Bank of England report on climate-related risks and the regulatory capital frameworks

Executive summary

1. In October 2021, the Bank of England (the Bank) published its Climate Change Adaptation Report (CCAR), which set out early thinking on climate change and the regulatory capital frameworks for banks and insurers. The CCAR set out that the current frameworks already capture climate-related risks (climate risks) to some extent, including through capital models and credit ratings. However, risk capture may be incomplete due to difficulties in estimating climate risks (capability gaps) and there may be challenges in capturing risks in the existing capital regimes (regime gaps).

2. The Bank committed to undertaking further work on these topics. It has engaged with a range of stakeholders to inform this work, including through a call for research papers and hosting a high-profile, focused research conference, as well as undertaking its own internal work. This report sets out an update and key findings from that work. It does not set out any policy changes but sets out the Bank’s thinking and identifies areas for future work.

Key findings

• Existing capability and regime gaps create uncertainty over whether banks and insurers are sufficiently capitalised for future climate-related losses. This uncertainty represents a risk appetite challenge for micro and macroprudential regulators. Regulators, including the Bank, need to form judgements on whether quantified and unquantified risks are within risk appetite – and act accordingly.
• Effective risk-management controls within PRA-regulated firms (firms) can reduce the quantum of capital required in the future for resilience, but the absence of controls might suggest a greater quantum of capital will be required. As a short-term priority, the Bank is focused on ensuring firms make progress to address ‘capability gaps’ to improve their identification, measurement, and management of climate risks.
• The Bank has explored conceptual issues to better understand the nature and materiality of ‘regime gaps’ in the capital framework. The unique characteristics of climate risks mean that their capture by capital frameworks requires a more forward-looking approach than used for many other risks. Scenario analysis and stress testing will play a key role in this. Regulators, including the Bank, need to focus on the development of these frameworks and how they can inform capital requirements. Firms will be expected to make further progress in this regard.
• Current evidence suggests that the existing time horizons over which risks are capitalised by banks and insurers are appropriate for climate risks. Therefore, there does not appear to be sufficient justification for regulators, including the Bank, to make a policy change to these time horizons. The Bank will continue to explore how climate risks can be calibrated within the timelines embedded in existing capital frameworks.
• Further work is needed to assess whether there may be a regime gap in the macroprudential framework. Any use of macroprudential tools would need to be assessed carefully against how well they mitigate climate risks, their behavioural impacts, and the potential for unintended consequences. Calibration of macroprudential tools would also be challenging given uncertainties around climate risks and the need for them to help facilitate an orderly transition to net zero. The Bank will explore the nature and materiality of such regime gaps as part of its ongoing policy work and consider whether action to address them would be appropriate.
• Research on the conceptual challenges of incorporating climate risks into the capital frameworks appears to be limited based on submissions to the Bank’s call for research papers. Further research and greater public dialogue would therefore be valuable, notably in sparsely covered areas listed within this report. More detailed research/analytical questions and how to notify the Bank on future relevant projects is set out in the Annex.

3. The Bank will undertake further analysis to explore whether changes to the regulatory capital frameworks may be required. In particular, the Bank will progress work to:

• ensure firms continue to make progress to address capability gaps;
• build its capabilities and forward-looking tools to judge the resilience of financial system to climate risks;
• support initiatives to enhance climate disclosures;
• promote high quality and consistent accounting for climate risks;
• build understanding of and address material regime gaps in the capital frameworks;
• supervise against SS3/19 – ‘Enhancing banks’ and insurers’ approaches to managing the financial risks from climate change’, including work to build an understanding of banks’ evolving approaches to Pillar 2 add-ons; and
• engage in relevant discussions at international fora to inform domestic policymaking.

4. Diagram A sets out how this work fits alongside progress to date on work to ensure banks and insurers are resilient to the risks of climate change.

Diagram A: High-level timeline of the Bank’s key work on climate-related risks

Section 1: Overview of the Bank’s work on climate-related risks and the capital frameworks

5. In October 2021, the Bank published the Climate Change Adaptation Report (CCAR), which set out the Bank’s early thinking on climate-related risks (climate risks) and the regulatory capital framework. The purpose of the CCAR, and the work that contributed to it, was to deepen understanding of the links between climate change and the regulatory capital frameworks. This forms part of the Bank’s climate objective to ensure that the financial system is resilient to risks arising from climate change.

6. The CCAR set out that capital frameworks should be a core tool within the broader regulatory frameworks to ensure that PRA-regulated firms are resilient to climate risks.^{footnote [1]} This includes risks that might arise through two key channels – physical risks and transition risks. The CCAR noted that the regulatory capital framework already captures climate risks to some extent, for example through capital models and credit ratings. However, this risk capture is potentially incomplete due to capability gaps and regime gaps.

• Capability gaps refer to the difficulties inherent in identifying and measuring climate risks. These challenges might arise due to lack of relevant granular data from firms on their own climate risks, or limitations in modelling techniques to fully incorporate and estimate the impact of climate factors on their counterparties. These gaps encompass broader challenges related to firms’ climate risk management in areas such as scenario analysis, accounting, and regulatory valuations.
• Regime gaps refer to challenges in capturing climate risks due to the design or use of methodologies in the capital frameworks. For example, in microprudential capital frameworks, methodologies are typically calibrated using historical data to capture risks that crystallise over a relatively short-term horizon. This is often around a year ahead. While this ensures that capital is set in a standardised and quantifiable way, it might underestimate future climate risks that could emerge over longer horizons. Another example is that there is currently no explicit consideration of climate risks in the macroprudential framework.

7. The CCAR highlighted that identifying the nature and estimating the materiality of these gaps in the regulatory capital framework is complex. The CCAR proposed to develop further understanding of the following questions:

• where is it most appropriate to reflect climate risks within existing regulatory frameworks?
• what is the appropriate time horizon to reflect climate risks within regulatory frameworks?
• how to understand and manage uncertainty over the time horizon and scale of risks, including through the use of scenarios?
• how exposed are PRA-regulated firms or the system as a whole to these risks?

8. In the CCAR, the Bank committed to undertaking further work on these issues, drawing on lessons from the 2021 Climate Biennial Exploratory Scenario (CBES), insights from supervisory work, research, and a focused conference on Climate and Capital. It also committed to publishing an update on its thinking to contribute to domestic and international policy discussions.

The Bank’s work on climate change and the regulatory capital frameworks

9. This report sets out the Bank’s latest thinking on the focus areas identified in the CCAR. It has been informed by a range of Bank workstreams, including:

• The 2021 CBES exercise, which explored the extent to which PRA-regulated firms and the system as a whole are exposed to climate risks.
• Insights from experience of supervising against the PRA’s expectations of firms’ approaches to managing climate risks. A letter to CEOs published in October 2022 set out thematic observations on firms’ levels of embeddedness.
• A detailed review of written auditor reporting for major UK banks, which considered those firms’ progress in developing capabilities to capture the impact of climate risks on their balance sheet accounting valuations. A letter to Chief Financial Officers published in October 2022 set out thematic feedback relevant to both firms and auditors from this review.
• Research and a conference focused on conceptual issues identified in the CCAR. The Bank sought a diverse range of views from academia, international regulators, and the financial services sector to inform and advance thinking on these issues via a Call for Research in early 2022, which received over 60 submissions. This research fed into a Bank hosted Climate and Capital conference on 19-20 October 2022, which brought together a wide range of participants to discuss these conceptual issues further. A note to support the Bank of England’s Climate and Capital conference was published and included background information, key points from relevant research, and broader public commentary.

10. In parallel to this work, the Bank continues to engage and contribute to discussions in international standard setting fora such as the Basel Committee on Banking Supervision (BCBS), the International Association of Insurance Supervisors (IAIS), and the UN Sustainable Insurance Forum (SIF). The Bank’s understanding has also benefitted from the work of other international central banks and regulators – both work that has been independently undertaken and through the Financial Stability Board (FSB) and the Network for Greening the Financial System (NGFS).

11. This report summarises findings from the Bank’s work on climate risks and capital to date. It highlights updated thinking. Any future proposals to change the regulatory capital frameworks would be taken through the PRA’s formal policymaking process. This report also identifies areas where further work is needed and the Bank would welcome collaboration and independent insights from external parties in these areas.

Section 2: Key finding on materiality of gaps in the capture of climate risks by capital frameworks

12. Climate risks create a challenge for both microprudential and macroprudential regulatory authorities when assessing the risks relative to their risk appetites. Capability gaps create uncertainty over the distribution and materiality of the risks, both across individual firms and at a system level. This makes it difficult to form a view on whether the size of the risks are within regulators’ institutional risk appetites. Furthermore, combined with potential regime gaps, this could mean that firms are insufficiently capitalised with respect to future climate-related losses.

13. This challenge to regulators’ risk appetites might increase with time as the transition to net zero progresses and more physical climate risks crystallise. While some of the gaps will be addressed over time, particularly with progress on addressing capability gaps, they will not all be fully addressed for some time. Therefore, it will be necessary for prudential regulators to continue to form judgements on the extent that relatively unquantifiable risks are within their risk appetites and act accordingly to address them throughout this period.

14. The Bank will continue its work to encourage and support firms’ steps to address capability gaps, while maintaining a focus on whether climate risks are adequately captured by the capital frameworks. This will inform whether further policymaking or supervisory action is required.

15. The Bank will take a judgement-based approach to policymaking given the uncertainties about the adequacy of risk capture and capital required to ensure resilience to climate risks. This might involve taking a gradualist approach to policymaking as risk identification and measurement capabilities develop, while raising a question about the burden of proof required for policy action if regulators are faced with material uncertainties about resilience to climate risks.^{footnote [2]}

16. The Bank will also explore whether a broader range of prudential tools might be appropriate, including those that may be more robust to uncertainty than microprudential tools. This includes concentration limit approaches and broader macroprudential policies, and is explored in Section 4 of this report.

Section 3: Key finding on capability gaps in the capital frameworks

17. The prudential approach for all risks, including those from climate change, relies on firms’ abilities to identify, measure and manage risks. Banks and insurers may face losses that can be conceptually split into those that are “expected” or “unexpected”. Firms manage expected losses through actions such as robust and consistent financial/regulatory accounting or by taking strategic action to manage their exposures. The regulatory capital framework is used to provide resilience against the unexpected losses that might not be caught by these processes. A firm with an effective risk management system will likely be better able to identify, measure and manage expected losses from climate change. This might, in turn, better position those firms to take actions that would reduce the size or probability of unexpected losses.

18. The PRA expects firms to develop and embed effective risk management processes to manage their climate risks. Further progress by firms to address capability gaps is important in the short term to provide a better foundation for the identification, measurement, and management of climate risks. This will provide a better basis for identification of material regime gaps that need to be addressed and for the design and calibration of capital tools. Therefore, the Bank’s view is that addressing capability gaps is an important short-term priority.

19. In October 2022, the PRA noted in its letter to CEOs that while some firms have made considerable progress in embedding climate risks within their risk frameworks, the levels of embedding vary and all firms need to make further progress. While the PRA does not expect firms to have embedded an end-state framework for climate risk measurement at this time, it does expect to see continual progress and ambition.

20. Firms also have a responsibility to determine their own capital adequacy. The PRA highlighted in its letter to CEOs that firms’ internal capital adequacy assessment processes (ICAAPs) and own risk and solvency assessments (ORSAs) should now provide sufficient contextual information to allow a reader to understand the firm’s analysis of climate risk and capital. This includes disclosure of methodologies and underlying assumptions, judgements, proxies, and consequent uncertainties. Firms should be able to explain how they got comfortable that any material climate risks are appropriately capitalised. Where firms are not able to provide this analysis, the PRA will consider if further supervisory action is required to ensure any deficiencies are addressed appropriately.

21. Banks and insurers need high quality and comparable information about the climate risks faced by their counterparties to support robust risk assessments and valuations of their exposures. For example, for banks and insurers respectively, this includes whether borrowers and policyholders could be exposed to climate-related losses. In addition, market participants will need information from banks and insurers to understand the impact of climate risk on their financial and capital positions. The PRA will continue to support international and domestic efforts to promote the implementation of consistent and comparable disclosure standards for climate risks, including by the International Sustainability Standards Board (ISSB).

22. Accounting values are foundational to the banking capital framework. The development of capabilities to support high quality and consistent accounting practices for climate risks will help mitigate the risk of gaps in the capital framework. The PRA will play an active role in promoting high quality and consistent accounting for climate change. The letter to CEOs set out that firms’ financial reporting-related priorities for 2022 and beyond should include enhancing their climate-related data and modelling capabilities, governance, and controls. The Bank will continue to make use of the work of external auditors in reviewing firms’ climate-risk assessments and share thematic feedback to support robust planning and early action. Development in this area will also feed into the broader consideration of capital adequacy.

Section 4: Key findings on regime gaps in the capital frameworks

23. Climate risks typically manifest through traditional risk channels. However, they have unique characteristics that may have implications for how the risks are incorporated into regulatory capital frameworks for banks and insurers. These characteristics typically relate to the complexity, uncertainty, and likely unprecedented nature of the risks. For example:

• Historic data sets are not likely to be a good predictor of how climate risks may affect firms’ future losses;
• climate risks are simultaneously uncertain in where, when, and by how much they will impact firms and yet the fact that there will be impacts is foreseeable; and,
• the size and balance of future risks will be determined by actions that are taken now and in the future.

24. While the prudential framework is designed to help manage such uncertainty, research highlights that the uncertainty of climate risks may be more pronounced due to non-linearities and feedback loops that may emerge, including those relating to future policy actions by governments.

25. These characteristics give rise to a greater degree of uncertainty over measurement and timing of climate risks. As a result, the calibration of capital to mitigate climate risks is more complex than for other risks. For example:

• Modelling challenges: the lack of relevant historical data makes quantification of climate risks and the validation of predictive models more challenging. This challenge also applies to the calibration of capital for climate risks. The understanding of some physical risks is likely to improve over time, which can aid modelling. However, as transition risks will often be one-off, modelling the impacts relies on clarity over the transition pathway to net zero.
• Available information: the lack of comparable disclosure across the economy about exposures of non-financial firms and individuals to climate risks makes granular risk assessments and valuations by banks and insurers more challenging. Initiatives are underway to improve climate disclosures, including standards being developed by the ISSB to set a comprehensive global baseline.
• Reliable valuations and metrics: regulators rely on robust and consistent financial/regulatory accounting by banks and insurers to underpin the capital frameworks, with capital requirements predicated on valuations and expected losses being properly accounted for and deducted from capital resources. In the absence of robust and consistent financial/regulatory accounting, there is a risk of understating the contribution of climate risk to expected losses, leading to the overstatement of capital resources available to absorb unexpected losses from climate risks.

26. Forward-looking tools that are flexible to exploring a range of possibilities, such as scenario analysis and stress testing, will be necessary to understand the potential impact of climate risks on firms. Regulators have already used exploratory scenarios to try and understand the extent to which regulated firms and the system as a whole may be exposed to climate risks. However, these exercises have been necessarily exploratory, partial, and are limited by some of the capability gaps discussed in Section 3.

27. The Bank will continue to develop its strategy and capabilities to use climate scenario exercises to understand firms’ exposures to climate risks and drive further improvements in risk management. In particular, it will seek to develop a deeper understanding of the materiality of risks that were outside the scope of the CBES, such as trading losses for banks. It will also continue to contribute to the work underway at the NGFS on sizing risks by developing short-term adverse scenarios.

28. Following publication of the 2022 CBES results, a PRA letter to CEOs set out thematic feedback on how to enhance scenario analysis and further embed SS3/19 expectations. In recognition that this feedback will take time to embed, the Bank will not launch a concurrent exercise that explores climate risk stresses in the near-term. Over this period, the PRA expects firms to demonstrate improvements in their internal capabilities and progress will be reviewed as part of the supervisory process.

29. The Bank will also continue to explore over the medium-term how the use of climate scenarios can support existing or new approaches for calibrating capital, including through the Bank’s approach to stress testing the UK banking system.^{footnote [3]} In doing so, it will consider a number of underlying issues, for example:

• The extent to which known expectations of how physical and transition risks will unfold over the typical stress test horizon could be incorporated into the baseline scenarios used within stress testing, and if so, how to calibrate those changes. For example, this might include incorporating the impact of already announced government transition policies, such as energy efficiency requirements on commercial properties, into the stress test.
• Exploring different severities and forms of climate scenarios, as well as how they could be used by firms and the Bank to calibrate capital requirements. Stress tests typically use a scenario that tests resilience against a severe but plausible stress. Consideration needs to be given to the extent that this single scenario approach is appropriate to calibrate capital given the uncertainties around the evolution of climate risks.
• Determining how to assess climate risks on a dynamic balance sheet basis. The loss projections used in the CBES assumed that participants’ balance sheets were fixed – allowing the exercise to assess the vulnerability of today’s business model against future shocks. This approach does not lend itself to the calibration of regulatory capital as, for example, firms might take actions over the stress test time horizon to reduce the climate risks they face.
• Reducing variation in firms’ modelling calculations. This is important given the variation in losses modelled under the CBES and the associated uncertainty of climate risks, linking back to challenges arising from capability gaps.
• The conceptual case for using scenario analysis and stress testing to test the resilience of the financial system to climate risks, which may then inform the calibration and use of macroprudential tools to mitigate those risks.

30. The PRA will also continue to supervise firms’ progress to develop capabilities for using scenario analysis and stress testing for climate risks in their ICAAPs and ORSAs. The Bank has identified that further work is needed by firms as existing capabilities are not sufficiently well developed to support the internal calibration of capital and broader decision making. The PRA’s supervision will evolve over time considering firms’ approaches and will be informed by the broader work on scenario analysis and stress testing highlighted above.

31. The CCAR asked whether the existing typical one-year time horizons for setting capital are appropriate for climate risks, given the risks might crystallise over longer time horizons than other risks. It highlighted a number of factors to consider in exploring this issue, for example: whether changes to the framework might give rise to unintended consequences; and whether differences in approach might be considered for different parts of the capital framework. The Bank has considered this issue further.

32. Regulatory capital frameworks have been calibrated to ensure firms are resilient against the risks of near-term unexpected losses to given degrees of stress.^{footnote [4]} Under the current frameworks:

• Both the banking and insurance regulatory frameworks largely calibrate microprudential capital requirements over a one-year period. This takes into account the period over which firms might need to respond to stresses, including through recapitalisation, as well as a number of practical considerations such as data availability.
• Longer time horizons are considered to ensure that firms maintain adequate financial resources against long run and stressed events. For example, banking stress testing frameworks often consider a time horizon of 3-5 years. This is intended to ensure that firms are able to absorb losses that may arise in a given year under a severe stress scenario. The period is intended to be consistent with the duration of losses that firms might face from a typical macroeconomic shock.
• While the capital frameworks explicitly focus on shorter-term time horizons, their calibrations also implicitly capture forward-looking information on expected and unexpected losses beyond these horizons. For example, banks and insurers use information such as credit ratings, which may reflect views on future risks including from climate change. Bank capital resources also capture, to some extent, a forward-looking view of climate risks through accounting, including by using market consistent valuation approaches for trading positions, and by capturing expected credit losses on loans under IFRS 9. Under Solvency II, the value of an insurer’s liabilities reflects the value of expected future cashflows over the duration of the insurance contract (which in some cases can be over 30-50 years).

33. A change to the time horizons used to set regulatory capital would represent a fundamental change to the regulatory capital frameworks. As the existing regulatory capital frameworks are already used to manage long-term risks, any justification would need to explain why they do not adequately address the idiosyncrasies of climate risks.

34. The conceptual arguments put forward by some commentators (including at the conference) for extending the time horizons used to set regulatory capital for climate risks are to avoid cliff-edge effects and to support a more long-term assessment of risks. For example, a different approach might be justified by the non-cyclical nature of climate risks and challenges in capturing the risks through current exposures and valuations (meaning that firms might be undercapitalised now). Earlier recognition of the risks in the capital frameworks could help to incentivise longer-term risk measurement and mitigation, which may in turn support a reduction in future risks if firms take strategic actions to respond to those risks in the meantime.

35. On balance, the Bank does not think that the arguments for changing the time horizons for regulatory capital setting are supported at this point. This is consistent with the majority of views put forward in the call for research papers and conference discussions. The following considerations contribute to this view.

• Time horizons have been calibrated to provide time for firms to recapitalise following a loss. It is not clear that firms would need longer to recapitalise for climate risks than other risks. There are also tools within the existing frameworks that could be used to test firms’ resilience to the risks in a given year over longer time horizons (for example, stress testing), reducing the likelihood of cliff-edge effects.
• Increasing use of forward-looking information to capture climate risks in existing regulatory frameworks in a timely way also reduces the need for capitalising against climate risks arising over an extended time horizon. As new information on future physical or transition risks becomes available, including as a result of initiatives to enhance measurement and disclosure of climate risk exposures, firms should be able to better recognise expected losses through accounting and market valuations that feed into capital resources. Actions by firms to address capability gaps should thereby help to reduce the risk of firms’ capital resources being overstated relative to the risks that they face. Over time, climate risks will also be better caught within forward-looking components of firms’ capital requirement calculations, including through the use of external and internal credit ratings.
• Firms are also already expected to evidence how they would manage risks over longer scenarios than those in the capital frameworks. For example, banks and insurers are already expected to consider longer-term adverse scenarios in their ICAAPs and ORSAs and be able to evidence how they would manage those risks appropriately. Therefore, when climate risks are fully embedded within firms’ risk management frameworks, they should be able to evidence how they would manage those risks over all the relevant horizons.

36. Changing the regulatory time horizons for capital setting would also introduce challenges and costs for firms that need to be appropriately considered as follows.

• Extending the regulatory time horizon would lead to earlier recognition of risks in the capital frameworks, which would (all else equal) increase capital requirements for banks and insurers. The size and impact of the resulting costs (relative to any benefits) are challenging to quantify now given the uncertainty over when, and to what extent, climate risks might crystallise in practice for a firm. This may have unintended consequences for the transition to net zero. For example, for insurers, it may increase the so-called ‛protection gap’ for some risks, with a withdrawal or limitation in cover for certain types of risks (see work to understand regime gaps in the insurance capital framework for more on this issue).
• Extending the regulatory time horizon may amplify the existing capability gaps with modelling and capitalising against uncertain future climate risks. For example, the breadth of possible transition pathways would likely increase over a longer period, introducing greater uncertainty into firms’ modelling of climate risks.
• Changing the regulatory framework for climate risks alone could introduce greater complexity into the overall prudential framework. For example, given that climate risks typically manifest through traditional risk categories, there would need to be a way to apply a different time horizon for climate risks within those existing risk categories. This complexity would need to be justified alongside the coherence of the framework.

37. The Bank will continue to evaluate any further evidence on timeframes as it becomes available.

38. The Bank has worked to deepen its understanding of regime gaps by considering the conceptual links between climate risks and the different elements of the banking and insurance capital frameworks. The Bank has also considered where different manifestations of climate risk might be optimally reflected in the capital frameworks, and the latest evidence on where gaps may be most material. This section summarises findings across these different areas.

Conceptual factors that influence where climate risks might be reflected into capital frameworks

39. The following considerations are relevant to determining where climate risks might be best reflected into the microprudential and macroprudential capital frameworks:

• Policy actions should take into account the differences in how banks and insurers may be impacted by climate risks and the policy tools available in the banking and insurance capital frameworks to mitigate those risks. For example, insurers may be impacted by climate risks in different ways to banks, given that their liabilities can also be impacted by climate risks. The policy tools in the insurance capital framework also differ, for example, the insurance regime does not explicitly require capital to be held for macroprudential purposes. It is important to understand these differences as part of assessing how best to address gaps in each framework. See Box A below for more information on these differences.
• The extent that climate change creates systemic risks that will not be addressed by microprudential capital requirements alone. A macroprudential response may be justified by the fact that climate change creates foreseeable risks at the system level, but these risks are largely unquantifiable and the Bank is unable to identify which firms they will impact. Furthermore, climate change might create or increase market failures and externalities. For example, individual firms may not take into account the impact of their actions on markets or other firms (eg concurrent actions by firms to divest from carbon intensive sectors). Further work is also needed to understand the impact of climate change on factors such as information asymmetries, moral hazard, and free-rider effects. At the same time, the potential for unintended consequences of any macroprudential action should also be considered, for example, whether it might lead to a transfer of risk outside the regulatory perimeter or lead to behavioural actions by firms that might not support the transition to net zero. These issues would benefit from further research to inform ongoing policy work.
• Retaining coherence within the broader capital frameworks and broader policymaking objectives. Given climate risks typically manifest through existing risk drivers, any changes to capital frameworks should avoid making the overall framework less coherent. As part of this, it will be important to assess the conceptual link between climate risks and the underlying policy objective of the relevant part of the framework, as well as the relative benefits of adjusting to particular parts of the framework. This might include for example, assessing whether to reflect climate risks into minimum capital requirements or buffers within the banking capital framework. This choice has implications for policy issues such as the availability and usability of capital in a stress.^{footnote [5]} Finally, any changes should be consistent with broader regulatory reform initiatives, such as work to design a ‘strong and simple’ regime for small banks and building societies, and the outcome of the PRA’s upcoming review of its Pillar 2A methodologies.
• Institutional remits, and the trade-offs that they provoke. A financial authority with a microprudential remit might make a different choice to one that also has a remit that also covers wider financial stability. The Bank would consider its full set of objectives and matters which it must have regard when considering this issue. For example, the PRA has a general objective of promoting the safety and soundness of the firms it regulates, through ensuring that firms’ business is carried on in a way which avoids any adverse effect on the stability of the UK financial system and minimising the risk they pose to financial stability.^{footnote [6]} The PRA also has an objective to contribute to the securing of an appropriate degree of protection for insurance policyholders and considers the impact of protection gaps on policyholders, as well as insurers’ resilience to climate risks.

Box A: Conceptual differences between climate change and the banking and insurance capital frameworks

1. Insurance companies are impacted by climate risks in different ways to banks. It is important to understand these differences when considering how and to what extent climate risks are adequately captured by the different capital frameworks.

Diagram B: Stylised regulatory balance sheet comparison

Some of the different ways that insurers can be impacted by climate risks include the following.

• Climate risks can have an impact on the value of an insurance company’s assets and liabilities. However, climate risks typically have their most pronounced impact on the value of a bank’s assets. For example, the value of an insurer’s promise to pay someone the value of their house changes if the house is destroyed by flooding or subsidence caused by climate change. However, a bank’s promise to pay someone the value of their deposit is not impacted by climate risks.^{footnote [7]}
• Insurers’ liabilities are based on regulatory requirements that rely heavily on actuarial calculations and judgement. Under Solvency II, an insurers’ assets and liabilities are measured for regulatory purposes at market value which typically differs from their accounting value, unlike for banks. However, valuation of an insurers’ liabilities is complex because the market price of a liability rarely exists. The Solvency II risk margin is intended to add to the central estimate of the value of liabilities (‛best estimate’) a margin, akin to a risk premium, with a view to reach a total value that is equivalent to the value the market would demand to purchase the liabilities from the insurer. As a result, and given the absence of market prices, the valuation of liabilities relies on judgement, modelling, and actuarial expertise. This is important because a policy change that did not appropriately reflect the complexity of valuations could result in inconsistent treatment between firms, with the potential for undercounting of risks.
• The Solvency II regime is based on an overall regulatory balance sheet approach and there can be complex interactions between insurers’ assets and liabilities. Both the assets and liabilities of an insurer are stressed simultaneously to calibrate capital requirements. As a result, correlated risks could negatively impact both the assets and liabilities of an insurer at the same time (ie reduce the value of an insurer’s assets while also increasing the value of its liabilities). For example, a general insurer that underwrites claims for windstorms in one country may also hold investments that may be indirectly affected by a flood in another country eg through supply chain disruptions. Emerging evidence suggests physical risk events could become increasingly correlated as a result of climate change, which may lead to instances of greater exposures to physical risks on both an insurers’ assets and liabilities.
• Insurers could also be vulnerable to climate risks in more ways than traditional banks. For example, because they underwrite protection for physical risks such as natural catastrophe cover.

2. The banking and insurance capital frameworks contain different policy tools to manage risks. This means that different approaches might be required to address any material regime gaps in the capture of climate risks across the different capital frameworks.

Diagram C: Stylised capital framework comparison

For example, an important difference between the banking and insurance capital frameworks is that there is no explicit role for buffers within the insurance capital framework.

• From a macroprudential perspective, there is no explicit macroprudential insurance buffer framework that regulators can use to mitigate systemic risks on an industry-wide basis. Macroprudential risk is captured implicitly, and to an extent, within the Solvency II valuation framework. The Solvency II matching adjustment and volatility adjustment can act as countercyclical tools that dampen volatility in a stress, mitigating the risk of asset fire sales. As a result, they might help mitigate market volatility that could occur as a result of disorderly policy action. However, they are calculated by individual insurers and so are not a macroprudential tool that can be used to set additional resilience for systemic climate risks more broadly.
• From a microprudential and macroprudential perspective, there is a parallel (to an extent) between the capital conservation buffer and the split between the Solvency Capital Requirement and Minimum Capital Requirement, given the graduated implications for supervisory action (including on retention of distributions to maintain resilience). However, unlike in the banking framework, the results of the Insurance Stress Test (IST) do not inform capital setting.
• Under the existing Solvency II framework, any policy action to address gaps in the capture of climate risks may need to address both microprudential and macroprudential risks together. Therefore, as understanding of the systemic risks of climate change improves, it will also be important to understand whether the existing framework sufficiently allows for the mitigation of systemic risks, or if this represents a material regime gap to address.

Update on work to understand and address regime gaps in the banking capital framework

40. This section sets out emerging evidence on the potential regime gaps in banking capital framework and ongoing work to address those gaps.

Microprudential use of Pillar 1

41. The Bank is actively contributing to work at the BCBS Taskforce on Climate-Related Financial Risks on the extent to which climate risks can be addressed within the Pillar 1 framework, whether there are regime gaps, and possible measures to address those gaps. As part of this work, on Thursday 8 December 2022, the BCBS published a set of FAQs with clarifications on how climate risks may be captured in existing Pillar 1 standards.

42. The Bank has also begun to look at the evidence for climate-related gaps in the measurement of risk-weighted assets (RWAs). An adjustment to the RWA framework might be justified on a prudential basis where climate change affects the relative riskiness of assets.^{footnote [8]} The challenges here are extensive as data and models remain limited. Further work is required to explore the correlations between risk and assets of varying greenhouse gas intensity or exposure to broader climate risks – and importantly how those risks might change over time.^{footnote [9]}

43. Firms may also capture any changes in risk across assets through their own modelling as they overcome capability gaps. In that case, the need for changes to the regulatory framework might be reduced. There are distinct challenges in capturing climate risk in the Internal Rating Based (IRB) or Standardised Approach (SA) to credit risk. For example, under SA, external ratings are used for corporate RWAs but not for retail RWAs. Therefore, to the extent that external ratings capture climate risks, there may be greater challenges in capturing climate risks in retail exposures. For IRB models, climate risks could potentially be reflected through modelling of risks to exposures, although data and modelling challenges are likely to remain. The Bank will continue to build its understanding of the extent to which climate risks are captured in firms’ modelling through its core supervision and policy work.

44. Some external commentators have suggested that the PRA should intervene to require firms to have more capital against assets exposed to carbon-intensive energy sources now, either directly or through a penalising factor. This is because they argue that these assets will necessarily become less economic and more risky as the UK transitions towards net zero. The PRA does not believe that such an intervention is currently supportable on prudential grounds as risks of individual assets will be correlated with the economic viability of the assets, and this will in turn be influenced as the detail of countries’ transition and energy strategies becomes clear. In addition, the assets in question might only become more risky beyond the time horizons currently used for regulatory capital setting. This area will be kept under review, and calibrations will continue to be informed by government policies and firms’ own risk assessments.

45. Accounting values are foundational to the Pillar 1 framework for banks. As noted above, the Bank has begun to consider progress by banks to develop capabilities to capture climate risk in accounting values that feed into Pillar 1 capital resources, and will continue to play an active role in promoting high quality and consistent accounting for the risks of climate change. This work will help improve the measurement of exposures and mitigate the risk of gaps in the Pillar 1 framework that would need to be filled by regulatory interventions. Any policy changes to address regime gaps should also consider the interaction with the accounting framework and whether the change might lead to a double counting of climate risk initially, or over time.

Microprudential use of Pillar 2

46. Banks are expected to be able to explain how they are comfortable that any material climate risks are appropriately capitalised for. In particular, climate risks that are not adequately captured within the existing Pillar 1 framework should be considered by banks in their ICAAPs under Pillar 2. The Bank’s view is that the Pillar 2 framework will be a key tool to address microprudential regime gaps within the existing regime, while policymaking discussions on gaps in the Pillar 1 framework take place internationally at the BCBS.

47. The PRA has set out that this is an area that firms need to make further progress as they address capability gaps. As noted in the letter to CEOs, the methodologies that firms used to demonstrate they have adequate capital against material climate exposures were generally maturing. The PRA is aware firms are likely to face ongoing challenges in measuring climate risks and related capital needs due to gaps in disclosure by counterparties across the economy, the associated paucity of data, and the absence of established best practices. In such cases, firms should use reasonable proxies, assumptions, and judgement to work around these issues and not leave known material risks uncapitalised.

48. The PRA will continue to assess whether firms are capitalising for all material climate risks and take action where necessary. Further work is needed to better understand the risks that are not being captured by Pillar 1 and, therefore, should be captured within Pillar 2. In addition to ongoing international work, this could be an area for further research. The PRA will continue to build its understanding of firms’ evolving approaches to Pillar 2 capital add-ons through ongoing supervisory review and evaluation process and, based on that, make an assessment of whether any changes to Pillar 2 methodologies for these purposes are warranted.

Macroprudential framework

49. There is a growing body of evidence on the systemic nature of climate risks, but work to understand macroprudential climate risk channels is still developing. These channels are complex and efforts to size them are unevolved.

50. To aid its assessment of where risks should be capitalised, the Bank is monitoring governments’ plans for their transitions to net zero. In theory, the more clarity that is provided on the transition to net zero, the more likely it is that transition risks could be captured by the microprudential framework. For example, clarity of transition pathway will make it easier for entities to measure and assess their climate risk exposures and it will allow banks and insurers to gather better data on counterparties’ transition plans and their evolving climate exposures. This will support banks to incorporate climate risk exposures into their risk-management processes. However, inaction, or a lack of clarity on the transition to net zero (particularly given the complex pathway of dependencies on the transition) might lead to a build-up of risks across the financial system that are not adequately captured by Pillar 1 or Pillar 2 microprudential capital requirements. This might justify the use of macroprudential tools as a way to ensure that the financial system is resilient to the risks. It is noted that even where full clarity of transition polices is provided, there may still be system-wide risks to address. For example, a build-up of physical risks might also warrant the use of macroprudential tools. Understanding the scale of these risks and the potential for unintended consequences should action be taken requires more work.

51. The current application of macroprudential tools is unlikely to fully capture climate risks as a result of the unique nature of climate risks, which build gradually and continuously over longer time horizons. Consequently, some commentators argue for refinement of existing macroprudential tools or the development of new tools that can more effectively capture climate risks. This could represent a regime gap that needs to be addressed.

52. To the extent that additional resilience is needed to mitigate systemic risks, calibration of macroprudential tools will be challenging in practice given ongoing uncertainty over the size of the risks. Care also needs to be taken to ensure that the calibration and use of macroprudential tools help to facilitate an orderly transition to net zero rather than acting against it, as a result of unintended policy consequences. Work is needed to consider whether further tools or approaches would help ensure that the policy framework is robust to the uncertainty of climate risks, alongside traditional risk measurement approaches. For example, in addition to the macroprudential framework, this may include the development of approaches that manage system-wide risks by preventing a build-up of risk concentration through limits or disincentives to the build-up of excessive climate-related exposures. In addition, the potential for unintended consequences of actions to address these risks needs to be central to any consideration. Some illustrative questions that need to be addressed prior to any policy development are set out in Table 1.

Table 1: Questions to explore in any future consideration of policies designed to manage the build-up of system-wide risks in the banking framework

53. In addition to the considerations set out in Table 1, a common consideration across all approaches to managing system-wide risks is the impact of the Bank’s policy action on the transition to net zero. While the Bank wants to ensure that institutions and the system are resilient to climate risks through this transition, it also wants to ensure that the move to mitigate risks is orderly, and does not unnecessarily create stranded assets, or other system-wide risks. For example, actions to increase banks’ resilience to systemic risks via a macroprudential tool may create behavioural lending effects that exacerbate transition risks or create risks that spillover outside the regulatory perimeter. This risk of unintended consequences was highlighted within the Bank’s CBES exercise. Therefore, consideration would need to be given to how policy tools can be calibrated in a manner that mitigates all these system-wide risks.

54. The open questions around how best to mitigate systemic risks from climate change will take time to answer. However, the longer firms might have to adjust to any changes, the smaller any unintended consequences might be. Therefore, while further work is required to establish whether changes to the capital frameworks may be required to provide additional resilience to systemic risks, it is important that progress is made in working through these issues soon. There is particular value in this issue being explored within international fora as well as domestically. The Bank’s ongoing work in this space will draw on international discussions at the FSB and the NGFS, for example, in relation to better understanding short-term stressed climate scenarios.

55. Further research on the trade-offs on the use of macroprudential tools, including on how to ensure resilience to systemic risks and facilitate a smooth transition, as well as on the pros and cons of different approaches to managing system-wide risks would be welcome to inform further thinking on this issue.

Update on work to understand and address regime gaps in the insurance capital framework

56. The Bank received materially fewer research submissions on regime gaps in the insurance capital framework. As a result, this is an area that would benefit from further research and analysis to inform the Bank’s ongoing policy work.

57. In the absence of detailed research and evidence on potential regime gaps, the Bank explored the conceptual links between climate change and the insurance regulatory capital frameworks. As part of this, it considered how differences between the banking and insurance capital frameworks, some of which are illustrated in Box A, might lead towards different approaches to regime gaps in the two frameworks.

58. The Bank also undertook early work to consider whether there might be conceptual regime gaps in the capture of climate risks within the existing Solvency II framework. This early work suggests that the Solvency II framework’s relative flexibility should support its ability to capture climate risks. However, the Bank will continue to progress analysis in areas where there could be gaps to address.

Microprudential considerations

59. The Bank will continue to explore whether there may be any material regime gaps in microprudential capital requirements for insurers that need to be addressed.

60. In particular, the Bank will explore whether any of the calibrations and/or definitions in the Standard Formula calculation for capital requirements need to be adjusted to take account of climate risks over time. For example, the existing framework has no explicit reference to climate risks, which should be captured through existing risk channels, and some natural catastrophe events linked to climate change are not currently captured but may need to be in the future. As more information about the nature and materiality of risks becomes available, it may become necessary to make adjustments to these calibrations.

61. The Bank will also consider whether the way firms assess risks within the matching adjustment (MA) adequately allows for the capture of climate risks.^{footnote [10]} Following the Solvency II review, the current design and calibration of the MA will be retained by HM Treasury in legislation, and the PRA will implement various supervisory measures to ensure firms consider the appropriateness of the resulting MA benefit for their portfolio. Within this framework, the Bank will explore, for example:

• whether and how climate risks are reflected in external credit ratings (or firms’ own internal ratings). The MA is designed to absorb market movements in asset prices that are not accompanied by a credit rating downgrade. If the riskiness of an asset increases as a result of climate change but this is not reflected in the asset’s credit rating, then the benefit of the MA could be too large as it does not reflect climate risks; and
• how firms consider whether the calibration of the MA for their portfolios sufficiently allows for the distinction between assets that might necessarily become more risky over time during the transition to net zero and other non-financial firms. The standard MA framework is also based on calculations that use 30 years of historical data, which, as identified in this report, is unlikely to sufficiently reflect future climate risks.

Macroprudential and broader considerations

62. In line with its mandate, the Bank will continue to consider whether insurers might be vulnerable to systemic risks from climate change that might not be captured by the existing Solvency II framework. As noted in Box A, this is an area where policy tools are less immediately available. As a next step, the Bank will continue to build its understanding of how systemic risks of climate change could affect insurers. Then, following this analysis, the Bank will consider whether further policy tools may be necessary to mitigate those risks for insurers.

63. The Bank’s work has also given rise to a broader question about whether further policy tools may be needed to address wider prudential issues arising from climate change for insurance companies. This includes the vulnerability to ‛protection gaps’.

64. A protection gap captures the difference between the amount of insurance that is economically beneficial for both policyholders and insurers, and the amount of coverage actually purchased or offered. The Bank’s CBES found that (in the No Additional Action scenario) households and businesses vulnerable to physical risks would be significantly affected, as general insurers might either pass on the cost of higher claims into premiums, or otherwise refuse to renew insurance for some customers.

65. It is important to consider this issue for two reasons. First, policymaking to address regime gaps for climate risks should also be sensitive to the possible unintended consequences on the availability and affordability of insurance and reinsurance. Second, the Bank will also continue to assess whether climate change itself might create or increase protection gaps, which in turn could have an impact on policyholder protection, or possible macroprudential implications.

66. This is an area where further research and consideration is needed. In particular, the following areas of research on the insurance capital framework would be beneficial:

• regime gaps in the insurance capital framework;
• analysis to enhance the regulatory view of protection gaps, including whether the regulator has an appropriate view over the time horizon for insurers’ strategies as part of the ORSA; and
• the potential exacerbation of market failures and externalities in light of climate risk.

67. The Bank received over 60 submissions to its call for research papers but only a fraction directly addressed the questions identified to help understand the materiality of any gaps in the capture of risks by the regulatory capital frameworks for banks and insurers. Of the papers received, those that explored microprudential policy typically focused on evidencing climate risk differentials for specific assets (eg residential mortgages). Those that explored macroprudential policy typically considered the evolution of systemic risk channels. These papers helped to inform the agenda of the research conference and consideration of these issues, but as highlighted in this document, there were key areas that were not covered and questions that remain unanswered.

68. Regulators are making progress to understand whether and where supervisory or policy actions are needed in this space. Further targeted research will be essential to inform ongoing thinking. In particular, the Bank has identified the following areas as priorities to make progress and as such would welcome further research:

• Cross sectoral issues on risk measurement and modelling, such as how to effectively design and calibrate forward-looking approaches for capital setting, particularly in the period where there is insufficient clarity on transition paths.
• The microprudential framework for banks, including evidence for where risks might not be adequately reflected in Pillar 1 and so should be caught by the Pillar 2 framework.
• The macroprudential framework for banks, including on the extent that existing macroprudential tools capture climate risks, whether existing tools might be appropriate or whether new tools are needed, including concentration limits. Research on how to balance resilience to systemic risks against the need for a smooth transition and the potential for unintended consequences would also be welcome.
• The insurance capital framework, including evidence on possible regime gaps, as well as broader issues around the potential exacerbation of market failures and externalities, and the issue of protection gaps.

69. A more detailed set of questions and information for how researchers can contact the Bank on their research are included in the annex of this report.

Section 5: Conclusion and next steps

70. This report updates on the Bank’s work on climate risk and the capital frameworks, building on the analysis set out in its Climate Change Adaptation Report (CCAR). It sets out areas where the Bank’s work has progressed since the CCAR, including on time horizons, ways of tackling uncertainty, and the impacts of climate risks on the existing microprudential and macroprudential frameworks. Substantial further work is needed and there remain many open questions, notably on potential regime gaps to capture systemic risks from climate change and unintended consequences. The Bank will continue to address these questions as part of its supervision and policymaking.

71. It is important to progress this work to ensure firms and the system remain resilient to the risks of climate change. Therefore, the Bank has identified a number of next steps and will:

• Maintain its focus on ensuring that firms make progress to address capability gaps to improve the identification, measurement, and management of climate risks. This is the fundamental first step in any work on this topic.
• Further develop its capabilities to test the resilience of the financial system to climate risks. This includes how scenario exercises and stress tests can help the Bank and firms understand the exposure of the financial system to risks not captured by the CBES. The Bank will work domestically and internationally to develop the toolkit required to facilitate this analysis.
• Progress work to understand material regime gaps in the capital frameworks. The Bank will consider the potential systemic risks of climate change and explore whether any changes to the macroprudential framework might be appropriate.
• Support ongoing international and domestic initiatives to enhance climate disclosures by the wider economy needed by banks and insurers to manage and measure risk, including the adoption of ISSB standards.
• Play an active role in promoting high quality and consistent accounting of climate risks, given accounting is a foundational building block of the capital framework for banks.
• Continue business as usual supervision against firms’ progress at embedding supervisory expectations in these areas as set out in SS3/19. This will also involve work to build an understanding of banks’ evolving approaches to Pillar 2A capital add-ons through ongoing supervisory review and evaluation process. It will also consider whether further guidance or updates to the expectations set out in SS3/19 are warranted over time.
• Alongside this, the Bank will continue to engage in relevant discussions at international fora on how regulatory frameworks and supervisory practices need to be adjusted to account for climate risks.

72. The question of whether the capital frameworks adequately capture climate risks is complex. The Bank’s work to date has benefited significantly from research submitted to the Call for Papers in 2022. The Bank continues to welcome further analysis and evidence from researchers and academics to supplement thinking as this work is taken forward. In particular, research would be beneficial on the financial stability implications of climate change for the banking macroprudential framework and to the insurance framework.

Annex: Areas where further climate-related research would be welcome

1. As set out in the report, substantial further work is needed on the issue of climate change and capital frameworks. The Bank’s work on this issue has greatly benefited from research and views of external stakeholders. Therefore, the Bank would welcome continued collaboration and independent insights from external parties, including academia, regulators, and financial services firms on areas where further work has been identified as being needed.

2. As part of this, the Bank is interested in hearing about new and evolving climate-related research and analysis. In particular, climate-related research focused on the macroprudential framework, measuring, and modelling climate-related risks and insurance would be welcome. Table 2 gives further details about specific questions of interest.

3. If you have ongoing research or plans for new research on these questions that might be of interest to the Bank, please send us details of your work to
ClimateCapital@bankofengland.co.uk.

Table 2: List of research/analytical questions that are of interest to the Bank