This latest ECB 2022 exercise aims to enhance the capacity of both banks and supervisors to assess climate risk. It will enable the regulator to undertake a peer-to-peer comparison and identify best practices, as well as challenges banks face regarding climate risk. It will provide banks with a comprehensive analytical toolkit which they could use for their future 2023 ICAAP submission.
In this blog we compare this upcoming ECB climate stress test to the Bank of England’s Climate Biennial Exploratory Scenario (CBES), focusing on the similarities and differences between the two exercises, as well as identifying the key challenges banks will face from this ECB stress test.
Similarities and differences between the CBES and ECB climate change stress test
- This ECB stress test is the latest iteration in regulatory climate change stress testing and its primary objective is to enhance the capacity of banks and regulators in assessing climate risk.
- It will provide banks with a comprehensive analytical framework which they can use for their own internal stress tests and for their upcoming ICAAP submission.
- It will enable the regulator to undertake a peer-to-peer comparison of climate risk metrics identify best practices, as well challenges banks face. This may indirectly impact, via the SREP scores, the Pillar 2 requirements.
- The overarching aim of the CBES is like the ECB exercise, as both are exploratory stress tests. However, the key focus of the Bank of England’s stress test was to assess the resilience of the UK financial system to physical and transition risk and explore vulnerabilities in participants’ current business models to future climate policy pathways.
Scenarios & Timelines
- Both supervisory stress test exercises incorporate three identical long-term scenarios which are sourced from the Network for Greening the Financial System (NGFS): 1) An Orderly Transition scenario towards achieving carbon emission targets by the end of the forecast horizon, which has a muted impact on GDP; 2) A Late Action scenario which, as per the previous scenario, assumes carbon emission targets are achieved at the end of the forecast horizon. However, policies to achieve these are not implemented until halfway through the scenario, which leads to a disorderly transition and adverse impacts on GDP; and 3) A No Additional Action/Hot House scenario assumes CO2 emissions are not reduced and leads to adverse weather events, repercussions on both living and working conditions, as well as uncertain macroeconomic conditions.
- Whist the forecast horizon for both these supervisory exercises are identical (30 years), their starting points differ. All ECB scenarios commence in 2022. However, through the calibration of the scenario and in the absence of any policy actions, the physical risks in the No Additional Actions scenario are prevalent between 2050 – 2080.
- Asides from the three long terms scenarios, the ECB exercise includes a short term, three-year transition scenario, which is triggered by a sharp increase in carbon prices. This disorderly scenario is front loaded rather than spread across the forecast horizon and is the basis on which banks are to report the impact of climate change (via the ECB template) on credit and market risk metrics.
- The ECB stress test uses a hybrid approach:
- The short-term scenario assumes a static balance sheet.
- The long-term scenarios assume a dynamic balance sheet and banks are required, via the ECB qualitative questionnaire and other supporting documentation, to outline their strategies.
- For the CBES, like the ECB short term scenario, banks are to assume a fixed balance sheet. This provides transparency to the regulators over the scale of the adjustments required by banks in achieving long term climate objectives. However, as part of the qualitative questionnaire, banks can deviate from this fixed balance sheet assumption and outline the actions and business opportunities they would take for each scenario.
- The ECB stress test captures the latest iteration of the NGFS scenarios, which incorporates the National Institute for Economic and Social Research (NIESR) quarterly econometric model, to capture the connectivity between transitional and physical impacts on macroeconomic variables at the granular country level.
- Furthermore, this latest version of these NGFS scenarios also incorporates damage estimates as a result of potential physical risk implications. Based on the scenario, these shocks have been calibrated on a country-by-country basis and are reflected in the macroeconomic variables.
- For the CBES stress test, as climate change stress testing was at its infancy, there was considerable uncertainty in terms of modelling the GDP impacts from both transition and physical risk. Consequently, these impacts were modelled separately, and participating banks were provided with a set of harmonised transition pathways, chronic climate impacts and indicative economic impacts for each of the scenarios.
Output and deliverables
- This ECB exercise comprises three separate modules. Modules 1 and 2 are mandatory for all significant institutions, whereas module 3 is only required for a subset of banks.
- This stress test is rather prescriptive in terms of the data granularity required for this exercise. For module 2 (climate risk metrics) and module 3 (bottom-up stress test projections), banks are required to:
- Split their corporate exposures between 22 industries at the NACE two-digit level and at the NUTS 3 1 regional classification.
- In module 2, banks will be required to provide specific greenhouse gas emission metrics for their top 15 counterparties for certain industries.
- In module 3, banks will be required to split their corporate exposures into 1) Not secured by real estate, 2) Secured by real estate where the collateral is within the scope of the Energy performance certificate (EPC), and 3) Secured by real estate where the collateral is not within the scope of the EPC.
- In terms of output deliverables, for the ECB stress test, banks are to provide risk metrics for the short-term scenario on an annual basis and on a ten-year cycle for each of the long-term scenarios.
- For the CBES stress test, banks are required to provide wholesale and retail projections by the standard industrial classification level 1 sectors and level 2 sectors where relevant. Furthermore, these projections are to be disaggregated at country level depending on materiality, or alternatively amalgamated to regional level.
- Additional separate template requirements for the CBES exercise are:
- Banks to splice their actual and projected exposures for UK residential and mortgage properties by EPC.
- Banks are required to report their projected risk and climate metrics for their top 100 non-financial non sovereign counterparties.
- In terms of output deliverables for the CBES exercise, banks are required to provide risk metrics on a five-year basis for each of the long-term scenarios.
- Both the CBES and ECB stress tests incorporate a qualitative questionnaire.
- The ECB’s questionnaire focuses on banks’ internal risk frameworks and assumptions around the bottom-up exercise. Banks are limited in that they can only select the predefined responses.
- The CBES questionnaire is more holistic and focuses on the narrative around the results, methodology, proposed management actions, as well as qualitative views on existing and future climate risk management practices. Banks are limited to a word count in their response and are able to provide supporting documentation.
Challenges banks face
- For banks participating in modules 2 and 3, this ECB stress test poses several challenges for them, specifically:
- Accurately identifying greenhouse emissions for their corporate exposures.
- Classifying current and projected exposures and risk metrics spliced by geography, sector, and EPC.
- Developing justifiable dynamic balance sheet assumptions to conform to the economic scenarios.
Although global regulators continue to assess banks’ resilience to climate risk through stress tests, they are also becoming much more attentive to how banks are capturing these risks on a regular basis.
Through the completion of these stress tests, banks have continued to enhance their analytical framework. However, alongside these supervisory exercises, and as part of their wider risk framework, banks also need to demonstrate the policies, data architecture and governance they are implementing to mitigate climate risks to the regulators.
We have devised a tool, based upon the UNEP FI framework, which helps bridge the gap between banks’ existing modelling capabilities and regulatory deliverables. This flexible and transparent toolkit can be calibrated to each respective banks’ data to generate the key risk and climate metrics and can be adapted for both regulatory and internal stress tests. For more information reach out to an expert member of our team.
1NUTS Maps - NUTS - Nomenclature of territorial units for statistics - Eurostat (europa.eu)