Skip to main content

Climate and bank capital requirements: where now and where next?

At a glance

  • Integrating climate into Pillar 1 bank capital requirements is complex. In our view, the current design of the Pillar 1 framework, and the technical and political challenges that would accompany a fundamental re-design, mean that policymakers have little appetite to revise the globally agreed Basel framework in the short term.
  • That said, our conclusion from the BoE’s conference on climate change and capital is that supervisors will prioritise three areas to ensure that banks are resilient to the crystallisation of climate risks. These are: (1) using Pillar 2 to accelerate improvement in banks’ risk management capabilities; (2) continuing to put pressure on firms to take the initiative to gather relevant counterparty data; and (3) driving improvement in scenario analysis and modelling capabilities, with a view to integrating climate risks into solvency stress testing in the medium term.
  • Pursuing these objectives will allow supervisors to gain some comfort that banks are holding adequate capital against climate risks that are likely to crystallise in the short to medium term, and make the eventual introduction of a Pillar 1 treatment more realistically achievable.
  • This provides a clear indication of the priority for banks – to continue to fill the data and capability gaps identified by supervisors in the thematic reviews undertaken by the ECB and PRA this year.
  • In the meantime, we think it is possible that financial stability authorities use macroprudential tools such as exposure limits, sectoral capital requirements or the Systemic Risk Buffer, in particular to manage risks caused by concentrations and changes in the accessibility of credit arising as a result of “herd behaviour”.

Regulatory work is ongoing across Europe and internationally to explore how climate risks should be best captured in the capital framework. There is live debate on the integration of climate risks into the Pillar 1, Pillar 2 and macroprudential frameworks, and whether or how these frameworks might need to change to ensure that banks are sufficiently resilient to such risks. The Bank of England recently hosted a conference on the topic, which has reinvigorated the debate. This blog distils our view of what we have learned from EU and UK policymakers’ analyses, and sets out our view on policymaker and supervisory priorities in the near-term.


Integrating climate risk into the Pillar 1 capital framework is technically and politically complex

The EBA discussion paper on the prudential treatment of environmental risks highlighted some of the technical challenges and demonstrated that there are limited easy options available in Pillar 1. In our view, the Pillar 1 framework will be a difficult vehicle for fully addressing climate risks without a significant redesign to capture all risks in a more forward-looking manner.

Policymakers are considering the Pillar 1 framework holistically, covering credit risk, market risk and operational risk. Taking credit risk as an example (given that it is arguably most material risk for banks), the Standardised Approach is already capable of capturing climate risks to some extent through external credit ratings, which increasingly take into account climate factors. However, there is no standardised methodology that credit ratings use to capture climate risks, meaning that outcomes vary depending on the rating used (and credit ratings agencies also face many of the same data and methodological challenges that firms experience).

The IRB framework could also conceptually incorporate climate risks through the valuation (and re-valuation) of collateral, and consequently LGD. However, banks’ assessments of the impact of climate factors are likely to be variable, and in many cases banks do not yet have the required data. PD could in principle incorporate climate-related threats to the solvency of borrowers, but it is currently calibrated based on historical losses and over a one-year time horizon. This means that, to capture climate risks in full (for which historical losses are unlikely to be a good indicator of future losses), it would need fundamental revision, which would in turn require significant lead times to bring to fruition.

A significant redesign of the Pillar 1 framework would also face political obstacles. Our view is that the PRA in particular is in no rush to front-run action by the BCBS on Pillar 1, and a revised framework led by the BCBS would take years to agree and implement. There may be more appetite in the EU to act ahead of global agreement being reached (as was the case with Pillar 3 requirements for ESG risks) - although, given the June 2023 timing of the EBA’s final report, it may be challenging for the Commission to include any proposed revisions in the finalised CRD6/CRR3 package without causing further delays to that package’s implementation.

More fundamentally, basing Pillar 1 on less reliable forward-looking data could increase undue variability in banks’ capital requirements – likely to be an unpalatable outcome for the BCBS (and the EBA) given efforts over the last decade to achieve the opposite outcome.

Political challenges also arise from current concerns around energy security driven by Russia’s invasion of Ukraine. This may temper policymaker enthusiasm for comparatively simple policy options that penalise banks’ exposures to fossil fuel companies. This includes, for example, assigning a risk weight of 150% to existing fossil fuel exposures and a 1250% risk weight to exposures to new fossil fuel exploration activities – an option proposed by a number of MEPs as amendments to the EU’s CRD6/CRR3 package.
 

Supervisors will focus on the “foundational tasks” around climate risk management, making use of the Pillar 2 framework

Our view is that the technical and political challenges set out above all but rule out wholesale revisions to Pillar 1 in the near term (while not necessarily precluding action longer-term). However, it is clear to us that supervisors recognise that they cannot afford to “wait and see”. A conclusion that we drew from the BoE’s October conference on climate change and capital was that supervisors will focus on the following interlinked “foundational tasks”:

  • Using the Pillar 2 framework to drive improvement in banks’ climate risk management capabilities

The use of Pillar 2 to address climate risks is comparatively well established in the supervisory framework. Banks in the UK and EU are already required to include climate risks in their ICAAP to the extent that they are material, while the ECB has indicated repeatedly that the SREP will incorporate assessments of banks’ climate risk management capabilities, including deficiencies identified through recent supervisory climate stress tests. Through influence on banks’ SREP scores, climate risk could have an impact on banks’ eventual Pillar 2 Requirements. The PRA could choose to use Pillar 2 Risk Management and Governance (RMG) scalars in a similar way, penalising firms whose capabilities are lagging significantly behind its expectations.

  • Pushing firms to gather the right data

All of the policy options being considered by policymakers rely on banks having a better understanding of their counterparties’ exposure to physical and transition risks than they have today. Banks should work to address data gaps, such as counterparty level emissions data, corporate sector classification, buildings’ energy performance, or information on the physical location of counterparties’ assets, which will need to be filled in all implementation scenarios. Engaging with counterparties on transition plans will also help banks come to a view on the short-, medium- and long-term transition risk profile of those counterparties.

  • Driving improvements in firms’ scenario analysis and modelling capabilities, and shifting from climate “scenario analysis” to “stress testing”

The BoE’s Climate Biennial Exploratory Scenario (CBES) and the ECB’s Climate Stress Test both demonstrated that firms have a long way to go before they can model climate risks in a forward looking manner – a point reiterated by the PRA in a Dear CEO letter on banks’ climate risk management practices. Further development of firms’ climate risk modelling capabilities will facilitate a greater industry understanding of climate risks and where risks to firms’ capital might arise.

Although supervisory climate stress testing exercises in the UK and EU have, up to now, not been used to set Pillar 2 capital add-ons, we expect that supervisors will look to do so in the future – not least given that the BCBS principles on climate risk management and supervision instructed supervisors to use climate scenario analyses to test the resilience of firms’ financial positions. This would most likely take the form of integration of climate risks into existing solvency stress testing scenarios, rather than standalone exercises. We expect that this could be one of the BoE’s near-term climate-related priorities – although it may take some time to develop an appropriate methodology to capture climate risks at the shorter (three-five year) horizon that solvency stress tests tend to cover. The EBA has also said in its 2023 work programme that it is working on incorporating climate risk into its stress testing framework.
 

Progress in these three areas has both short- and longer-term benefits

Firstly, as covered above, robust scenario analysis and stress testing will provide supervisors with the comfort they need that banks are holding sufficient capital against climate risks that are likely to crystallise in the short to medium term – through the ICAAP and potentially through supervisory stress testing exercises.

Secondly, progress made now on data and modelling capabilities would make the eventual introduction of a Pillar 1 approach more realistically achievable, as banks that have made progress on climate modelling will have relatively less work to do to put the right capabilities in place.

For banks, the message is “keep doing your homework”

Banks should already know what supervisors expect them to do to improve, following thematic reviews and feedback from both the ECB and PRA. The ongoing policy debate around Pillar 1 capital is, to some extent, a red herring. Banks should ensure that making further progress on data and modelling is their top priority. Supervisors’ indications that they are prepared to use Pillar 2 add-ons to speed up progress on risk management should provide sufficient incentive.


Financial stability authorities could use macroprudential tools to address the systemic elements of climate risks, but there is no consensus yet on the right tool to use

Another relevant conclusion from the BoE conference was that the systemic dimensions of climate risks may require a macroprudential response. This view is shared by the EBAECB and the European Systemic Risk Board, as well as the Financial Stability Board. European authorities have highlighted that the use of tools such as exposure limits or the Systemic Risk Buffer merit further exploration. The choice of macroprudential tool will ultimately depend on the nature of the risks that authorities are looking to address. Our expectation is that the PRA and Financial Policy Committee will focus on concentrations and changes in the accessibility of credit arising as a result of “herd behaviour”, as financial services firms execute their transition plans in tandem – a risk also identified by the CBES. This may require intervention to either constrain or incentivise lending to certain sectors, although it will require more thought to establish whether any tools are available to compel banks to lend to sectors that are at risk of becoming stranded.


Conclusion

In our view, policymakers are unlikely to pursue or implement a Pillar 1 approach in the short term. Instead, supervisors will prioritise using the tools that are already available – Pillar 2, supervisory engagement to improve risk management capabilities, scenario analysis and the macroprudential toolkit. Using a combination of available capital and non-capital tools outside the Pillar 1 framework will buy policymakers time to pick through the complex technical and political challenges associated with revising the globally agreed Pillar 1 framework.

Our thinking