Measuring climate-related financial risks using scenario analysis

June 2024

Quantifying the risks of climate change is an on-going challenge for financial services (FS) firms and a consistent and expanding area of focus for regulators.

The BCBS noted at the end of 2023 that banks were lagging on the quantification of climate-related risks — for more see KPMG in the UK's article here. The PRA and ECB are already supervising firms against their expectations on climate modelling — the PRA is due to refresh Supervisory Statement 3/19 in 2024 and the EBA has recently closed its consultation on new Draft Guidelines for the management of ESG risks, including minimum standards and a methodology for identifying, managing and monitoring ESG risks. The International Association of Insurance Supervisors (IAIS) launched a consultation in 2023 on climate risk scenario analysis and is expected to publish final comments at the end of the year. 

As part of its April 2024 Quarterly Bulletin, the Bank of England (BoE) published an article that explores how central banks and financial institutions can use scenario analysis to quantify climate-related risks. The article focuses on `extending' macro-climate scenarios to carry out more granular asset-level analysis and references examples across sovereign bonds, corporate bonds and residential mortgages.

Banks and insurers can benefit from reviewing the BoE bulletin and assessing how their existing approach compares with the central bank’s guidance. Modelling at asset level is particularly important given the supervisory scrutiny highlighted above. Alignment with the BoE’s guidance will likely go a long way in helping to achieve this. 

Standard risk modelling approaches do not capture the unprecedented nature and uncertainty of climate-related risks, hence the need for further guidance from the BoE. 

The article focuses on the FS-specific NGFS climate scenarios and how to measure the asset-level impacts of climate-related financial risks. Firms can use toolkits such as:

• Special granularity and resolution — e.g. proximity of the asset to the source of physical risk;
• Related variables — e.g. including variables that are not usually provided in climate scenarios, such as sovereign debt / GDP ratios;
• Temporal misalignments — e.g. using varying time horizons in analysis given the broad range of maturities, from months to years, for financial assets; and
• Intra-sectoral variabilities — e.g. developing toolkits to apportion sectoral impacts at asset level.

1. Sovereign bonds

Macro climate scenarios often only provide a short-term and a 10-year interest rate projection. However, as investment portfolios include sovereign bonds with intermediate or much longer maturities, it is important to extrapolate the macro projection to measure financial risks across the whole yield curve. Important factors to consider when doing this are: 

• Changes in the `risk-free' component of interest rates — firms should estimate how yields change on bonds with maturities other than short-term and 10 years. The BoE provides an example assuming that the yield of a 30-year (or other maturity) bond today is simply determined by the pathway of short-term interest rates in a given scenario over the next 30 years (or other maturity); and
• Changes in the `credit risk' component of interest rates — firms wishing to measure how debt/GDP ratios or sovereign credit ratings change over climate scenarios primarily need to overcome the `related variables' challenge i.e. what are the other considerations that are not usually provided in climate scenarios. Factors to consider in developing debt / GDP projections include changes in tax revenues, carbon tax revenues, fuel taxation, green investment, debt interest, and acute physical spending (i.e. government expenditure on supporting households to respond to or mitigate against the physical risks of climate change).

2. Corporate bonds

Physical and transition risk transmission variables for corporate bonds include regional carbon prices, country-level GDP and global temperature rises. 

Macro scenarios do not provide sufficient granularity to assess impacts at individual firm level. The BoE notes that asset-level analysis should ideally incorporate: 

• The impact of carbon pricing on the specific issuer, with consideration of the corporate's future decarbonisation plans;
• The impact of other transition risks and opportunities on the specific issuer, e.g. regulatory costs, reductions in income, any financial benefits;
• The impact of physical risks on the specific issuer, e.g. location of operations, supply chains and consumer markets; and
• Economy-wide interdependencies.

3. Residential mortgages

Residential mortgages form the largest asset class for most retail banks but, with typical maturities of 25 years or longer, climate-related risks may not crystallise for several decades. When extending macro scenarios to asset-level analysis the BoE advises firms to consider the following key challenges:

• Transition risks: not knowing how energy consumption is distributed across different households and how different households might respond to rising energy prices by improving energy efficiency — combining energy prices from macro scenarios with the energy consumption data from Energy Performance Certificate (EPC) ratings can provide an estimate of each borrower's total energy cost.
• Physical risks: not knowing how climate change impacts the frequency and severity of floods in the UK. The BoE analysis uses data from a catastrophe modelling firm to estimate this — and thus indicates the potential damage to specific properties.

Firms will need to build out their scenario analysis capabilities to reflect the approaches set out by the BoE in its report.

KPMG in the UK has a dedicated Risk and Regulatory Advisory practice with extensive climate risk expertise, including climate-related modelling. We can support financial institutions in addressing the business and regulatory challenges of climate-related risk, applying our experience, our understanding of industry bodies and evolving regulatory expectations, and our combined expertise across different markets and risk types. 

Our services include:

Modelling and analytics: including benchmarking and gap analysis of existing capabilities, incorporating climate risk into stress testing and credit decisioning models, and climate-adjusted asset valuation modelling. 

Scenario analysis: including climate risk driver identification and materiality assessment, tailored climate scenarios and climate risk quantification training. 

Model validation: including validation of climate risk and ESG models, model enhancement identification and planning, and internal audit support.

To discuss your requirements, please get in touch.