Part 2: Climate Risk Management - Methodology and Assumptions

[Ankit_Pal]

Welcome back to Part 2 of our blog series where we (Ankit and Saeed) delve into the process of executing a climate risk scenario analysis within SAP Profitability and Performance Management (PaPM). PaPM comes with a wide range of sample business scenarios that can be run out of the box. In this series, we aim to guide you through the using PaPM's Sample Content titled Climate Risk Scenario Analysis to assess the financial impacts of climate risks on various real estate assets and portfolios. The series will cover the methodology, data integration, and calculation steps of the sample content.

This blog series contains 3 parts:
• Part 1: Climate Risk Management Introduction.
• Part 2: Climate Risk Management - Methodology and Assumptions.
• Part 3: Financial KPIs and Calculation Steps.

In the first part of our blog series, we covered the impact of climate risks on economies through various channels, the importance of understanding the relationship between climate and the economy for informed financial decision-making, and the role of advanced methodologies and Integrated Assessment Models (IAMs) in managing climate risks.
Part 2 of this blog will focus on the methodology and assumptions that form the foundation of the climate risk scenario analysis, which quantifies the impact of climate change on real estate investments from an investor's perspective.

Climate Data for Analysis

Climate variable data is crucial for accurately assessing the risks and impacts of climate change on various sectors. This data provides essential insights into potential environmental changes, helping stakeholders make informed decisions to mitigate financial and operational risks. The climate data used for this analysis is sourced from Climate Analytics, a global climate science and policy institute recommended by the Network for Greening the Financial System (NGFS). While the sample content in this blog uses the data outlined below, PaPM is a flexible tool that can be adapted to work with the data you have available.

The following variables were considered in this model:

Variable Name	Definition
Mean Air Temperature	Air temperature refers to the temperature of air masses near the Earth's surface (2 metres above the ground in this case).
Precipitation	Precipitation is defined as the mass of water (both rainfall and snowfall) falling on the Earth's surface, per unit area and time. The data used for this variable have undergone a bias-adjustment procedure to correct for deviations between modeled and observed values over the time period where they overlap.
Pressure Adjusted	Atmospheric pressure quantifies the force that would be exerted by the weight of the column of air situated above a given location, per unit area.
Radiations Adjusted	Downwelling longwave radiation is defined as the downward energy flux in the form of infrared light that reaches the Earth's surface.
Relative Humidity	Relative humidity is defined as the ratio of water vapour in the air to the total amount that could be held at its current temperature (saturation level)
Specific Humidity	Specific humidity is defined as the mass of water vapour contained in each kg of air. Here we consider specific humidity at 2 metres above ground.
Snowfall	Snowfall is defined as the mass of water falling on the Earth's surface in the form of snow, per unit area and time.
Temperature Maximum Adjusted	Daily maximum air temperature is defined as the peak air temperature reached in a day, in this case at 2 meters above the ground.
Temperature Minimum Adjusted	Daily minimum air temperature is defined as the lowest air temperature reached in a day, in this case at 2 meters above the ground

Wind Speed Adjusted	Velocity of an air mass at 10 meters above ground.
One in Hundred Years Expected Damages from Tropical Cyclones	Damage expected to occur once in 100 years, assuming constant GDP size and distribution as of 2005.
Annual Expected Damages from River Floods	Annual expected damage from river floods, assuming constant GDP size and distribution as of 2005.
Annual Expected Damages from Tropical Cyclones	Annual expected damage from tropical cyclones, assuming constant GDP size and distribution as of 2005.
Annual Maximum River Flood Depth	Depth during the most severe flood of the year.
Land Fraction Annually Exposed to Heatwaves	Part of the population experiencing a heatwave annually, in a grid cell of 0.5° resolution.
Land Fraction Annually Exposed to River Floods	Fraction of land area flooded during the annual maximum event.
Land Fraction Annually Exposed to Wildfires	Annual aggregate of land area burnt at least once a year by wildfires.

Table 1 - Definition of the variables used in modeling 

Analysis Methodology

Variables are clustered around specific perils to streamline the assessment of climate-related risks by grouping related factors that collectively influence each hazard, such as
floods or heatwaves. This approach enhances the accuracy of risk evaluation and allows for more targeted mitigation strategies, ensuring that peril-based products are effectively tailored to address the most relevant risks.
• Climate Change Variables : Relative Humidity, Specific Humidity, Precipitation, Snowfall, Atmospheric Pressure, Radiations Adjusted, Wind Speed Adjusted, Mean Air Temperature, Daily Maximum Air Temperature, and Daily Minimum Air Temperature.
• Wind : Changes in Wind Speed and Pressure.
• Precipitation : Precipitation, Snowfall, Relative Humidity, and Specific Humidity.
• Cyclones : One in Hundred Years Expected Damages and Annual Expected Damages from Tropical Cyclones.
• Heatwaves : Annual Exposure to Heatwaves.
• Floods : Annual Expected Damages from Floods and Land Fraction Exposed to River Floods.
• Wildfires : Land Fraction Annually Exposed to Wildfires.

With the climatic variables effectively clustered around key perils, we can now move on to the critical step of calculating the potential financial impacts. This calculation will help translate the identified risks into actionable insights, enabling a clear understanding of how climate change could affect asset values and financial performance.

Climate Delta Risk Calculation Methodology

Climate Value at Risk (CVaR) is the primary financial metric used in this Sample Content to quantify the potential financial losses that could result from climate-related risks. It works by estimating the expected costs associated with climate impacts on an asset, portfolio, or company, and comparing these costs to the total value of the underlying asset.
To calculate CVaR, we first estimate costs due to chronic and acute risks for eight scenarios. Both types of risk are considered because chronic risks represent the long-term, gradual impacts of climate change on assets, while acute risks capture the immediate, severe consequences of extreme events. Together, they provide a comprehensive view of potential vulnerabilities, ensuring more robust financial and strategic planning.

Chronic Risks:

Chronic risks are associated with long-term impacts like asset damage and business interruption. The costs associated with these risks are modeled using:

Cost = number of exceedances × vulnerability × vulnerability reduction × optimal revenue

• Number of Exceedance: Number of days a climatic variable exceeds a threshold.
• Vulnerability: Impact of change in climatic variables on a geography.
• Vulnerability Reduction: Adaptation of the local society and economy to climate extremes over time.
• Optimal Revenue: Revenues of a firm in ideal conditions without climate change exposure.

For our analysis, we have calculated vulnerability for specific locations. Selecting location-based vulnerability is beneficial for the analysis as it incorporates the relative characteristics of local weather. For example, 35°C can be very hot for people of Northern Europe but could be a normal temperature in the equatorial region. The same holds for lower temperatures. We rely on classical statistics to calculate vulnerability, making it a unitless parameter. To ensure comparability and remove the scaling effect, we have further normalized it to a range of 0 to 1. This normalization highlights outliers across different regions.
Although all climatic variables are inter-related, one often emerges as the primary driver of climate change in a specific area. Therefore, variables carrying more variability are assigned higher weights, indicating their significance. For instance, if the temperature is changing significantly, it indicates that temperature change is the primary driver of climate change in that area, mathematically.

Acute Risks:

Cost = annual damage × asset value + annual share of revenue lost due to business interruption × annual revenue.

Acute risks are short-term, severe impacts of climate events, such as extreme storms or floods, that can cause immediate damage to assets and disrupt operations. They are crucial to consider because they represent sudden financial shocks that can significantly affect an organization's stability and resilience.

Data Integration- Linking Climate, Location, and Financial Data:

A common question asked is about the linkage of climate data, location data, and financial data, specifically how these diverse datasets are integrated to inform comprehensive real estate analysis and decision-making. Financial Data provides specific income information, linking financial performance metrics with real estate properties. Real Estate Data then utilizes this financial information. Climate Data contributes location-specific scenarios, detailing how climate changes might impact different areas. These location-specific climate scenarios are incorporated into the Real Estate Data, providing insights into potential environmental impacts on properties. The combined data allows for comprehensive analysis and forecasting, considering both economic and environmental factors in real estate decision-making.

Figure 1 -  Linkage of Financial, Real Estate, and Climate Data for Scenario Analysis

Climate Value at Risk (CVaR):

Climate Value at Risk is a key metric for assessing the potential financial impact of climate-related risks on assets. It estimates the proportion of an asset's value that could be lost due to climate change. It is defined as the percentage ratio of the expected climate costs (or income) to the total value of the underlying asset. Using appropriate rates, the expected physical climate costs are discounted and divided by the total market value to provide an overview of the risk. This indicator, proposed by Carbon Delta, is widely used for scenario analysis as it is simple and easy to understand.

CVaR(year) = Value of climate costs or profits(year) / Market value of enterprise(year).

Conclusion
The PaPM Sample Content titled Climate Risk Scenario Analysis leverages the described methodology to allow investors to quantify the impacts of climate change on real estate by analyzing chronic and acute risks. By understanding potential vulnerabilities, investors can make informed financial decisions to mitigate climate-related risks.
Stay tuned for Part III of our blogpost series about Climate Risk Management, where we will cover the dashboard and the results of analysis.

References

• TCFD Technical Supplement 2017: The Use of Scenario Analysis in Disclosure of Climate-Related Risks and Opportunities
• IPCC Special Report 2000: Emissions Scenarios
  Carleton, T.A. and Hsiang, S.M., 2016. Social and economic impacts of climate. Science, 353(6304), p. aad9837.
• UNEP FI Changing Course: A Comprehensive Investor Guide to Scenario-Based Methods for Climate Risk Assessment, in Response to the TCFD., 2019