Banks’ business planning and credit assessments have changed significantly since the 2008 financial crisis. Firms must now use a “forward-looking” approach for loan loss reserving and, for those banks using IFRS 9, this should be implemented under a range of forward-looking scenarios.
Banks also routinely generate forward-looking scenarios as part of their medium to long-term business planning and stress testing obligations.
With the financial world increasingly facing climate change risk and responding with scenario analysis we ask: How similar is climate risk scenario analysis to IFRS 9 scenario analysis? Also, is the experience gained with IFRS 9 relevant to banks, given the current focus on climate and with climate impacting credit risk? We explore several aspects of this below:
- Forward-looking projections: Both climate risk and credit scenario analysis rely on forward-looking scenarios. IFRS 9-ready banks are adept at generating economic scenarios and estimating financial consequences. What is new for climate scenario expansion is the requirement to generate detailed macroeconomic and technological factors that drive the financial risks at credit facility level on the back of complex but high-level environmental and social narratives. Examples might be a substantial rise in the average global temperature above pre-industrial levels and a breakdown in global trade with regional rivalries prevailing in food and energy supply.
- Financial risk drivers and data: Firms will need to re-think portfolio segmentation and vulnerabilities from the traditional credit perspective. The emphasis for credit analysis is normally on a borrower’s financial health and broad economic conditions, assuming long-term energy, sector and environmental conditions are relatively stable.2 Climate data requirements are steeper and firms will need to include additional data on physical risk hazards (including, say, the impact of local air pollution and the energy profile of properties) and transition risk hazards (such as the impact of rapid technological change as the economy is forced to transition to low carbon solutions, disruption to supply chains with implied threats to their and others’ business models).
- Non-linearity of outcomes: IFRS 9 and climate risk scenario analysis will both encounter “non-linearity” – the effect where the impact of a scenario significantly increases once certain thresholds are crossed and leads to disproportionately adverse outcomes. For climate risks, this might happen when evaluating physical hazards, such as high greenhouse emissions. Once emissions cross certain thresholds the effects on average global temperature and ocean acidification could jump significantly. Similarly, for credit risk, a fall in asset prices (such as house prices) below certain percentage drops can result in a steep increase in default risk.
- The past as a guide to the future: Credit scenario analysis tends to rely on historical relationships between economic and credit factors (although this is being challenged in the current environment). Climate risk scenario analysis differs by giving significant weight to future changes (physical and transitional); deviation from the past will need to be recognised and factored into every scenario.
- Multiple scenarios: This is a common expectation of both IFRS 9 and climate risk scenario analysis. For IFRS 9, firms typically have a highly weighted “base case” scenario and then, usually, one “upside” and one or two “downside” scenarios. For climate risk work, no firm expectations have been agreed on the minimum number of scenarios. However, given the uncertainty around climate outcomes, there is expectation that firms will develop more than one scenario and consider a range of transition and physical risk hazards in each.The guidance from regulators is to make reference to both the temperature target and the transition pathway taken to get there in each scenario.3 This makes climate risk scenario analysis more involved than a collection of exogenous macroeconomic stress shocks.
- Horizon: Credit scenario analysis is typically reliable over five to six years. Climate risk analysis is necessarily much longer-term with modelling expected over 50 years and significant events (such as major technological change or geographical changes) needing to be included as disruptors, even several years out.
- Modelling: Credit risk scenario modelling is mature having been given a significant focus by regulators after the 2008 financial crisis (although IFRS 9 has thrown in some new complexity in the last couple of years). Climate risk scenario modelling for banks is in its early stages and is significantly less developed than credit modelling.
We conclude that, although there is some common ground between IFRS 9 and climate risk analysis, particularly in the use of multiple forward-looking scenarios, there are also marked differences.
The planning experience gained by banks will be beneficial, but there are elements of climate risk scenario analysis that are new, complex and will need significant investment in data, infrastructure, modelling and reporting to develop a mature strategy and implementation framework for climate risk management.
We expect climate risk scenario analysis to evolve rapidly as banks (and other financial services firms) are pushed by investors and regulators for richer climate assessments and disclosure.