Response to discussion paper on the role of environmental risk in the prudential framework
Go back
We agree with EBA concluding remark that “Prudential regulation should remain risk-based and evidence-based". We also agree with EBA proposal that ESG risks are not specific risks, but ‘risk drivers’ that may impact “traditional” categories of risks such as credit, market, operational or strategic risk. We also fully share EBA views that historical data and evidence is not yet available to support an adaptation of the Pillar I framework.
ADDITIONAL CAPITAL REQUIREMENTS WILL HAMPER EU BANKS’ ABILITY TO FINANCE THE TRANSITION
• We believe that increasing banks’ capital requirements is not the right approach as banks need to be able to finance the transition of their clients, in a context of increasing financing needs to support the economy’s transformation shift. This is all the truer in the EU where the financing of companies remains mostly bank-loan based.
• “Front-loading” capital requirements would not necessarily deliver a tangible difference in resilience. Instead, they could constrain capacity to extend credit and investment to key sectors in need of transition finance, which could potentially ‘bring forward’ transition-related disruptions to the economy (e.g., exacerbating inequality by unaffordable energy prices), and negatively impact climate outcomes.
• Banks are part of the solution to achieve the objective of net-zero greenhouse gas (GHG) emissions in the EU economy by 2050 but they should not be the primary enforcers of the EU climate policy. There is a political responsibility in defining the relevant industrial and tax policies to ensure an orderly transition and limit transition and physical risk levels. EBA also strengthened in the Discussion Paper that “The primary responsibility and most effective tools for dealing with environmental-risk-related externalities lie within the remit of political authorities”. In this context, the priority is the definition by the European Union of a detailed transition path towards a decarbonized economy by 2050, at a granular level, by industrial sector and by country, considering the industrial implications of a successful transition. This is a key pre-requisite for the European banking supervisors to assess, as part of the Supervisory Review and Evaluation Process (“SREP”), the way banks adapt their climate risk management frameworks, as relevant, to support the financing of these transition paths.
• Last, in a globalised economy, punitive changes to EU banks’ prudential requirements would only result in a substitution of the financing, which will be taken over by non-EU banks and/or non- bank players, subject to less stringent regulatory standards. This may put the related risks beyond the reach of EU regulators and supervisors.
THE DEFINITION OF THE APPROPRIATENESS OF CAPITAL, IN TERMS OF NATURE AND CALIBRATION, IS NOT MATURE
• The Regulatory capital framework is designed to cover unexpected losses with a time horizon of 1 year, without the intervention of material mitigation and strategic actions. Banks, supervisors and regulators face huge challenges to include the long-term horizon and the forward-looking nature of climate, and more broadly, environmental risk factors.
• We share BCBS’s and EBA’s view that climate factors are not a new category of risk per se: they are ‘risk drivers’ of the existing prudential risk categories, especially credit risk, with a positive or a negative impact. However, given the nascent nature of the collective understanding of how the climate risk drivers may impact these traditional risks, it is deemed premature to define a regulatory capital treatment. Indeed, given gaps in the evidence-base, necessary data, and methodology, it is very challenging, at this stage, to calibrate risk parameters such as PDs and LGDs, in credit risk, taking into account climate related risks drivers. As long as robust risk-based methodologies have not been established and experienced, reliable counterparty data is not available and the results of supervisory exercises is not stabilised, it would be far too early to foresee any additional capital requirement.
• We also believe that non-risk-based Pillar 1 adjustments, such as the Brown Penalizing Factor or the Green Supporting Factor (GSF) would be pure political measures. The underlying objectives should be addressed through public policies or taxes and not through banking regulation. The anchor of the prudential framework must remain risk-based as rightly pointed out by the EBA in its discussion paper. The only consideration of the characteristic of a transaction being sustainable or not does not imply a low or high credit risk.
• The potential interplay between macroeconomic cycles and climate risk factors has not been established, so the use of macroprudential tools in this area would not be appropriate at this stage. In February 2022, the Financial Stability Institute (‘FSI’) highlighted in its Brief No. 16 ‘The regulatory response to climate risks: some challenges’ that applying the macroprudential framework to systemic climate-related financial risks is likely to be ineffective and potentially counterproductive for financial stability. It also notes that ‘macroprudential measures aimed at reducing exposures to carbon intensive firms and sectors may not always be conducive to reducing aggregate climate-related financial risks. In particular, a significant increase in capital requirements for brown exposures, by curtailing the availability of credit to carbon intensive industries would increase the vulnerability of those sectors and hinder affected firms from adjusting their business models’. Ultimately, a macroprudential buffer linked to climate could thus delay the financing of transition by banks, delay transition as a result and increase the potentially systemic nature of climate risk and financial stability as a result.
In addition, as highlighted below, proactive work is ongoing to adapt Pillar 2 frameworks, as need be, to capture climate-related impacts if deemed material. Hence, it will be key to ensure that overlap or double counting does not occur over time – as well-acknowledged by the EBA.
PILLAR 2 WOULD BE THE MOST APPROPRIATE FRAMEWORK, BUT THE DEVELOPMENT OF ROBUST CLIMATE SCENARIOS, METHODOLOGIES AND DATA ARE STILL GLOBALLY SHARED CHALLENGES
• Pillar 2 approaches could be considered, but stress testing is not the right approach to calibrate climate-related Pillar 2 requirements. Rather, FBF is supportive of climate scenario analyses, which are by design best suited to apprehend climate risk drivers, given their forward-looking nature. In this context, large French banks have participated in a number of industry-wide exercises, notably the French supervisor, ACPR climate stress test in 2020, which was done on a voluntary basis, and the ECB 2022 Climate stress test.
• Climate scenario analyses are useful tools to capture changes in business models that banks will need to undertake in climate scenarios. Indeed, transition scenarios may result in a rebalancing of economic activity across sectors and, within sectors, across counterparties. Climate scenarios are not designed, like traditional stress test scenarios, to measure losses, and thus capital, in adverse macro-economic circumstances. Indeed, some sectors or players in sectors can be expected to benefit from climate change. Climate scenarios are designed to anticipate sectoral evolutions that climate change will trigger and to help banks adjust to these changes accordingly. The horizon of climate change and of climate scenarios is also radically different from that of capital sizing stress tests. There are various initiatives to develop climate scenarios, such as those undertaken by the Network for Greening the Financial System (NGFS) or the International Energy Agency (IEA). However, while welcome, these scenarios are still incomplete. In particular, they do not offer a complete scenario framework including the modelling and projection of industrial and technological developments, as well as their translation into macro-economic variables. Until this is achieved, scenario analysis will remain exploratory and hypothetical in nature or even unrealistic on short-term horizons and not suited for capital sizing stress testing.
• Pillar 2, through SREP assessment, should focus on the quality of banks’ risk management framework in supporting well-defined transition plans from public authorities.
• The results published recently by the ECB of its 2022 climate stress test confirms the multiple challenges that both supervisors and banks face in developing reliable and plausible supervisory stress test exercises.
• The ECB notes that while banks make progress in developing climate stress testing infrastructures, many challenges remain for coherent results to be produced. Data availability and model developments remain unequal among banks, which is more a reflection of the specificities of climate risk (absence of historical observations, lack of uniform reporting of climate data by clients and, in the case of GHG emissions, lack of a uniform framework to report scope 3 emissions in particular) than banks’ reluctance to invest in climate stress testing frameworks.
• The approach by supervisors of climate scenarios is also widely different: for example, the ECB transition scenarios included a 3-year scenario, which is not the case of any previous climate stress test exercise; the ECB proposed two physical risk scenarios (flood and drought & heat) while the PRA considered that a “no policy action” scenario constitutes a physical risk scenario. It is very difficult for banks to develop frameworks in the absence of common scenario approaches by supervisors.
In addition, the time horizon of climate change inherently brings uncertainty to projections in climate scenarios. The ECB notes in its publication report that, as far as the 30-year transition scenarios are concerned, “it should be borne in mind that the 30-year projections are exploratory and subject to significant uncertainty. Therefore, these long-term loss projections should be interpreted as a qualitative yardstick for the direction of travel rather than as a robust quantitative measure.” The ECB also notes that, in the long-term scenarios, “the aggregate pattern of loan loss projections over time is masked by notable differences across banks with respect to projected losses in the long term”. This is particularly true as banks make different assumptions in balance sheet projections under a dynamic balance sheet methodology, which is the preferred option in 2050 projections and the only way to properly capture the business model dimension of climate stress testing.
• Most regulators acknowledge, the exercises are not robust enough yet in terms of data, scenarios, and methodologies. A progressive and iterative development of such methodologies, scenarios and data would enable banks to strengthen their risk management frameworks (for example through the building of risk and IT infrastructure and the development of climate specific scenarios) and effectively continue to include climate drivers in their Pillar 2 frameworks.
• In the meantime, a climate-related risk concentration framework, leveraging off the existing ones for pillar 2 and/or large exposures, could be useful to monitor specific areas of concentration and take actions, as relevant, to prevent a small number of financial institutions from unduly accumulating exposures to climate-related risks. However, any potential measure such as concentration limits should be cautiously considered to avoid harming specific regions or sectors and impeding efforts to scale up transition finance to sectors that need it most.
• Irrespective of the sector, clients demonstrating strong willingness to transition, with solid transition plans and commitments to reach net-zero by 2050 should be supported and best in class players should have incentives to pursue their transition shift.
CONCLUSION
• The focus should remain on bank’s adequate risk management and creating a policy environment that does not create disincentives and impediments to finance the transition. Tempting to move forward with “precautionary measures” such as capital buffers, would inevitably require a departure from the core risk-based foundations of the prudential framework and introduce subjective choices pertaining to broader policy objectives and potential unintended consequences.
• To cope with the challenges associated with time horizons, data, and methodologies, we believe that the banking industry together with regulators and supervisors should pursue dialogue and collaboration factoring in technical work and scientific research benefitting from industrial and sectoral experts.
• Scenario analysis is an appropriate tool to assess the consequences of climate change in various transition scenarios and anticipate business model changes. However, given the above, it is far too premature to size capital buffers on that basis.
• Banks should keep their ability to finance the transformational shift of the economy and the sizeable needs stemming from the transition plans expected from Public Authorities.
Q1: In your view, how could exposures associated with social objectives and/or subject to social impacts, which are outside the scope of this DP, be considered in the prudential framework? Please provide available evidence and methodologies which could inform further assessment in that regard.
The French Banking Federation (FBF) contributed to EBF response to this consultation and supports it. Moreover, we would like to emphasize the following key views (in this text box for technical reasons)We agree with EBA concluding remark that “Prudential regulation should remain risk-based and evidence-based". We also agree with EBA proposal that ESG risks are not specific risks, but ‘risk drivers’ that may impact “traditional” categories of risks such as credit, market, operational or strategic risk. We also fully share EBA views that historical data and evidence is not yet available to support an adaptation of the Pillar I framework.
ADDITIONAL CAPITAL REQUIREMENTS WILL HAMPER EU BANKS’ ABILITY TO FINANCE THE TRANSITION
• We believe that increasing banks’ capital requirements is not the right approach as banks need to be able to finance the transition of their clients, in a context of increasing financing needs to support the economy’s transformation shift. This is all the truer in the EU where the financing of companies remains mostly bank-loan based.
• “Front-loading” capital requirements would not necessarily deliver a tangible difference in resilience. Instead, they could constrain capacity to extend credit and investment to key sectors in need of transition finance, which could potentially ‘bring forward’ transition-related disruptions to the economy (e.g., exacerbating inequality by unaffordable energy prices), and negatively impact climate outcomes.
• Banks are part of the solution to achieve the objective of net-zero greenhouse gas (GHG) emissions in the EU economy by 2050 but they should not be the primary enforcers of the EU climate policy. There is a political responsibility in defining the relevant industrial and tax policies to ensure an orderly transition and limit transition and physical risk levels. EBA also strengthened in the Discussion Paper that “The primary responsibility and most effective tools for dealing with environmental-risk-related externalities lie within the remit of political authorities”. In this context, the priority is the definition by the European Union of a detailed transition path towards a decarbonized economy by 2050, at a granular level, by industrial sector and by country, considering the industrial implications of a successful transition. This is a key pre-requisite for the European banking supervisors to assess, as part of the Supervisory Review and Evaluation Process (“SREP”), the way banks adapt their climate risk management frameworks, as relevant, to support the financing of these transition paths.
• Last, in a globalised economy, punitive changes to EU banks’ prudential requirements would only result in a substitution of the financing, which will be taken over by non-EU banks and/or non- bank players, subject to less stringent regulatory standards. This may put the related risks beyond the reach of EU regulators and supervisors.
THE DEFINITION OF THE APPROPRIATENESS OF CAPITAL, IN TERMS OF NATURE AND CALIBRATION, IS NOT MATURE
• The Regulatory capital framework is designed to cover unexpected losses with a time horizon of 1 year, without the intervention of material mitigation and strategic actions. Banks, supervisors and regulators face huge challenges to include the long-term horizon and the forward-looking nature of climate, and more broadly, environmental risk factors.
• We share BCBS’s and EBA’s view that climate factors are not a new category of risk per se: they are ‘risk drivers’ of the existing prudential risk categories, especially credit risk, with a positive or a negative impact. However, given the nascent nature of the collective understanding of how the climate risk drivers may impact these traditional risks, it is deemed premature to define a regulatory capital treatment. Indeed, given gaps in the evidence-base, necessary data, and methodology, it is very challenging, at this stage, to calibrate risk parameters such as PDs and LGDs, in credit risk, taking into account climate related risks drivers. As long as robust risk-based methodologies have not been established and experienced, reliable counterparty data is not available and the results of supervisory exercises is not stabilised, it would be far too early to foresee any additional capital requirement.
• We also believe that non-risk-based Pillar 1 adjustments, such as the Brown Penalizing Factor or the Green Supporting Factor (GSF) would be pure political measures. The underlying objectives should be addressed through public policies or taxes and not through banking regulation. The anchor of the prudential framework must remain risk-based as rightly pointed out by the EBA in its discussion paper. The only consideration of the characteristic of a transaction being sustainable or not does not imply a low or high credit risk.
• The potential interplay between macroeconomic cycles and climate risk factors has not been established, so the use of macroprudential tools in this area would not be appropriate at this stage. In February 2022, the Financial Stability Institute (‘FSI’) highlighted in its Brief No. 16 ‘The regulatory response to climate risks: some challenges’ that applying the macroprudential framework to systemic climate-related financial risks is likely to be ineffective and potentially counterproductive for financial stability. It also notes that ‘macroprudential measures aimed at reducing exposures to carbon intensive firms and sectors may not always be conducive to reducing aggregate climate-related financial risks. In particular, a significant increase in capital requirements for brown exposures, by curtailing the availability of credit to carbon intensive industries would increase the vulnerability of those sectors and hinder affected firms from adjusting their business models’. Ultimately, a macroprudential buffer linked to climate could thus delay the financing of transition by banks, delay transition as a result and increase the potentially systemic nature of climate risk and financial stability as a result.
In addition, as highlighted below, proactive work is ongoing to adapt Pillar 2 frameworks, as need be, to capture climate-related impacts if deemed material. Hence, it will be key to ensure that overlap or double counting does not occur over time – as well-acknowledged by the EBA.
PILLAR 2 WOULD BE THE MOST APPROPRIATE FRAMEWORK, BUT THE DEVELOPMENT OF ROBUST CLIMATE SCENARIOS, METHODOLOGIES AND DATA ARE STILL GLOBALLY SHARED CHALLENGES
• Pillar 2 approaches could be considered, but stress testing is not the right approach to calibrate climate-related Pillar 2 requirements. Rather, FBF is supportive of climate scenario analyses, which are by design best suited to apprehend climate risk drivers, given their forward-looking nature. In this context, large French banks have participated in a number of industry-wide exercises, notably the French supervisor, ACPR climate stress test in 2020, which was done on a voluntary basis, and the ECB 2022 Climate stress test.
• Climate scenario analyses are useful tools to capture changes in business models that banks will need to undertake in climate scenarios. Indeed, transition scenarios may result in a rebalancing of economic activity across sectors and, within sectors, across counterparties. Climate scenarios are not designed, like traditional stress test scenarios, to measure losses, and thus capital, in adverse macro-economic circumstances. Indeed, some sectors or players in sectors can be expected to benefit from climate change. Climate scenarios are designed to anticipate sectoral evolutions that climate change will trigger and to help banks adjust to these changes accordingly. The horizon of climate change and of climate scenarios is also radically different from that of capital sizing stress tests. There are various initiatives to develop climate scenarios, such as those undertaken by the Network for Greening the Financial System (NGFS) or the International Energy Agency (IEA). However, while welcome, these scenarios are still incomplete. In particular, they do not offer a complete scenario framework including the modelling and projection of industrial and technological developments, as well as their translation into macro-economic variables. Until this is achieved, scenario analysis will remain exploratory and hypothetical in nature or even unrealistic on short-term horizons and not suited for capital sizing stress testing.
• Pillar 2, through SREP assessment, should focus on the quality of banks’ risk management framework in supporting well-defined transition plans from public authorities.
• The results published recently by the ECB of its 2022 climate stress test confirms the multiple challenges that both supervisors and banks face in developing reliable and plausible supervisory stress test exercises.
• The ECB notes that while banks make progress in developing climate stress testing infrastructures, many challenges remain for coherent results to be produced. Data availability and model developments remain unequal among banks, which is more a reflection of the specificities of climate risk (absence of historical observations, lack of uniform reporting of climate data by clients and, in the case of GHG emissions, lack of a uniform framework to report scope 3 emissions in particular) than banks’ reluctance to invest in climate stress testing frameworks.
• The approach by supervisors of climate scenarios is also widely different: for example, the ECB transition scenarios included a 3-year scenario, which is not the case of any previous climate stress test exercise; the ECB proposed two physical risk scenarios (flood and drought & heat) while the PRA considered that a “no policy action” scenario constitutes a physical risk scenario. It is very difficult for banks to develop frameworks in the absence of common scenario approaches by supervisors.
In addition, the time horizon of climate change inherently brings uncertainty to projections in climate scenarios. The ECB notes in its publication report that, as far as the 30-year transition scenarios are concerned, “it should be borne in mind that the 30-year projections are exploratory and subject to significant uncertainty. Therefore, these long-term loss projections should be interpreted as a qualitative yardstick for the direction of travel rather than as a robust quantitative measure.” The ECB also notes that, in the long-term scenarios, “the aggregate pattern of loan loss projections over time is masked by notable differences across banks with respect to projected losses in the long term”. This is particularly true as banks make different assumptions in balance sheet projections under a dynamic balance sheet methodology, which is the preferred option in 2050 projections and the only way to properly capture the business model dimension of climate stress testing.
• Most regulators acknowledge, the exercises are not robust enough yet in terms of data, scenarios, and methodologies. A progressive and iterative development of such methodologies, scenarios and data would enable banks to strengthen their risk management frameworks (for example through the building of risk and IT infrastructure and the development of climate specific scenarios) and effectively continue to include climate drivers in their Pillar 2 frameworks.
• In the meantime, a climate-related risk concentration framework, leveraging off the existing ones for pillar 2 and/or large exposures, could be useful to monitor specific areas of concentration and take actions, as relevant, to prevent a small number of financial institutions from unduly accumulating exposures to climate-related risks. However, any potential measure such as concentration limits should be cautiously considered to avoid harming specific regions or sectors and impeding efforts to scale up transition finance to sectors that need it most.
• Irrespective of the sector, clients demonstrating strong willingness to transition, with solid transition plans and commitments to reach net-zero by 2050 should be supported and best in class players should have incentives to pursue their transition shift.
CONCLUSION
• The focus should remain on bank’s adequate risk management and creating a policy environment that does not create disincentives and impediments to finance the transition. Tempting to move forward with “precautionary measures” such as capital buffers, would inevitably require a departure from the core risk-based foundations of the prudential framework and introduce subjective choices pertaining to broader policy objectives and potential unintended consequences.
• To cope with the challenges associated with time horizons, data, and methodologies, we believe that the banking industry together with regulators and supervisors should pursue dialogue and collaboration factoring in technical work and scientific research benefitting from industrial and sectoral experts.
• Scenario analysis is an appropriate tool to assess the consequences of climate change in various transition scenarios and anticipate business model changes. However, given the above, it is far too premature to size capital buffers on that basis.
• Banks should keep their ability to finance the transformational shift of the economy and the sizeable needs stemming from the transition plans expected from Public Authorities.