Both the Bank of England (BoE) and the Basel Committee on Banking Supervision (BCBS) published papers in April 2024, discussing the use of scenario analysis in measuring firms’ climate-related financial risks.

Overview

Financial institutions are increasingly looking to scenario analysis to assess the financial risks posed by climate change. While some firms have developed in-house scenarios, others utilize those produced by Network for Greening the Financial System (NGFS), which are widely used across the industry.

These papers follow growing criticism of current scenario analysis. Many of the issues and limitations raised echo those highlighted in publications such as the Institute and Faculty of Actuaries' (IFoA) Emperor’s New Climate Scenarios. It is becoming gradually clearer that many scenarios do not adequately capture the downside risks of climate change. Therefore these scenarios cannot be used effectively to control the risks of climate change or optimise portfolios and investment decisions to support positive change.

Article from the Bank of England

On 17 April 2024, the BoE released a paper exploring the use of scenario analysis for measuring firms’ climate-related financial risks. The context was the increasing focus on climate-related factors by financial institutions and the BoE considered the application of scenario analysis to the financial risks to which it is exposed. We believe the conclusions, if not the specific examples, are relevant to all financial institutions considering these risks.

Our key takeaways from this bulletin are that the BoE believes that current mainstream scenario analysis:

1. provides a useful starting point for assessing climate-related financial risk exposures but they do not generally provide the level of detail firms require
2. only captures a subset of chronic and acute physical risks, rather than all relevant climate risks faced by a firm and that climate risks may be underestimated as these scenarios don’t allow for tipping-points or the possible non-linearity of risks
3. doesn’t account for the heterogeneity of assets within a portfolio whereas risks and opportunities will differ between companies within the same sector or sub-sector
4. doesn’t include projections of all variables needed for granular asset-level analysis, so firms will need to construct or acquire additional variables.

The paper focuses on the asset classes the BoE is most exposed to, ie sovereign bonds, corporate bonds, and residential mortgages, and includes examples of how they have approached extending the NGFS macro-scenarios to undertake asset-level analysis.

Sovereign Bonds

The NGFS scenarios only provide projections of short-term rates and a 10-year rate. The BoE has sought to build a yield curve through interpolation and extrapolation of these rates. They have also highlighted the need to extend scenarios to include the data needed to evaluate changes in credit rating of the sovereign over time, such as gross domestic product (GDP) and government debt factors. With these extensions to include various maturities and forecast credit rating changes, and along with assumptions about the impact of sovereign credit rating changes on their yield, the BoE were able to calculate the changes in sovereign yields along the scenarios. Results showed a portfolio of 20-year sovereign bonds losing:

• over 9% of value in all scenarios
• over 20% in the Divergent Net Zero scenario, ie a scenario reflecting divergent policies across sectors and regions represented as carbon price variation, leading to a disorderly transition to net-zero.

Looking at implied changes in credit ratings, the largest downgrades are seen under scenarios where current policies persist, leading to a higher degree of warming and greater physical risks. Under these scenarios, some sovereign credit worthiness would be downgraded by a couple of whole letter ratings.

Corporate Bonds

The BoE noted that key challenges in applying climate scenario analysis to bond-level assessments included the level of granularity needed and capturing intra-sectoral variability. It is noted that this data is often developed by third party providers rather than in-house due to the data intensive approach required.

The BoE suggests that scenario analysis for corporate bonds should incorporate:

• the impact of carbon pricing on the issuer
• the impact of other transition risks and opportunities on the issuer
• the impact of physical risks, including to the issuer’s office location and supply chain
• the economy-wide interdependencies between different corporates, recognising that the scale of costs and benefits will be a function of impacts across the issuer’s supply chain.

Having calculated these factors' impact on relevant financial metrics, the BoE expects financial institutions will be able to integrate them into existing financial modelling toolkits, albeit evaluating credit changes may require further assumptions. The biggest impact was seen on the electricity, industrial and transport, and energy sectors, while the BoE found the most striking result to be the significant variability between firms within the same sector, demonstrating the critical need to incorporate firm-specific considerations.

Residential Mortgages

Similar challenges were noted for residential mortgages, particularly the need to extend macro scenarios to capture impacts at a sufficiently granular level. For example, incorporating Energy Performance Certificate (EPC) ratings is required to assess the impact of changes in energy costs on specific households, and possible costs to improve the energy efficiency of properties. The estimated impacts vary by EPC rating, from a 14% increase in the proportion of household income used to meet mortgage payments for F-rated properties to 4% for B-rated properties.

Similarly, incorporating flood data can improve assessments on changes in the cost and availability of insurance for specific properties following the run-off of FloodRe. FloodRe ensures flood insurance is widely available and that premiums are capped for many properties in flood prone areas. However, this guarantee will end in 2039. The BoE’s scenario analysis estimates changes in insurance availability and cost could lead to a fall in house prices for the 10% of areas that are most vulnerable to flood risk that is 6.5 times larger than that for an area with the median level of flood risk.

Discussion paper from the Basel Committee on Banking Supervision

In the same week as the BoE published their paper, the BCBS issued a discussion paper seeking feedback on the use of climate scenario analysis to strenghten the monitoring and management of climate-related financial risks. The BCBS’s focus is on banks and their supervisors, however there is a clear read-across to the use of climate-related financial analysis by insurers in both the key features and the usage-specific considerations that they are discussing. Similar to the BoE, the BCBS recognises there are challenges with the application of climate scenario analysis and have requested feedback in a number of areas. However, the paper also notes the objectives and benefits of effective use of climate scenarios, such as helping firms identify risk exposures, informing risk management processes, and assessing the resilience of their business model to climate risks. The closing date to provide feedback on the discussion paper is 15 July 2024.

What are we doing at Hymans Robertson?

Hymans Robertson has been carrying out climate scenario analysis for our clients for a number of years. Our in-house climate scenario model assesses investment portfolios under three climate scenarios and has been used by a range of clients to assess their climate strategies, and for inclusion in their Task Force on Climate-Related Financial Disclosure (TCFD)-reports. We’ve long been vocal about the importance of interpreting climate scenario modelling with care, recognising that it’s an area where both understanding and methodologies will develop through time. We recognise the need to evolve.

In 2023, we developed our approach to climate scenario modelling to include more detailed narrative scenarios – ‘stories’ that help explore the complex interactions that traditional modelling cannot capture. With the help of these narratives, we are helping to educate our clients on limitations in current modelling approaches and to develop their own scenario analysis approach. We published details of our approach in September 2023, inviting feedback from the industry.

We see many benefits to firms in developing narrative-based scenarios, including gaining insights into the broad drivers of risk, the behaviour and interconnectedness of risks, and more granular risk management.

Conclusion

It’s clear that climate scenarios are firmly on the regulatory agenda, and we can expect further guidance on their use over coming months. Firms have an opportunity to get ahead of regulations by starting to:

• ensure that those that rely on the output of such scenario analysis are fully aware of the limitations inherent in the scenarios
• explore what is needed to expand their scenarios to overcome some of these limitations
• consider developing narrative-based pathways.

While we recognise the development of asset-level climate scenarios is challenging and resource-intensive, we believe the benefits far outweigh the challenges.

If you would like to discuss climate scenario modelling and/or our narrative-led approach, please get in touch with your usual Hymans Robertson contact or the authors of this blog.

This blog is based upon our understanding of events as at the date of publication. It is a general summary of topical matters and should not be regarded as financial advice. It should not be considered a substitute for professional advice on specific circumstances and objectives. Where this blog refers to legal matters please note that Hymans Robertson LLP is not qualified to provide legal opinion and therefore you may wish to obtain independent legal advice to consider any relevant law and/or regulation. Please read our Terms of Use - Hymans Robertson.