physical risks in an alternative pathway (e.g., a Hot House World).

Question 14: Do you have any additional comments on the draft Guidelines on ESG Scenario Analysis?

• We encourage the EBA to clarify the overarching objectives of the guidelines, specifically in relation to the interaction between business opportunities and stress testing. As it could lead to confusion for financial institutions in identifying which perspective they should adopt.
• Although section 3.2 states in page 9-10 that there are some limitations of the IEA scenarios for long-term scenario analysis, not much information can be obtained through NGFS and other scenarios. Therefore, it would be desirable to provide alternative approaches.
• In the absence of data / methodology on ESG, conducting scenario analysis for entire ESG is difficult. In that context, it should take into account the outcome of the omnibus package, and its effect to the data accessibility.
• Moreover, careful consideration should be given to ensure that requirements do not overlap for institutions, particularly with other regulations and standards such as ISSB.

Upload files

Name of the organization