Practical issues in climate change modelling for lenders

The focus on climate change from regulators around the globe continues to intensify. The Basel Committee on Banking Supervision released a report last month (April 2021) on the challenges that banks face in measuring the financial risks associated with climate change.  Below we share our thoughts on the key modelling challenges highlighted in the report.

Identifying Risk. Credit risk modelling of the financial impact of climate change has focused on physical and transition risks. In the UK, the most material physical risks identified by many lenders include flooding, subsidence and coastal erosion. For retail mortgage lenders, energy efficiency represents the most immediate transition risk, however regional and sector-level economic changes (e.g. phasing out of coal and oil production) will also impact the risk of the portfolio.

Lenders are approaching these challenges from two different directions – top-down approaches start with sector, regional and industry average impacts, which are applied downwards onto individual mortgage accounts. Bottom-up approaches look at the risks affecting individual borrowers and properties and aggregate upwards. While top-down approaches are easier to implement, they frequently overlook the risks affecting individual properties and the interactions between risk factors. Top-down approaches also risk overlooking concentrations to physical or transition risks in the portfolio. Bottom-up approaches require highly granular data and far more complex models, however they provide a deeper understanding of risk and far better information for managing risk appetite. Lenders opting for the bottom-up approach are increasingly turning to geolocational data to identify and quantify such potential hazards as flooding, subsidence, and coastal erosion that may affect the properties in their portfolio. As in the early days of IRB and IFRS9, firms may find themselves balancing the advantages of increased granularity against the cost and complexity of the models required to take advantage of this detail.

Risk in a Changing Climate. Measuring climate-related financial risks requires approaches to extrapolate these risks into the future under different climate emissions scenarios. Until recently most climate-related financial risks measured by banks have focused on short-term transition risks to counterparty and portfolio-level exposures e.g. estimating the likelihood of more stringent regulation, exposure to sectors, etc. Climate impacts are expected to take place over a longer time frame than is used in traditional macroeconomic exercises and assessment horizons of up to 50-60 years are now being considered.

While models of how physical risks will evolve with climate change have existed within the insurance industry for years, the financial industry has only recently started investigating how to incorporate these forecasts into credit risk models and link this data to traditional risk parameters such as probability of default and loss estimation.

Managing Uncertainty.  A particular challenge of modelling climate change impacts is the high degree of uncertainty involved. Unlike traditional credit risk modelling, which relies on understanding and projecting historical trends, there are limits on the both the availability and relevance of historical data relating climate to financial risks.

Climate models, for example, rely heavily on assumptions about changes in climate, advances in technology, and the behaviour of both political and economic actors. Models have to predict across long time horizons and account for the potential impact of climate tipping points and feedback loops.  The direction that climate change will take is a further unknown, with multiple scenarios and impacts ranging from mild to extreme. In addition, these impacts, while often considered in isolation, will likely lead to linked events e.g. increasing physical risks driving more aggressive energy efficiency legislation, leading to larger transition risks. In response, banks will need the capability to implement forward-looking models that can be applied to multiple climate and regulatory scenarios.

Climate change risk is introducing a step change to the requirements for risk management in the financial industry, perhaps commensurate with the introduction of IRB. Assessing climate risks requires new approaches, using types of data not used before in the credit risk industry, with less reliance on historical information and with greater levels of uncertainty. Despite these challenges, robust solutions are already emerging in the UK financial industry, which are embedding climate change solutions into existing risk management frameworks and models.

For more information on how 4most can assist you in developing climate change risk measurement and scenario modelling solutions, please get in touch with thomas.clarke@4-most.co.uk or ivelina.nilsson@4-most.co.uk.