Why is it important to develop climate stress testing capabilities to accurately assess climate impact on organizations?

Climate change impacts real economy and financial system due to adverse consequences of rising temperature. Linkage between anthropogenic emissions and rising temperatures have been established scientifically (IPCC reports). To accomplish net zero future, green house gas (GHG) emissions are required to be reduced by at least 45 per cent by 2030; and to net zero by 2050. This shift to a low-carbon future brings with them distinct challenges and risks for economic sector.

Climate risk is primarily transmitted in two different channels, i.e., physical risk, and transition risk. Physical risk emanates from the interaction of climate-related hazards and corporate assets resulting in financial losses. Financial losses primarily arise from increasing frequency and severity of climate-related events (e.g., storms, floods, heatwaves) and significant shifts in the existing climatological patterns (e.g., ocean acidification, rising sea levels and changes in precipitation). Transition risk, however, is associated with the uncertain financial impacts that could result from a rapid low-carbon transition, including policy changes, reputational impacts, technological breakthroughs, and shifts in market sentiments, consumer preferences and social norms. A transition to a low-carbon economy has a direct impact to the fossil fuel industry, gasoline-fueled automobile manufacturers, electric utilities, and other industries, leading to significant financial implications for their creditors and shareholders including banks and financial services companies.

Financial sector needs tools to help it understand and manage the risks and rewards associated with climate change including transition and physical risks. It becomes a business imperative to identify, measure, and manage climate risk drivers that can impact their financial performance through financial or non-financial prudential risks such as credit, market, counterparty, operational, liquidity and funding risks. This can be accomplished by developing climate stress-testing programs similar to capital stress testing programs that have become now common across banks since the 2008 financial crisis. As a forward-looking exercise, climate stress tests measure a firm’s exposure to climate risks using severe climate risk scenarios.

Also, regulators are pursuing stress testing (top down or bottom up) exercises in multiple jurisdictions to assess the climate risks and how vulnerable banks are to these risks. In response, financial institutions are developing climate stress testing capabilities to implement them to assess the impact. Climate stress testing results can inform institutions’ decisions regarding their business models, planning and strategy.

Specifically, insights are derived on the impact of physical and transition risks at different points in the future, along the emission transition pathways. These insights allow financial institutions to identify which sectors and counterparties are most at risk in portfolios under the scenario assumptions and identify the climate drivers for these risks. Importantly, they help in updating risk appetite statements, develop mitigation and adaptation strategies through client engagement in vulnerable sectors, identify gaps in the firm’s infrastructure and data related areas and develop appropriate remediation plans and provide good quality information for external disclosures.

What are the advantages of being able to leverage scenarios to create a meaningful analysis?

Scenario analysis has been in use within financial institutions as part of capital stress testing programs (CCAR and ICAAPs) for many years to explore potential implications of changes in macro-economy and financial markets. Similarly, climate scenario analysis is becoming a powerful tool that can help financial firms to explore the potential implications of physical (e.g., storms, floods, heatwaves) and transition risks (e.g., including climate policy, availability of clean technologies and changes in customer behaviours).

Financial institutions consider a variety of scenario narratives to understand the magnitude and impact of low carbon transition activities on their financed emissions portfolios. For instance, net zero 2050 scenario narrative, if all emission intensive sectors adopt low carbon policies and implement them in an orderly fashion then temperature rise will be limited to 1.5^0C.. However, at this time, there is only 50% chance to limit temperature per update from IPCC. Now, there are other scenario possibilities that could be realized. They will have different order and magnitude of physical and transition risks impacting financial performance. Currently in various supervisory regulatory examinations delayed transition and hot house world scenario narratives are being used to assess exposure to climate risks.

Also, each of these scenarios needs to be assessed by considering specific circumstances of the business. Specifically, transition scenario analysis to include possibilities such as a utility company phasing out coal and transportation company going electric. Hence, scenarios should be considered based on their suitability and fit for purpose. For instance, for risk management purposes, downside scenario or the ‘Disorderly’ adjustment scenario is likely to result in a high degree of transition risk, given the sharp and abrupt adjustment required to limit the increase in temperature to 2^0C. This scenario can help evaluate the impact of transition on valuations of assets and collateral and the default risk of loans and to calibrate risk appetite and explore the actions necessary to manage risk and alter the risk profile of the institution. Additionally, when firms are making commitments about decarbonising portfolios and developing business plans, alignment scenarios could be used to identify appropriate business strategies and measure the achievability of plans to build sustainable businesses.

Overall, climate scenario analysis can serve a variety of functions including setting climate-related limits and targets, developing a climate strategy, making internal and external climate disclosures, improving underwriting and investment criteria, and meeting regulatory requirements.

What are the challenges associated with designing scenarios and practically implementing stress tests?

Financial institutions are increasingly adopting Integrated assessment models (IAMs) in designing scenarios due to their ability to describe the interactions between economic activity, GHG emissions and climate change. IAMs provide global coverage, have long time horizons and give adequate details on mitigation pathways that provide high-level overviews of tradeoffs and constraints. These scenarios provide insights into different configurations of the energy system, industry, land use and agriculture, they need to be further extended to create scenarios that consistently describe the associated financial risk and reward conditions. These models are utilized by Network for greening financial system (NGFS) in their scenario design that are being used by regulators and financial industry to conduct physical and transition risk analysis as part of climate stress testing.

There are some major challenges associated with the design of climate scenarios based on the IAMs. Some of these include:

• Climate scenario analysis due to its unique nature, brings new challenges of people with skills including various areas of climate science, financial risk, engineering technologies, and energy markets. They are essential to perform the assessment of the assumptions on future state of the world (Policy guidance, Technology advances, consumer behaviours) embedded in the IAM model, development of bespoke climate scenarios and communication of the scenario analysis outputs. This will help when communicating the climate scenario stress testing outputs to variety of stakeholders.
• Scenarios are in general have long term horizons (10 – 30 years) and needs to be supplemented by short term (1-5 years) scenarios to eliminate smoothening over long term horizon. IAM model outputs physical and transition risk variables at global level and are required to be further downscaled to sector, regional and country level and where necessary subject to variable expansion to be fit to use at firm level. This kind of downscaling with the help of models and proxy data (lack of quality data) usage in addition to IAMs could increase the model and data risks.
• Most of the current practice is around assessing physical and transition risks independently that is contrary to real world experience. The impact of physical risks on transition pathways arises due to idiosyncratic characteristics of each sector. This manifests in the form of direct impact to the transition related investments (infrastructure) and indirect effects from impact to other resources including natural resources and supply chain. This requires incorporating non negligible impact of the physical risks on transition risks in the scenario analysis.
• Uncertainty associated with interactions among climate, macro-economic and financial drivers is complex and impact countries, regions, sectors and companies disproportionately. Therefore, within the organization, proper review, challenge and assessment of scenarios, model assumptions is critical in selecting candidate scenarios from multitudes of scenarios to ensure the stress testing results are meaningful.

Once these climate risk scenarios (physical and transition risk) have been designed and their impact on macro (e.g GDP, Unemployment rate) and micro variables is performed in conjunction with financial risk models for implementing stress testing programs.

How can financial institutions prepare their data, governance and methodologies for climate stress testing?

Climate stress testing execution is an enterprise wide program that needs preparation across multiple fronts including data, governance and methodologies and coordination among the front line business units, risk and audit functions. Specifically,

Data: In general, an institution requires the following information 1) macro financial (macro-economic, financial markets) 2) Portfolio exposures (climate sensitive) 3) climate change related (decarbonizations plans, emission pathways, weather, geospatial data etc) as key inputs in the climate stress testing program. There are many challenges in gathering the climate related data including limited availability, insufficient granularity, lack of good quality and not adequate comparability to other firms. Therefore, Institutions are required to develop policies and procedures for collection, processing and use of the data. Best practices to be put in place for data collection including identification of data needs, understanding the availability of data sources, implementing industry standards, validating data, identifying data gaps and adapting institutional systems. Also, filling the data gap by development of in-house and ensuring access to external opensource and proprietary information on climate risk exposures and physical and transition risks data.

Governance: Board and senior management of the financial institution should have understanding of the financial risks arising due to climate change to overall business strategy and risk appetite. There should be sufficient governance and oversight through appropriate policies and procedures. An effective design of policies and procedures are needed to integrate the climate risk into business decisions and operations across the enterprise. These policies and procedures should address four distinct priority areas at enterprise level, as follows: (i) the corporate strategy in addressing the impacts of both physical and transition risk in existing and future assets and financial positions of the institution through the development of a new climate risk management framework; (ii) the compliance framework for both voluntary and mandated climate risk regulations including stress testing components in the jurisdictions that the institution operates in; (iii) corporate strategy in monetizing the climate risk-related opportunities in the markets, including but not limited to trading the climate-related derivatives and financial instruments; and (iv) the scope and scale of ESG activities as they relates to climate risk management process which need to be addressed under an overarching umbrella of enterprise risk management (ERM).

Methodologies: Models play an important role of translating climate related risks (through the scenarios) into financial impacts at firm level. Importantly, as part of the climate risk quantification of physical and transition risks impacting a firm’s financial position (cashflows, revenues and balance sheet) exposures are modeled. This analysis is highly dependent on the data including corporate decarbonization plans, balance sheet and financial statement information. To measure for instance, credit risk embedded in the financed emissions, credit risk parameters (PD/LGD/EAD) are estimated to quantify their exposures at individual counterparty level. Overall, these models can be used in conducting loan level analysis on impact of climate risk drivers. Biggest challenge lies in adapting financial risk models (credit risk (PD, LGD) dynamic balance sheet (revenues), equity and bond valuations) to capture the transition risk dynamics in long term scenarios along with data challenges (firm level data on balance sheet, energy expenditures and decarbonization plans).

Therefore, firms should lay out a strategy as part of their usage of methodologies for climate stress testing. Determine the portfolios exposures (energy, real estate), corresponding parameters and model inputs required to be estimated. Followed by a review of inhouse, vendor and other modelling capabilities define an approach to deploy appropriate modelling methodologies to identify, measure, and manage credit, market, counterparty, operational and liquidity risks emanating from climate-related risks and conduct the stress testing.

Additionally, Model risk management should understand how climate risk assessment models are fit for given purpose. Climate models have their own set of challenges. One of them being that there is deep uncertainty associated with climate risk models. Deep uncertainty usually involves decisions that should be made over time in dynamic interaction with the system. For climate risk models there are many plausible and even “unknown” system models and actual outcomes.  Therefore, the current model risk management frameworks should be adapted to climate risk models as it will help to evaluate the validity and degree of uncertainty of model outcomes at the firm level. Internal audit to partner and provide independent view on the climate risk management.

How must organisations adjust methodologies to convert data into impact?

Financial institutions utilize model methodologies to identify, measure, and manage macro financial risk drivers that can impact their financial performance through financial or non-financial prudential risks, such as credit, market, counterparty, operational, liquidity and funding risks. Additionally, to capture climate risk drivers, these methodologies are required to be either updated or redeveloped depending upon the purpose within the stress testing. For instance, credit risk models such as internal risk rating models used in institutional credit decisioning purposes take into consideration macroeconomic environment and client’s financial position when estimating the model parameters. These models can be adapted by incorporating climate risk (physical risk data such geospatial information, transition risk data such as emission profile and mitigation strategies).  In the case of retail loans to household sectors (mortgages, auto financing), modelling approaches need to be modified to incorporate impact from physical hazards (sea level rising, flooding etc) and transition risk variables (electric car and energy efficiency measures).

On a wider basis firms need to undertake a gap analysis of current inventory of model methodologies (top down, bottom up and alignment) in the light of significant hurdles in obtaining the data to be consumed by these models and identify areas requiring further development or refinements.  Furthermore, institutions need to engage with external modelers, vendors and academics to develop internal models and improve their linkages to external models (IAMs and others) to align with guidance and specifications provided by regulators.

Why is it important for financial institutions to understand the impact of climate events and calibrate this data with long term goals?

Many countries and institutions have pledged to net zero emissions either before 2050 or after. Financial institutions are focused on transitioning their portfolio financed emissions to net zero by 2050 in alignment with Paris agreement. Fossil fuel laden energy system should be rewired to reduce the emissions. A transition to a low-carbon economy has a direct impact to the fossil fuel industry, gasoline-fuel automobile manufacturers, electric utilities, and other industries, leading to significant financial implications for their creditors and shareholders including banks and financial services companies. Financial institutions with exposure to the energy and utilities sectors face the starkest challenges in navigating the transition. These sectors have long had the difficulty of trying to plan for long-term capital assets in an uncertain policy and resources landscape. This transition is fraught with risks and deeply uncertain. There are risks and opportunities arising from this transition. Therefore, financial institutions need to understand climate risk drivers and their impact.

According GFANZ³ analysis public and private investments of USD $ 125 trillion is flowing to low carbon energy supply to attain net zero by 2050. Also, approximately 32 T across high emission sectors to implement decarbonization strategies by 2030. IPCC confirmed that there are technology options available today in the areas of renewables, carbon capture and storage, Hydrogen and alternative fuels to reduce emissions half by 2030.

However, there are challenges to capture this scale of investment and opportunity. Also to achieve portfolio financed emission reduction targets for financial institutions is easier said than done due to complexities involved with

• Engagement with clients in understanding their decarbonization plan.
• Collaboration with multiple stakeholders from regulators, and industry.
• Development of clear mandate with the key internal stakeholders.

This requires a framework comprising of climate risk strategy, scenarios analysis and methodologies to understand and mitigate the risks with good quality decision-useful information for assessment of climate risks to finance investments of this scale for climate-related opportunities.