Catastrophe Modelling: Introduction to MSMF-CatM Framework (Part A)
Publication Date: 2nd March 2025
Author: Vincent Su
Contributors: Lucas Enders, Daniel Giam
Executive Summary
In recent decades, the frequency and severity of natural disasters have surged, driven by climate change and a more interconnected global economy. This has highlighted the limitations of relying solely on historical data for risk prediction, giving rise to catastrophe modelling as a more comprehensive approach to risk estimation. The Monash Student Managed Fund (MSMF) has developed the MSMF-CatM framework, a catastrophe model designed to integrate advanced climate science, economic interdependencies, and probabilistic risk assessment. This framework is tailored specifically for student-managed funds, providing a robust tool for assessing and managing climate-related risks in investment portfolios.
Catastrophe modelling has traditionally been used by the insurance industry, but its application is expanding into other sectors, including asset management and sustainability. MSMF-CatM focuses on evaluating physical risks, such as extreme weather events, sea level rise, and prolonged droughts, while excluding transition risks like regulatory changes. By leveraging detailed asset-level data and advanced modelling techniques, the framework enables students to quantify the financial implications of climate hazards, including asset damage, business interruptions, and revenue losses. This makes MSMF-CatM a critical tool for improving portfolio resilience and informing investment decision-making in an era of increasing environmental uncertainty.
The framework aligns with broader scenario analysis processes, such as those outlined in the MSMF Scenario Analysis Plan, and integrates with open-source platforms like the OASIS Loss Modelling Framework (LMF). OASIS provides standardised data formats, modular architecture, and multi-peril capabilities, ensuring the accuracy and actionability of MSMF-CatM’s outputs. By combining these tools, MSMF-CatM equips the student-managed fund with the insights needed to navigate a complex risk landscape while offering students hands-on experience in cutting-edge risk modelling and sustainable investment practices. MSMF-CatM represents a significant step forward in catastrophe modelling, providing students with the tools and knowledge needed to manage risks effectively and make informed investment decisions in a rapidly changing world. This paper is the first in a two-part series introducing the MSMF-CatM framework. Part A provides an overview of climate catastrophe modelling and the MSMF-CatM framework, while Part B will offer a detailed implementation guide for student-managed funds. Together, these publications aim to advance the understanding and application of catastrophe modelling in the context of student-led investment initiatives.
MSMF Catastrophe Modeling Publication Series
This paper is part of a two part series that introduces the MSMF-CatM Framework.
1. Introduction
In recent decades, the frequency and severity of natural disasters have surged, driven by climate change and a more interconnected global economy. This has underscored the limitations of relying solely on historical data to predict future losses, giving rise to the field of catastrophe modelling—a more comprehensive approach to risk estimation. Among the emerging frameworks in this field is MSMF-CatM, a cutting-edge climate catastrophe model currently in development, designed to integrate advanced climate science, economic interdependencies, and probabilistic risk assessment to better anticipate and mitigate future disasters.
This report explores the key components and methodologies behind catastrophe modelling, with a particular focus on the innovative MSMF-CatM framework. It also examines the expanded application of such models in asset management, sustainability, and public policy for various stakeholders. Ultimately, this discussion aims to highlight how modern catastrophe modelling, including tools like MSMF-CatM, not only advances risk management practices but also informs strategic decision-making in an era of unprecedented environmental challenges.
1.1 Catastrophe Modelling
Catastrophe modelling has been used by the insurance industry for decades, as it was originally designed for insurers. And it is gaining unprecedented importance in today's world. Over the past 50 years, the frequency and severity of natural disasters—such as floods, earthquakes, and wildfires—have surged by a factor of five (World Meteorological Organisation, 2019). This trend has highlighted the vulnerabilities of infrastructure and communities, attracting increasing interest from other sectors (Fathom, 2024). The growing awareness of these risks has led to a heightened demand for robust tools, to assess and mitigate potential losses effectively.
Catastrophe modeling arose to overcome the limitations of using historical data alone for predicting natural disasters. The foundation for modern catastrophe modeling lies in the intricate interplay between mapping risk and measuring hazards, practices that have their roots in the 1800s. These early efforts were driven by the need to understand and mitigate the financial and human impacts of natural disasters, such as earthquakes, floods, and hurricanes. Over time, the field has evolved significantly, incorporating advancements in technology, data science, and risk management practices to create sophisticated models that are indispensable in today’s insurance, reinsurance, and disaster preparedness industries.
Major disasters, such as Hurricane Hugo in 1989 ($4 billion in losses) and Hurricane Andrew in 1992 ($15.5 billion in losses), underscored the inadequacy of traditional actuarial methods. These events spurred industry-wide adoption, transforming cat modelling into a standard risk management tool (American Academy of Actuaries, 2018; The Chartered Insurance Institute, 2009).
Since then, advancements in computer power and availability of historical data now allow for models to simulate stochastically, and adjust past data to reflect on current climate conditions (Clark, 2022). For instance, studies show that a 1°C increase in global temperatures correlates with a 2.5% rise in hurricane wind speeds, leading to an increase of 11% in insured losses. These changes also alter the shape of exceedance probability (EP) curves, with frequent, moderate losses rising faster than extreme losses. Clark points out that future models would aim to project risks under various climate scenarios, using data from reports such as the IPCC AR6 (IPCC, 2023). Therefore, faster updates are needed to keep pace with environmental changes, though (re)insurers often resist these due to significant shifts in loss projections.
2. Catastrophe Modeling Methodology
The model components typically described in catastrophe modeling are Event Set, Hazard, Vulnerability, and Financial Modules (Fathom, 2024; The Chartered Insurance Institute, 2009; American Academy of Actuaries, 2018).
The Event Set defines the simulated natural events or disasters (e.g., hurricanes, earthquakes) that the model uses for risk analysis. It encompasses the various scenarios, magnitudes, and frequencies, based on historical data and scientific research, to define the possible range of disasters that can happen.
The Hazard Module simulates the actual hazards, such as wind speed, rainfall, or seismic activity. It estimates the intensity of the hazard in relation to a given location (e.g., how severe the hurricane wind speeds are at Monash University Clayton Campus). The hazard module uses scientific models, such as meteorological data or geophysical models, to determine the likelihood and severity of different hazards.
The Vulnerability Module assesses how assets (e.g., buildings, infrastructure) react to hazard intensities and quantifies the extent of damage based on their characteristics. Using vulnerability functions or damage curves, it links hazard metrics (e.g., wind speed, ground shaking) to expected losses. This module relies on the Exposure Information, which provides \information such as the asset’s location, construction type, and value, to estimate the degree of physical damage caused by an event.
The Financial Module translates the damage estimates from the Vulnerability Module into economic terms. It incorporates the cost of damages to assets, alongside Policy Information (e.g., deductibles, limits, and coverage types), to calculate metrics like insured losses, out-of-pocket costs, and recovery costs. Additionally, it can generate critical risk metrics, such as loss exceedance probabilities (EP curves) and expected annual losses (AAL), which are essential for financial and risk planning.
While understanding the key components of catastrophe models is essential, it's equally important to address how these components are operationalised across different models. With diverse models available, each with unique inputs, outputs, and workflows, integrating and standardising these models can be challenging.
3. Implications to Industries
Catastrophe modelling plays a critical role in understanding and managing risks posed by acute and chronic physical risks. Historically, these models have been widely used by actuaries within the insurance and lending industries as a cornerstone of risk management. They enable insurers and financial institutions to quantify potential losses, set appropriate premiums, and ensure solvency in the face of catastrophic events. However, recent trends indicate a growing adoption of catastrophe modelling across various other sectors, including investment management and government planning, to assess the potential financial and operational impacts of natural disasters.
3.1. Insurance and Reinsurance
Catastrophe models have been extensively used by the (re)insurance industry for decades, providing critical insights into the financial impact of natural disasters. According to the American Academy of Actuaries (2018), metrics such as Average Annual Loss (AAL) and Probable Maximum Loss (PML) allow insurers to calculate appropriate premiums to its clients, reflecting the variability and uncertainty in potential losses on the insured asset based on established criterias. underwriting and risk selection by evaluating how new policies interact with existing portfolios, ensuring financial stability through metrics like PML that quantify the magnitude and likelihood of extreme losses. Furthermore, the Academy points out that by comparing AAL with and without specific mitigation features, catastrophe models identify cost-effective strategies to minimise risks. These models also streamline collaboration with reinsurers, enabling insurers to share financial burdens from extreme events, protect against insolvency, and maintain payments to policyholders.
3.2. Asset Management
Catastrophe modelling plays an essential role in managing the risks associated with investments in companies exposed to natural disasters. For asset managers, particularly in equity portfolios, catastrophe modelling offers several benefits.
Disasters can have a dramatic impact on stock prices, especially for firms with significant physical assets or supply chains vulnerable to natural disasters. A study from Edirisinghe & Xin Zhang (2011) on portfolio risk management shows that portfolios managed with a focus on catastrophic risk mitigation can yield significantly better out-of-sample performance—10 times or even better—than portfolios using traditional mean-variance frameworks.
Additionally, cat bonds have emerged as an attractive investment vehicle for diversifying portfolios and reducing risk. As Haffar & Le Fur (2022) shows, they are effective tools for reducing equity risk and providing uncorrelated returns to other market assets . This makes cat bonds valuable for investors looking to hedge natural disasters on their portfolios. Cat bond investors, investment banks, and bond rating agencies use cat modelling in the pricing and structuring of a cat bond (The Chartered Insurance Institute, 2009).
3.3. Operational Sustainability
Environmental risks, especially those related to climate change and natural disasters, are now a critical component of ESG evaluations. Enterprises worldwide face growing complexity in their operations, and extreme events can jeopardise their survival.
Cat models provide insights into the geographic and operational exposures of a company's assets to natural disasters, allowing investors to evaluate the potential disruptions to supply chains and business continuity (Komljenovic et al., 2016). With a framework that integrates risks from extreme events into overall risk management, companies can make more informed decisions regarding operational diversification.
3.4. Government Resource Allocation and Disaster Preparedness
Governments can use catastrophe modelling for strategic planning and response purposes, helping them prepare for disasters and protect vulnerable populations.
According to Grossi and Kunreuther (2005), by identifying high-risk regions, governments can direct emergency services and supplies where they are needed most, ensuring communities are better equipped to respond to crises. At the same time, it guides infrastructure investment, such as constructing more resilient buildings, upgrading flood defences, or improving power grids. These efforts reduce long-term economic damage from disasters.
In extreme cases, governments may encourage population relocating from extremely high-risk areas. This is particularly relevant when insurance coverage is reduced, and government intervention becomes necessary to protect lives and property. (Calandro et al., 2021)
4. Overview of Monash Student Managed Fund’s Proposed Catastrophe Risk Modeling Framework (MSMF-CatM)
4.1 Introduction to MSMF-CatM
MSMF-CatM is a climate catastrophe framework designed to provide a comprehensive and dynamic approach to assessing and managing the risks posed by climate-related disasters, tailored specifically for the MSMF Student Managed Fund. By integrating advanced climate science, economic interdependencies, and probabilistic risk assessment, MSMF-CatM aims to address the limitations of traditional catastrophe models, offering a more robust and forward-looking tool for risk estimation. The framework is particularly focused on evaluating physical risks associated with climate catastrophes, such as extreme weather events, sea level rise, and prolonged droughts, while excluding transition risks like regulatory changes or market shifts.
MSMF-CatM is built to support the student-managed fund in understanding and mitigating the financial and operational impacts of climate risks on its investment portfolio. By leveraging detailed asset-level data and advanced modelling techniques, the framework enables students to quantify the potential financial implications of climate hazards, including asset damage, business interruptions, and revenue losses. This makes MSMF-CatM a critical tool for improving portfolio resilience, informing investment decision-making, and enhancing risk management practices in an era of increasing environmental uncertainty.
4.2 Scenario Assessment Process
For the majority of organisations, climate risk assessments fall under the scenario analysis phase, a critical component of understanding and preparing for the potential impacts of climate change. Scenario analysis enables businesses to evaluate how different climate-related pathways could affect their operations, supply chains, and financial performance and therefore gives rise to asset-level environmental risk assessments.
The MSMF Scenario Analysis Plan outlines a comprehensive approach to conducting scenario analysis, which includes the following key steps:
1) Adopt Reference Scenarios:
These scenarios provide a consistent and scientifically robust basis for assessing climate risks by adopting reference scenarios that align with industry best practices and global standards. Commonly used reference scenarios include:
Intergovernmental Panel on Climate Change (IPCC) Scenarios
Network for Greening the Financial System (NGFS) Scenarios:
International Energy Agency (IEA) Scenarios
2) APRA Assessment:
The Australian Prudential Regulation Authority (APRA) has emphasised the importance of conducting Climate Vulnerability Assessments as part of scenario analysis. This involves evaluating how climate-related risks could affect an organisation’s financial stability, operational resilience, and long-term viability. APRA’s guidance encourages organisations to consider both transition risks (e.g., policy changes, technological shifts) and physical risks (e.g., extreme weather events, sea level rise) in their assessments.
3) Assess Climate Risks:
A critical step in scenario analysis is the identification and assessment of specific climate risks. These risks can be broadly categorised into:
Physical Risks: These include acute risks, such as floods, bushfires, and cyclones, as well as chronic risks, such as sea level rise, extreme temperatures, and prolonged droughts. In the Australian context, physical risks are particularly relevant due to the country’s exposure to extreme weather events and changing climate patterns.
Transition Risks: These arise from the shift towards a low-carbon economy and include regulatory changes, technological advancements, market shifts, and reputational impacts.
4) Model Financial Implications and perform Climate Catastrophe Modeling
4.3 MSMF-CatM Process
For the purpose of assessing climate catastrophe risk towards a portfolio of assets, the focus is particularly relevant to parts (3) and (4) of the Scenario Assessment Process. Specifically, part (3) involves the assessment of physical risks associated with climate catastrophes, such as extreme weather events, sea level rise, and prolonged droughts, while excluding transition risks (e.g., regulatory changes, market shifts, or technological disruptions). Part (4) then builds on this by modelling the financial implications of these physical risks on the portfolio, quantifying potential impacts on asset values, operational costs, revenue streams, and overall financial stability.
In the context of part (3), the assessment of physical risks requires a detailed analysis of the portfolio’s exposure to climate-related hazards. This includes identifying assets located in high-risk areas, such as floodplains, bushfire-prone regions, or coastal zones vulnerable to sea level rise. For example, in Australia, assets in northern regions may face heightened risks from cyclones, while those in southern areas could be more susceptible to heatwaves and droughts.
Moving to part (4), the financial modelling of these risks involves translating the physical risk assessment into quantifiable financial impacts. This includes estimating potential costs related to asset damage, business interruptions, increased insurance premiums, and recovery expenses. For instance, a property located in a flood-prone area may face higher repair costs and reduced market value, while a supply chain dependent on vulnerable infrastructure could experience disruptions leading to revenue losses and negative market reaction.
Based on the asset being analysed, a detailed analysis of their operations and supply chains at an asset level are used as input and limits for the catastrophe model. This identification of input involves:
Identifying Critical Assets: Mapping out key assets, including physical infrastructure, facilities (fixed and long-term asset), and supply chain nodes, that are vulnerable to climate risks based on location data.
Assessing Exposure: Evaluating the geographic and operational exposure of these assets to specific climate hazards, such as flooding, extreme heat, or bushfires by region.
Once input values have been determined:
Using a climate catastrophe modelling tool to quantify impacts: Estimates the potential financial impacts of climate risks on each asset, considering factors such as replacement costs, downtime, and revenue losses to which is translated as an environmental risk factor.
4.4 Catastrophe Modeling Platform: OASIS LMF
The OASIS Loss Modelling Framework has emerged as a pivotal tool in the field of catastrophe (cat) modelling, addressing the growing need for standardisation, transparency, and interoperability in risk assessment processes. As an open-source platform, OASIS provides a unified environment for integrating multiple catastrophe models, enabling users to compare and combine results from different sources seamlessly. This is particularly valuable in an industry where diverse models, data formats, and methodologies can create challenges for consistency and accuracy. Backed by the global (re)insurance community, OASIS has become a cornerstone for advancing cat modelling practices, fostering collaboration, and enhancing the robustness of risk assessments.
Strengths of OASIS LMF
1. Standardisation of Data Formats and Processes
One of the key technical strengths of OASIS is its ability to standardise data inputs and modelling processes. The framework supports a common data format, which ensures that users can integrate various models without the need for extensive data manipulation or conversion. This standardisation reduces errors, improves efficiency, and facilitates the sharing of data and results across organisations.
2. Modular Architecture
OASIS is designed with a modular architecture, allowing users to plug in different components, such as hazard models, vulnerability functions, and financial modules. This flexibility enables organisations to tailor the framework to their specific needs while maintaining consistency in the overall modelling process. For example, users can incorporate region-specific hazard models (e.g., cyclones in Australia or earthquakes in Japan) alongside global models, ensuring that the outputs are relevant to their portfolios.
3. Support for Multiple Perils
The framework supports a wide range of perils, including floods, earthquakes, hurricanes, and wildfires. This multi-peril capability is essential for comprehensive risk assessments, particularly in regions like Australia, where properties may be exposed to multiple hazards simultaneously. By integrating models for different perils into a single platform, OASIS enables users to assess cumulative risks and interdependencies between events.
4. Probabilistic and Deterministic Modelling
OASIS facilitates both probabilistic and deterministic modelling approaches. Probabilistic models generate a range of potential outcomes based on historical data and statistical distributions, providing insights into the likelihood and severity of future events. Deterministic models, on the other hand, focus on specific scenarios, such as a 1-in-100-year flood event. This dual capability allows users to explore both broad risk landscapes and specific scenarios of interest.
MSMF-CatM Use Case
4.6 Limitations of MSMF-CatM
While the MSMF-CatM framework offers significant advancements in climate catastrophe modelling, it is not without its limitations. These challenges must be carefully considered to ensure effective implementation and utilisation of the framework.
1. Operational Cost:
Based on MSMF-CatM’s cost assessment and discussions with Australian-based data model providers, the base cost of a data model license can range from five to six figures, depending on the number of perils covered. This high cost may pose a barrier for student-managed funds with limited budgets, potentially restricting access to comprehensive risk assessment tools.
2. Complexity of Implementation:
The framework’s sophisticated design, while beneficial for accurate risk modelling, also introduces complexity in implementation. Integrating MSMF-CatM with existing systems, ensuring data compatibility, and configuring the model to specific portfolio requirements can be technically challenging and time-consuming.
3. Training Requirements:
Effective use of MSMF-CatM requires a solid understanding of catastrophe modelling principles, data science, and financial risk analysis. This necessitates significant training for students and fund managers, which may require additional resources and time investment which is a steep learning curve for most students.
-
MSMF acknowledges and pays respect to elders past, present and emerging peoples of the Kulin nation, the traditional owners of the land on which MSMF operates.
We are grateful to Prof. Deep Kapur and the team at Monash Centre for Financial Studies (MCFS) for their unwavering support on a student-led initiative and the delivery of our agenda in multiple ways and acknowledge their contributions to each and all releases. We also remain deeply grateful to the Faculty of Banking and Finance at Monash Business School for continuous support in facilitation of MSMF and our agenda.
We would like to thank Daniel Giam - MSMF Sustainability Chairman, who has been the main key topic advisor, critical supporter and contributor of this paper and the initiatives at MSMF ESG Portfolio Team.
In addition, we would also like to thank Matthew Donovan and Ben Hayes from the OASIS LMF team for their advice and knowledge on the topic of the research.
-
This material is a product of Monash Student Managed Fund (MSMF) and is provided to you solely for general information purposes. I understand that the information in these documents is NOT financial advice. Before making an investment decision to acquire shares, you should consider, preferably with the assistance of a financial or other professional adviser, whether an investment is appropriate in light of your own personal circumstances. If you can, you should obtain a copy of the Information Memorandum of the company that you are seeking to invest in, and consider their risks and disclosures. Subject to the Australian Consumer Law, Corporations Act, the ASIC Act, and any other relevant law, MSMF does not accept any responsibility for any loss to any person incurred as a result of reliance on the information, including any negligent errors or omissions. This information is strictly the personal opinion of an MSMF member and does not represent the views of MSMF. This information constitutes factual information that is objectively ascertainable such that the truth or accuracy of which cannot reasonably be questioned. MSMF does not intend to advertise any stock or financial product whatsoever. Past performance is not a reliable indicator of future performance. Past asset allocation and gearing levels may not be reliable indicators of future asset allocation and gearing levels. Performance data is just an estimation based on public market data and may not be a true reflection of actual fund performanceon text goes here
-
Actuaries Institute. (2015, July). Technical paper: The use of catastrophe model results by actuaries. https://www.actuaries.asn.au/Library/Standards/GeneralInsurance/2023/TP%20The%20Use%20of%20Catastrophe%20Model%20Results%20by%20Actuaries%20Jul%202015%20(reclassified%20Feb%202023).pdf
American Academy of Actuaries. (2018, July). Uses of catastrophe model output. https://www.actuary.org/sites/default/files/files/publications/Catastrophe_Modeling_Monograph_07.25.2018.pdf
Calandro, J., Hall, G., & Zheng, S. (2021). Property natural catastrophe risk analysis in an era of climate change: Some financial services, investment and public policy implications. The Journal of Impact and ESG Investing, 19–32. https://doi.org//10.3905/jesg.2021.1.024
Clark, K. (2022, May 16). The evolution of catastrophe modeling since hurricane andrew. Insurance Journal. https://www.insurancejournal.com/magazines/mag-features/2022/05/16/667461.htm
Edirisinghe, N. C. P., & Zhang, X. (2011). Portfolio risk management: Market neutrality, catastrophic risk, and fundamental strength. In G. Nota (Ed.), Risk Management Trends (pp. 109–128). IntechOpen. https://doi.org/10.5772/16270
Grossi, P., & Kunreuther, H. (2005). Catastrophe Modeling: A New Approach to Managing Risk. https://doi.org/10.1007/b100669
Haffar, A., & Le Fur, É. (2022). Dependence structure of CAT bonds and portfolio diversification: a copula-GARCH approach. Journal of Asset Management, 23(4), 297–309. https://doi.org/10.1057/s41260-022-00271-3
IPCC. (2023). Sixth Assessment Report — IPCC. Ipcc.ch; IPCC. https://www.ipcc.ch/assessment-report/ar6/
Jones, M. (2024, October 1). Introduction to catastrophe modeling. Fathom. https://www.fathom.global/insight/introduction-to-catastrophe-modeling
Komljenovic, D., Gaha, M., Abdul-Nour, G., Langheit, C., & Bourgeois, M. (2016). Risks of extreme and rare events in asset management. Safety Science, 88, 129–145. https://doi.org/10.1016/j.ssci.2016.05.004
The Chartered Insurance Institute. (2009). Climate change and its Implications for catastrophe modelling. https://www.cii.co.uk/media/4043795/ch4_catastrophe_modelling.pdf
World Meteorological Organisation. (2024). Global reported natural disasters by type, 1970 to 2024. Our World In Data. https://ourworldindata.org/grapher/natural-disasters-by-type
