Understanding the financial stability implications of climate risks: approaches to climate risk analysis in Financial Sector Assessment Programs (FSAPs)

• The analysis of climate-related financial risks has distinct characteristics that introduce new data and modeling challenges. These include longer time horizons; limited disclosure of climate-related financial data; assessment of climate-related financial risks extending beyond the economic and financial factors typical of conventional risk analysis; little to no guidance from historical trends (given that climate risk analysis is inherently forwardlooking); greater sectoral and geographical granularity; and a higher level of uncertainty and model complexity due to interactions between climate, anthropogenic activities, and economic dynamics.
• The modeling framework of the IMF’s climate risk analysis for both transition and physical risks has three stages: (1) climate risk diagnostics with a global and country-specific perspective, (2) design of country- and financial system-specific climate scenarios, and (3) financial stability.
• The macro approach to assessing the implications of climate-related risks for financial stability aims to quantify the impact at an aggregated economic/financial level and uses climate-augmented macrofinancial scenarios as inputs for standard stress-testing methodologies to assess the financial system’s resilience. The micro approach is an extension of the macro approach and uses granular income and balance sheet data of a large sample of individual firms and/or households (when reliable granular data is available) to assess the impact. The integrated micro-macro-modeling framework has been piloted in several FSAPs.
• Physical risks arise as the interaction of three components: hazard, exposure, and vulnerability. The macro approach incorporates the analysis of the impact of aggregate shocks, due to hazard damages, on macroeconomic and financial variables by inputting country-level aggregates of granular damages estimated by catastrophe models into macro-models. The micro approach requires granular transaction and loan-level data and thus has limited application due to data constraints.

The models can be improved by enhancing temporal resolution and refining short-term scenarios to better capture near-term shocks; enhancing sectoral and spatial granularity to better represent heterogenous climate-related risks across sectors and regions; combining transition and physical risk considerations in an integra