Gaining a Competitive Advantage Through Climate Risk Management
Risk management has long been established as a competitive advantage in financial services firms. It is the skill by which firms avoid exposures to inflated assets and avoid potentially large losses that typically occur abruptly. It is also a means by which firms attract clients and investors and satisfy regulators and other stakeholders. In the new era of climate risk management, we shall come to see the same ebb and flow of fortunes made and lost as some players grasp a lead over their competitors and gain a competitive advantage.
The science of climate risk is well documented and the incidents of impacts are well chronicled. The changes in laws and regulations impacting financial services firms and other industries are coming faster than many anticipate. So, as the world shifts its business practices to embrace all things sustainable, the future success of firms rests firmly on how they embrace climate risk management.
For the financial services sector, there is substantial momentum across a wide range of sustainability themes. For many firms, the discussion around the myriad of sustainability themes may still be at a high level, looking at board governance, group transition objectives, group research and strategic options. For other firms, they are already grappling with the challenge of how to take the first steps into a practical business operations process to measure, monitor, manage and report on specific elements of sustainability.
Given the embryonic state of solutions in this market, one might think that firms have the luxury of time to explore solutions and implementation. However, the momentum of government legislation and regulation combined with other stakeholder pressures (e.g. investors, customers, partners, employees and suppliers) is resulting in a faster timetable of change than many anticipated.
The Focus on Climate Risk
As always for such a broad subject, one has to start somewhere. In the case of financial services, the most pressing issues are coalescing around climate risk disclosure,with its origins in the TCFD (see https://www.fsb-tcfd.org/ ). The disclosures are now being pursued by national lawmakers to make them mandatory with, inter alia, two major objectives:
- Assess the GHG Emissions of a firm -to measure contribution to global warming;
- Assess the Capital Adequacy of a firm - to stress-test exposure to global warming.
Measure, Monitor & Manage Climate Risk
The cycle of risk management is a well established practice of measure, monitor and manage. Each of these apply to climate risk management in the same way. The first solutions may be rough, but getting going is essential to create a virtuous cycle of build-test-improve-repeat.
Table 1: Summary of practical aspects of the data process (input, derived, risk metrics)
|Metrics & Standards||Competitive Frequency||Information Delivery Mechanism|
|Data & Models||Intraday Monitoring||Presentation Layer|
|Modular Design||Machine-based Monitoring||Ex and Cum Climate Risk Metrics|
|Build versus Buy||Human-based Monitoring||Human-Machine Interaction|
|Buy Components vs. Buy Service||Data & Model Improvement||Human Commentary Input Mechanism|
Only by measuring risk does anyone know the exposure. One may have a great instinct for the quantum of risk a business is running, but only bymeasuring and aggregating risks across the business will one have evidence of the firm’s risk. To do this, one needs a framework, assumptions and a mathematical method, but these depend on data in order to create business value. We must define and source input data, define intermediate derived data and define the final derived risk metric data. Together, these form the basis of a risk management information and decision making tool. So, let’s consider practical aspects of achieving this objective.
Practical Point 1: Metrics & Standards Define a narrow set of metrics to start with. Even if the first attempt is poor, through iteration this can improve.
Practical Point 2: Data & Models Do not expect to continue to use the same data and models. All will be in flux for some time to come.
Practical Point 3: Modular Design Adopt a components-based approach to any solution. This will allow you to switch and change as data and models evolve.
Practical Point 4: Build versus Buy This is always a hard decision. Fortunately, we live in an enlightened era for finance, where collaboration and value-chain partners are becoming normal.
Practical Point 5: Buy Components versus Buy Service Even if one buys components, there is still the issue of assembling them which can present a significant build-and-maintain effort. It may be prudent to divide the platform into two elements: one that is internal and one that is an external business process service that plumbs neatly into the existing internal business processes.
Once we measure, then it’s necessary to re-measure at different time intervals to monitor how the risk is evolving. Determining the frequency of monitoring is always a challenge: too slow and risks can spiral out of control, too fast and effort can spiral out of control. It is also a function of the nature of one’s business (6, 7).
Practical Point 6: Competitive Frequency Higher frequency monitoring may well be a huge competitive advantage when a crisis occurs.
Practical Point 7: Intraday Monitoring The initial requirements of an embryonic market rarely require intraday or near real time monitoring. Yet, over time, history has shown us that as events transpire, crises occur and regulation tightens, higher frequency monitoring is inevitable.
Practical Point 8: Machine-based Monitoring Make automation central to the monitoring process. It can add significantly to the monitoring process, highlighting material moves and promoting them for scrutiny and management.
Practical Point 9: Human-based Monitoring Assign a small number of highly skilled humans who understand the measurement process to add value to the monitoring process. Just like any automated process, it needs a degree of oversight to identify when automation is not capturing the essential issues or is promoting the wrong issues.
Practical Point 10: Data & Model Improvement The Machine-Human monitoring is a basis for identifying strengths and weaknesses in the measurement process. Use it to drive iterative improvements.
Measuring and monitoring provides us with the information we need to make decisions on risk exposures - to increase, to decrease or to do nothing. Any effective management and decision tool delivers information both fast, meaningfully and in a manner where it can be assimilated, understood and acted on fast.
Practical Point 11: Information Delivery Mechanism Avoid delivering climate risk as a separate report, dashboard or tool - aligned it with existing tools.
Practical Point 12: Presentation Layer Leverage existing tooling for dashboards, visualisation, commentary and reporting.
Practical Point 13: Ex and Cum Climate Risk Metrics Provide context for the climate risk measurements by showing how a risk metric changes due to the inclusion of climate risk considerations.
Practical Point 14: Human-Machine Interaction Machine Learning is changing the effectiveness of management. We see NLP (natural language processing) being used to ingest human commentary to enhance data analysis and we see NLG (natural language generation) being used for promoting and reporting issues. Together, these can form a virtuous cycle whereby machines and humans promote and demote issues as they evolve and trend. Use this to create faster iterative improvements to intelligence.
Practical Point 15: Human Commentary Input Mechanism Provide ability for humans to add commentary on risk to be ingested into data using NLP. This will enhance the iterative process of improving and measuring the effectiveness of human risk management intervention.
Johnny Mattimore, First Derivatives