Comment Detail

Comment

3. What methodologies, datasets, variables, assumptions, future climate scenarios, and measurement tools are used to measure and monitor climate risk to the national housing finance markets? Describe any gaps in available data that limit the ability to measure such risks. How could such data gaps be resolved?

A common, agreed-upon set of scenarios, time horizons, acceptable risk levels, and metrics are among the criteria that enable companies to assess and manage changes in climate risk over time. These criteria also help facilitate and accelerate more meaningful disclosure reporting comparisons within and across industries. Today, analytics are sufficiently mature to support comparisons among companies within a given industry or sector.
However, more needs to be done to be able to accurately make comparisons across sectors. Such requirements should leverage private sector investments to help optimize relevant federal priorities and policies.
Overall, Jupiter believes that climate data:
Must be assessed for bias; 2) is downscaled; 3) can model extremes; 4) is up to date; 5) is transparent.
Moreover, Jupiter strongly believes that climate metrics should be based on models from the most recent phase of the ongoing Coupled Model Intercomparison Project (CMIP).
The federal government should lead by example and be held accountable to climate risks, ESG disclosure metrics and standards, as well. Jupiter strongly believes a nuanced view of the state of science and analytics should guide the metrics and tools used, rather than a desire for a simple one-size-fits-all approach.
4. What risk management strategies or approaches—including but not limited to those related to pricing, insurance, credit risk transfers (CRT), loss mitigation, and disaster response—do industry participants use to address climate and natural disaster risk?
At this time almost no industry participants have enterprise-wide deployments in place to address the issues raised in the question.
To a large extent, most industry participants use 30-year-old, backward-looking risk management models and approaches that do a poor job of accounting for how risk has already changed and continues to worsen due to climate change. They often do not consider the ongoing impact of climate change (or what some refer to as “future conditions'') on assets with durations over one year. During the last two years, some leaders in the financial system have begun to address this issue with approaches that vary from investigation and experimentation, to multi-year commitments to implement risk management systems that account for climate change in a rigorous, responsible, professional way.
6. With respect to the foregoing questions, FHFA invites interested parties to submit any studies, research, data, or other qualitative or quantitative information that supports a commenter’s response or is otherwise relevant to the regulated entities’ climate and natural disaster risk.
Jupiter’s high-resolution, forward-looking risk analysis from dynamic Earth System models, combined with dynamical and data-driven downscaling techniques, produce hyper-local hazard estimates based on projected changes in climate.
Jupiter products were created by global experts in climate and Earth science, data science, technology, risk management, business (with extensive experience in Fortune 100 companies), government, and academia.
Jupiter’s transparent methodologies employ dozens of the scientific community’s most respected physical models of the atmosphere and hydrosphere coupled with machine learning, land use, elevation data, and extensive observations of the systems we model. We build robust verification and validation into every step of the model chain, and our methods natively account for the changing frequencies and characteristics of extreme events in time.
Jupiter ClimateScore Global quantifies physical climate risk around the world. With its expansive and detailed projections of perils for the remainder of this century, it answers the fundamental question, “How will the current and future climate lead to extreme weather that impacts my locations of interest?”
ClimateScore Global employs dozens of the scientific community’s most respected climate models, coupled with machine learning, land use and elevation data, and data-driven models for hydrology, wildfire, severe weather, and more.
By distilling the complex interactions between expected changes in sea levels, surge, storm intensity, land and sea surface temperatures, and pressure and precipitation patterns, ClimateScore Global pinpoints what decision-makers need to know: the depth of the water, the speed of the wind, the intensity of the heat, and the probability of drought, wildfires, and hail. Those metrics, in turn, will drive damage and economic loss models and put a price on climate change.

II. Enhancing FHFA’s Supervisory and Regulatory Framework

Climate and Natural Disaster Risk Management Request for input
8. What specific processes and systems of a regulated entity should FHFA examine in its supervision of the regulated entities’ climate and natural disaster risk management?
At the highest level, FHFA should ensure that there is a Board, CEO and senior executive commitment to understand, manage, reduce, and disclose risks associated with climate change. In many cases throughout the global financial system, even this foundational step has not yet been taken.
Once that commitment is in place, best practices for integrating climate risk into overall risk management practices exist and are increasingly well documented. At least one of the top 3 US Banks already is far along in doing exactly this, and Jupiter sees signs that the remaining top 3 banks are accelerating their investments along the same lines.
At present, Jupiter is aware of one GSE at the forefront of these issues. However, it currently appears that the large banks are on the verge of eclipsing all the GSEs in this area.
FHFA should examine the following:
What is the board level process for reviewing climate risk management practices in the GSE?
What conclusions have the CEO and board drawn regarding the state of existing risk management practices?
What commitments & investments are in place to bring climate risk management practices to best practices level and for continuous improvement?
What processes, data, and standards are the GSEs using to support these activities?
What disclosures to regulators and shareholders are GSEs making, how do those disclosures compare to best practices in the U.S. and globally?
Generally speaking, many of these issues are addressed in broad conceptual ways by the Task Force for Climate-Related Financial Disclosure (TCFD).
13. Should FHFA implement a stress testing, scenario analysis, or similar program to assess the regulated entities’ climate and natural disaster risk? If so, what factors should FHFA consider in defining the purposes, design, and scenarios of any such programs?
FHFA should implement stress testing and scenario analysis to assess the regulated entities’ climate and natural disaster risk, when the GSEs have implemented sufficient internal processes to adequately respond. Jupiter’s view, based on our direct experience, with both the nation’s largest banks and GSEs, is that some period of time is required before any of these entities can pass a stress test and, consequently, FHFA should define a timeframe for satisfying any requirements.
This approach of providing a clear roadmap of expectations follows the path of the Bank of England, Prudential Regulation Authority (PRA) and similar regulatory bodies throughout Europe, Asia and Australia.
Jupiter recommends one year from the date FHFA definitively communicates this requirement.
16. Market discipline could potentially supplement FHFA’s supervision and regulation of the regulated entities’ climate and natural disaster risk appetite and management. Market discipline depends in part on the information that is available to shareholders, creditors, and other counterparties. Is the existing publicly available information sufficient for shareholders, creditors, CRT and other investors, and other counterparties to understand and exercise market discipline over a regulated entity’s appetite for and management of climate and natural disaster risk? If not, what changes are needed? Should each regulated entity be required to disclose additional information, including but not limited to the extent to which its underwriting practices take into account climate and natural disaster risk?
The existing publicly available information is frequently insufficient for shareholders, creditors, CRT and other investors, and other counterparties to understand and exercise market discipline over a regulated entity’s appetite for and management of climate and natural disaster risk.
However, the tools for satisfying the parties above exist, provided the parties implement them in a rigorous and professional manner consistent with generally accepted risk management practices.
As to the final question: Yes, each regulated entity should be required to disclose additional information, including to the extent to which its underwriting practices take into account climate and natural disaster risk.
18. Policies to manage climate and natural disaster risk could increase the cost of housing, making it more difficult for lower income households in some areas to obtain affordable housing. Are there policies the regulated entities could pursue to mitigate such adverse effects for lower income households in vulnerable areas without undermining efforts to manage climate and natural disaster risk?
The implementation of rules intended to ensure fairness and equal access to housing now have the unintended consequence of often limiting the ability of lower income families to relocate away from high-risk areas in a way that more affluent citizens can more easily manage.
The scope of this issue likely requires some congressional action to address. Jupiter is available to discuss our views on this issue in more detail and provide technical assistance to FHFA to better understand and address this critical topic.
21. What specific issues or topics should FHFA consider for future research on climate and natural disaster risk to the regulated entities and the national housing finance markets?
Jupiter is available to provide technical assistance and additional background on this topic.
22. What data or housing market information would be beneficial for FHFA to make available, to the extent permitted by privacy considerations, to researchers and other interested parties to support the assessment of climate and natural disaster risk to the regulated entities or the national housing finance markets?
FHFA should make available all data at the minimum level of aggregation necessary to preserve consumer privacy. This could be at the neighborhood or census level.
Jupiter provides similar information to researchers and some other interested parties at no charge.

Climate and Natural Disaster Risk Management Request for input
24. Are there existing or potential government agencies or programs that FHFA could partner with to enhance the Agency’s supervision and regulation of climate and natural disaster risk to the regulated entities?
The same week that the submission to this RFI is due the House Science Committee is holding hearings on a National Climate Service. It will be critical for FHFA to understand whether - and when - a National Climate Service can provide sufficient information to enhance the Agency’s supervision and regulation. Other alternatives include private sector services, leverage over $100m in capital invested into these services to date.
25. What, if any, other enhancements should FHFA consider to its supervision and regulation of each regulated entity’s management of climate and natural disaster risk? Other enhancements could include but need not be limited to: (i) regulatory capital requirements or other loss absorbing capacity requirements that ensure each regulated entity has the capacity to absorb impacts of climate and natural disaster risk; (ii) disclosure requirements to provide shareholders, creditors, CRT or other investors, and other counterparties with appropriate information about a regulated entity’s climate and natural disaster risk; and (iii) changes to FHFA’s supervisory program to enhance examination of or reporting on each regulated entity’s infrastructure and processes for identifying, assessing, mitigating, and monitoring the regulated entity’s management of climate and natural disaster risk.
FHFA should ultimately enhance its regulation and supervision so that regulatory capital requirements or other loss absorbing capacity requirements ensure each regulated entity has the capacity to absorb impacts of climate and natural disaster risk. However, GSEs and FHFA likely require substantial work to structure these requirements in ways consistent with rapidly evolving tools and stakeholder expectations. A clear roadmap for defining and implementing this and any other requirement should be established and communicated as a first step.
As noted above, FHFA should implement these disclosure requirements.
As noted above, FHFA should enhance its supervisory program to implement examination of and reporting on each regulated entity’s infrastructure and processes for identifying, assessing, mitigating, and monitoring the regulated entity’s management of climate and natural disaster risk.
26. To what extent, if any, should FHFA support efforts to develop standards of classification and data reporting on climate and natural disaster risk to the financial performance of companies, such as those by the Sustainability Accounting Standards Board, domestic and foreign government agencies, or others?
Standards for duration, risk tolerance, and climate scenarios considered are extremely important to ensure consistency of information across and within regulated entities. Jupiter can provide technical assistance in developing these standards. Jupiter also recommends that any standards or more general requirements be developed considering other emerging standards in the U.S. and globally to ensure consistency of requirements to the extent possible.