Despite the growing use of stress tests by central banks, it remains to be seen whether banks are fully aligning with these efforts and adequately preparing for the risks posed by climate change.
Investors and financial institutions face increasing pressure to improve their climate-related risk management and reporting practices to better prepare for potential impacts such as temperature increases and extreme weather events.
We explain how banks can improve climate stress test results to meet the growing demand of regulators.
Back to basics
With many central banks and regulatory authorities across the globe increasing the requirement for financial institutions to conduct climate-risk stress tests, some banks are better positioned than others. However, banks can only claim to be fully prepared at this point, attempting to meet the different approaches of central banks. Climate stress tests contrast with economic shocks modelled in traditional stress tests as they include a range of climate shock scenarios to be analysed on financial actors. As one of the measures driven by the Basel Accords, Pillar 2 of the Basel Framework reinforces the first pillar that sets minimum capital requirements to apply additional capital buffers to withstand stressed situations. The policy has been driven towards encouraging financial institutions to examine and disclose their climate risk exposures, primarily through the Task Force on Climate-Related Financial Disclosures (TCFD) but also more within scenario analysis and stress testing frameworks.
However, climate stress tests remain at an early-stage state as learning exercises for banks and supervisors. The results of each stress test do not have a direct implication for banks' capital requirements. Therefore, it is crucial to identify the distinction between climate stress tests and climate scenario analyses. Stress tests examine an institution's ability to withstand extreme conditions to assess the number of capital banks must hold. At the same time, scenario analyses focus on determining how financial institutions set themselves for future scenarios. Climate stress tests are based on scenario analysis in a set framework developed by the Network for Greening the Financial System (NGFS), shown in figure 1 below, including a scenario with high physical risk, as climate change remains largely unabated.
Who is who?
Differences between the central bank regulators include framework differentiations of bottom-up approaches, balance sheet assumptions, financial risk types covered, and types of risk, as the BoC has focused on testing for transition risks. In contrast, the European Central Bank (ECB), BoE, Autorité de Contrôle Prudentiel et de Resolution (ACPR), HKMA and Australian Prudential Regulation Authority (APRA) focused on both transition and physical risks. These differences may hamper investors’ ability to compare the full suite of climate-related risks faced by all institutions have been individually showcased below:
1. The European Central Bank (ECB)
The ECB climate stress testing learning exercises aim to improve on past initiatives by introducing innovations in terms of data and modelling, being the only climate stress test to capture the interactions of both long-term physical and transition risks with a top-down approach.
The test showcased that the banks progressed in climate stress-testing capabilities but also exposed deficiencies, including data gaps and inconsistencies. More prominent institutes are exposed to acute physical risks to varying degrees. Banks' risks in this regard are closely linked to the geographical location of their lending activities. Around 60% of banks need well-integrated climate risk stress-testing frameworks. Most banks envisage a medium to long-term time frame for incorporating physical and/or transition climate risk into their framework.
2. The Bank of England (BoE)
The Bank of England launched its Climate Biennial Exploratory Scenario (CBES) in June 2021 to capture the exposure of UK banks and insurers to both the transition and physical risks over the next 30 years based on the NGFS framework. The exercise considered two possible routes to net-zero UK greenhouse gas emissions by 2050: an 'Early Action' (EA) scenario and a 'Late Action' (LA) scenario. A third 'No Additional Action' (NAA) scenario explores the physical risks that would begin to materialise if governments worldwide fail to enact policy responses to global warming. It was identified that climate risks captured in the CBES scenario have the potential to drag UK banks' and insurers' profitability, and climate losses are uncertain with scenario analysis in its infancy and substantial data gaps. All participating entities have more work to improve their climate risk management capabilities.
3. The Autorité de Contrôle Prudentiel et de Résolution (ACPR) and Banque de France (BdF)
With a bottom-up approach, the 2020 climate pilot exercise with a group of French banking institutions and insurers assessing both physical and transition risks on credit risk, market risk and sovereign risk for nine banks and on the assets and liabilities of 15 insurance institutions over the next 30 years, drawing on the NGFS scenarios. The pilot exercise presented meaningful methodological innovations, such as adopting a dynamic balance-sheet assumption, allowing financial institutions to invest in and out of economic sectors based on climate risk-reward considerations revealing moderate exposure of French banks and insurers to climate risks. However, the exercise did not allow for the simultaneous impact of specific climate events on the economy and financial institutions.
4. The Bank of Canada
The Bank of Canada conducted a climate scenario pilot project with six Canadian financial institutions. The method to assess credit risk combined top-down and bottom-up approaches, but the project only focussed on variables from climate transition scenarios across sector-level financial impacts. The project estimated a relationship between climate transition information and credit risk using the transition scenarios' financial impacts and stressed credit outcomes.
5. Hong Kong Monetary Authority (HKMA)
The Hong Kong Monetary Authority discovered with the pilot climate risk stress test (CRST) that climate risks could potentially cause substantial adverse impacts on the banking sector under extreme scenarios and that banks need to take early measures to manage them. The test involved 20 central retail banks and seven branches of international banking groups participating, accounting for 80% of the banking sector’s total lending. Participating banks assessed their climate risk exposures under three scenarios: a physical risk scenario involving a worsening climate situation and two scenarios envisaging orderly and disorderly transitions to a low-emission economy. Overall, the test evaluation found that the banking sector remains resilient to climate-related shocks, given the substantial capital buffers built up for the banks within the tests over the years. Still, some of the banks’ operations would be impacted by physical risks of climate hazards.
6. The Federal Reserve (The Fed)
The Federal Reserve announced in September 2022 that they would conduct a pilot climate scenario analysis exercise on six of the largest banks in early 2023, including Bank of America, Citigroup, Goldman Sachs, JPMorgan Chase, Morgan Stanley, and Wells Fargo. The Fed Board's stress test exercises assess whether large banks have enough capital to continue lending to households and businesses during a severe recession focusing on the potential future physical risk drivers for the participating banks' real estate portfolios. The pilot is set to partially reflect the uncertainty of assessing physical risks by incorporating a range of potential physical risk shocks and mitigation assumptions rather than focusing on a single shock. In addition, by specifying possible emission pathways and a future date for an extreme event, the pilot exercise brings forward potential future climate risks from 2050, when the effects of physical risk drivers are likely to become more severe. The climate scenario analysis exercise is exploratory and does not have capital consequences focusing on both physical and transition risks.
Uncertain Timing
Criticism of climate stress tests has amounted. This includes the current approaches to stress tests taking too long, needing more complexity and being a poor choice of resources for more impactful measures. Stress testing is a valuable tool to understand the more considerable uncertainties of physical-climate-related risks. However, the quantitative modelling does not compensate for the more profound ‘unknowns’ attached to underlying socio-economic mechanisms. Moreover, the assessment of physical risks is subject to uncertainty and not suited to conventional financial risk analysis, which makes an assessment in the context of climate stress tests particularly difficult.
Uncertainty centre’s upon the timing of climate-related risk materialisation. Based on current balance sheets, the timing with physical climate change appears inconsistent with the time approach of traditional stress tests attempting to measure outcomes over much longer time horizons. Exemplified by the Bank of England biennial exploratory scenario, the stress test proposed a scenario within a 30-year timeline with a fixed 2020 balance sheet. The physical changes of climate change are proposed to be explored with the same 2020 balance sheet. This coincides with how regulators view the balance sheets of financial institutes. For example, the Bank of England (BoE), Bank of Canada (BoC) and Hong Kong Monetary Authority (HKMA) assumed a static balance sheet under which the size, composition and risk profile do not vary over the stress testing time horizon. But several other jurisdictions have used a hybrid approach, assuming a dynamic balance sheet for specific long-term horizons. However, excessively long planning horizons in climate stress tests create enormous challenges in designing and executing these exercises. A long horizon enables the impacts of climate change to materialise, but it can lead to uncertainty about the events that will impact financial institutions.
Lack of historical data
A key challenge for all climate stress tests is a need for historical granular data vital as a baseline for modelling the interactions between climate, the macroeconomy and the financial sector needed to design coherent climate scenarios to understand climate risk management fully. Historical data is an essential tool for understanding and preparing for the impacts of climate change. Caroline Liesegang, head of prudential regulation and research at the Association for Financial Markets in Europe, said that “climate risk analytics and climate risk data is still in its infancy,” with a lot of work to do closing data gaps. In particular, stress testing usually relies on comparing a limited set of scenarios over short periods with the ratio function of the agents based on historical data, limiting the range of possible outcomes. Furthermore, even with realistic scenarios, it is challenging to deal with unprecedented events based on historical events in the absence of any equation of state. More historical data and model validation is needed for longer-horizon projects to be practical when analysing climate scenarios. The need for understanding market data about climate risks is often cited as a reason for the limited progress. Still, analysts have questioned this rationale believing that it doesn’t make sense to expect research outcomes in the future to support policy decisions that must be taken in the present to have any impact.
Ask our Climate X experts direct questions about how we are the most relevant and effective data provider with our chat box.
Multiple scenarios
Another critical attribute stress testing relies on is assessing an explicitly limited number of scenarios. That is even its main advantage in the face of the multiplicity obstacle. Scenario analysis, particularly stress testing in finance, usually compares a limited set of scenarios over short periods with the agents' reaction function based on historical data. This inevitably limits the range of possible outcomes. Some future realisations may appear impossible or improbable that they are not worth considering. Still, even with realistic scenarios, it is difficult, if not impossible, to deal with unprecedented events based on historical events in the absence of any equation of state. Climate stress tests should have multiple scenarios to capture uncertainty and risks with climate change. Overall, there needs to be more capacity for models and modellers to capture the complexity at stake.
Assessing Climate Risk Credit Modeling capabilities
According to the results of previous climate stress tests, banks are in an early stage of factoring climate risks into their internal credit risk stress-testing models. The ECB reported that most banks do not include climate risk in their credit risk models, and just 20% consider climate risk as a variable when granting loans, posing a risk to financial stability globally. Moreover, credit risk parameters seem fairly insensitive to the climate shocks included in the scenarios. For this, banks will need to enhance their modelling capabilities. Positive actions by the ECB climate stress exercises were the inclusion of sectoral risk parameters following the asymmetric shocks across industry sectors, actual data usage and the inclusion of both direct and indirect transmission channels. However, climate stress tests are still in their infancy, with a lot more work to be tested and accomplished to fulfil their benefits as tools for financial institutions, with banks being able to improve the tests themselves.
Upping the Ante
For banks, climate risks will add a layer to risk management despite climate stress tests having no direct implications for capital requirements at banks. Yet, the UK's Prudential Regulation Authority (PRA) expects banks to capitalise on climate-related financial risks. It could also require an additional buffer if it deems a bank's risk management inadequate. But for climate stress tests, key challenges exist, including the planning horizon, scenario design, and a need for historical data. While central banks have recognised these challenges, rectifying these shortcomings has been limited to accomplishing little progress with integrating physical risk effects into climate stress exercises. For financial institutes, it is crucial for research to continue in this area and for banks to continue to make advancements in their internal scenario analysis. Such efforts will help advance the testing methodologies by publishing relevant work and more focused case studies on the impact of physical risks on bank portfolios. Although there are unsurmountable challenges to using climate stress tests to set capital requirements, the underlying analysis and research will gather valuable information for public policy on climate change. And while the costs of policies to slow climate change are immediate and relatively easy to see, the prices of not acting are harder to measure.
“Euro area banks must urgently step up efforts to measure and manage climate risk, closing the current data gaps and adopting good practices that are already present in the sector,” stated Andrea Enria. Climate X is a solution that provides entities with data to help them quantify and prepare for climate-related physical risks, an essential tool in building climate resilience to support the need for banks to rise to the challenge of developing modelling/methodology capacities with real data to meet the demands of increasing regulatory climate stress tests pressure and expectations for years to come.
Sources
- Jim Clayton, Steven Devaney, Sarah Sayce and Jorn Van de Wetering, The Journal of Portfolio Management Special Real Estate Issue 2021, 47 (10) 75-90; DOI: https://doi.org/10.3905/jpm.2021.1.278
- https://coast.noaa.gov/states/fast-facts/climate-change.html
- https://www2.deloitte.com/content/dam/insights/articles/us175539_cfs_fsi-outlook-commercial-real-estate/DI_CFS_FSI-Outlook-Commercial-real-estate.pdf
- https://www2.deloitte.com/us/en/insights/industry/financial-services/commercial-real-estate-outlook.html
- https://www.efrag.org/Assets/Download?assetUrl=%2Fsites%2Fwebpublishing%2FSiteAssets%2FFinancial%2520Institutions%2520slides.pdf&AspxAutoDetectCookieSupport=1
- https://www.knightfrank.com/research/article/2022-06-06-global-house-price-index-q1-2022
- https://www.nar.realtor/on-common-ground/real-estate-in-a-riskier-climate
- https://www.perenews.com/what-climate-change-means-for-real-estate-real-risk-big-impact/
- https://www.pwc.com/gx/en/asset-management/emerging-trends-real-estate/assets/Emerging%20Trends%20in%20Real%20Estate%20Europe%202023%20Report.pdf
- https://www.qra.qld.gov.au/sites/default/files/2022-07/dae_report_-_south_east_queensland_rainfall_and_flooding_event_-_8_june_2022.pdf
- https://www.realestate.com.au/news/how-australias-flood-crisis-could-impact-property-prices-in-high-demand-regions/
- https://reliefweb.int/report/world/climate-crisis-destroy-167-million-homes-next-20-years
- https://www.reri.org/research/files/RERI_Bond_McCoy_and_McDonough_2022.pdf
- https://www.us.jll.com/en/trends-and-insights/investor/how-regulators-are-tightening-the-screws-on-sustainability-reporting