Program and Abstracts
The workshop starts on Wednesday at 9:00 a.m.
and ends on Friday approximately at 1.00 p.m.
Details t.b.a. very soon
Overview (Draft)
Wednesday, September 4, 2024
- Two plenary talks
- Contributed talks
- Joint lunch
- Contributed talks
- Workshop "AI in Climate Change"
Thursday, September 5, 2024
- One plenary talk
- Panel Discussion
- Business Lunch
- One plenary talk
- Contributed talks
Friday, September 6, 2024, morning:
- Two plenary talks
- Contributed talks
Plenary speakers
Invited plenary talk: date/time t.b.a.
M. Carmen Boado-Penas (Heriot-Watt University, UK)
Climate emergency: understanding the risks
In this talk, we provide a comprehensive overview of climate risks and their consequences for insurers, covering both life and non-life sectors, along with the solutions insurers and governments are offering. We will also discuss how insurance companies can promote greener practices in the market and incentivise adaptation measures. Special attention will be given to climate scenarios and the potential influence of tipping points on long-term projections.
The second part of the presentation will focus on how climate-related risks affect the three pillars of pension systems. We will discus the impact of climate risks on social security programs, which are typically based on a pay-as-you-go basis. Additionally, we will examine how various climate policies affecting investment strategies can influence the accumulated capital at retirement in funding schemes, highlighting the different impacts on various birth cohorts.
About
Carmen Boado-Penas is a professor of Actuarial Sciences at Heriot-Watt University in Edinburgh. For many years, she was the Actuarial Mathematics BSc programme director at the University of Liverpool, UK. She holds a PhD in Actuarial Science (Doctor Europeus) from the University of Valencia, Spain, and an MSc in Quantitative Finance. She was also awarded a prize by the Foundation of Spanish Savings Banks for her PhD "Instruments for improving the equity, transparency and sustainability of pay-as-you-go pension systems".
She has published more than 40 peer-reviewed papers on pension finance in prestigious international journals and has cooperated on various projects related to pension systems at the Swedish Social Insurance Agency in Stockholm and at the Spanish Ministry of Labour and Immigration. In 2012, she worked as head of research on a project for the Spanish Ministry of Labour and Immigration, the aim of which was to evaluate the redistributive effects of the pension system reform in Spain. In 2020, she received the BBVA Longevia award (first prize on the economics section) to support pension research.
Her research interests are focused on life insurance, automatic balance mechanisms for state pensions, mixed pension schemes, redistribution, and, most recently, the impact of climate change on retirement.
Invited plenary talk: date/time t.b.a.
Jose Garrido (Concordia University Montreal, Canada)
Actuarial climate indexes: international comparative study and insurance applications
Climate risks are increasingly affecting the frequency and the severity of claims in different insurance branches. In order to help insurance companies predict and manage climate risks, North American actuaries defined the Actuaries Climate IndexTM (ACI), that combines information from several important weather variables in the historical records of United States and Canada. The ACI provides a factual and objective climate risk measure for North America; it shows a significantly increasing trend over recent years.
There is a need to test if a similar tool can measure the impact of climate change in other parts of the planet, and if the change is similar or not.
Despite the observed global nature of climate change, different regions and countries can be affected in different ways. In this presentation we make use of the same ACI methodology to calculate actuarial climate indices with climate data from France, Spain and Portugal and compare the results obtained to those of Canada and the US. An illustrative example uses the French index to design and price a parametric insurance product.
About
Dr. José Garrido is Distinguished Professor Emeriti at the Department of Mathematics and Statistics at Concordia University, in Montreal, Canada.
After working as an actuarial analyst for Towers Watson in Montreal (then called TPF&C), Prof. Garrido received a Masters from Université Catholique de Louvain, in Belgium, and his PhD in 1987 from the Department of Statistics and Actuarial Sciences at the University of Waterloo, Canada. He is an Associate of the Society of Actuaries (SOA) and was an Associate of the Canadian Institute of Actuaries (CIA), from 2012 to 2021.
His research interests are in Risk Theory, Loss Models, Insurance Statistics, Credibility Theory, Risk Management and Credit Risk, Machine Learning in Insurance, Predictive Modelling and Robust Methods. Prof. Garrido has published more than 50 articles in international refereed journals and conference proceedings. He is Associate Editor of several journals, including Insurance: Mathematics and Economics and the North American Actuarial Journal, as well as an Editor of the European Actuarial Journal and of the open access journal Risks.
Invited plenary talk: date/time t.b.a.
Robert Holzmann (Governor OeNB, Austria)
Damages by extreme weather and the case of insurance
The number and scope of extreme weather conditions is increasing in Europe and elsewhere which stimulates the discussion of how best to pay for damage and quick repair. The presentation explores the role and scope of insurance against the background why central banks are interested in the topic and EU financial market institutions, including the European Central Bank, participate in the discourse. While insurance is expected to play a major role in mitigating extreme weather outcomes, the role of private and – perhaps – mandated insurance may be limited, innovative models of risk management – such impact underwriting and cat bonds – can contribute to reduce the insurance protection gap. More innovations, including in public-private partnership may be needed.
About
Robert Holzmann is an Austrian economist, Governor of the Austrian Central Bank und member of the Governing Council of the European Central Bank (since 2019), and elected member of the Austrian Academy of Sciences (since 2014). He holds honorary positions at the South-Western University in Economics and Finance, Chengdu, University of Malaya, Kuala Lumpur and University of New South Wales, Sydney. Before his return to academia in 2011, he held various positions at the World Bank including for 12 years Sector Director and acting Senior Vice President. Before joining the World Bank in 1997 he was an academic in Austria and Germany, and senior economist at the IMF and OECD. He has published 39 books and over 200 articles on financial, fiscal and social policy issues. He has travelled to over 90 countries in the world.
Invited plenary talk: date/time t.b.a.
Chris Kenyon (MUFG Securities EMEA plc and University College London, UK)
Climate-economy scenario/probability construction for financial markets
Climate-economy models can be used to extend traded markets, but their integration into pricing has been limited. Limiting factors include: size of policy-level models, large fixed timesteps in simple models, lack of methods to generate probabilities when only scenarios are available, and lack of simple methods to generate financial market variables. Using DICE2023 and SSP-RCP scenarios as examples we show how these limitations can be addressed and overcome. Some limitations can be addressed by simple re-development techniques, other limitations can be addressed by a Bayesian mixture state transition model (BMST) generating probability overlays on scenarios that we introduce, and further simple techniques like the Taylor rule for central bank actions. These techniques enable construction of tractable models providing useful sets of market variables to extend financial markets from liquid traded horizons to longer term. Longer term is beyond 3-5 years as is the limit of traded carbon markets, and beyond 5-10 years as the limit of traded CDS markets.
About
Dr. Chris Kenyon is global head of quant innovation at MUFG, and global head of XVA quant modelling for MUFG. Chris is also honorary associate professor of mathematics at University College London, where he teaches on mathematical climate finance. Previously, Chris was head of XVA quant research at Lloyds Banking Group, worked at Credit Suisse, and at Depfa Bank plc he was the post-crisis head of structured credit valuation, after working on inflation-rates hybrids introducing new smile models. Chris formalized KVA and MVA with Andrew Green.
More recently he introduced a climate change valuation adjustment (CCVA), the carbon equivalence principle (CEP), and a CO2-equivalent Scope 3 valuation adjustment (CO2eVA). Chris has degrees from University of Texas - Austin, and Cambridge University. He is an author of the open source software QuantLib, holds 10 US patents, and published 21 papers in the peer-reviewed Cutting Edge section of Risk magazine.
Invited plenary talk: date/time t.b.a.
Ralf Korn (TU Kaiserslautern and Fraunhofer ITWM, Germany)
Optimal investment with sustainable assets – aspects for life insurers
Structural and long term investments are key ingredients for transforming our environment and our society to a sustainable state. As life insurers traditionally invest into assets with a long duration, they can enter appropriate positions that correspond to participations in e.g. solar and/or wind parks in their cover fund. We will therefore set up a control theoretic framework for a life insurer to deal with a demand for sustainable assets.
Further, we consider the problem of optimal taxation that should automatically lead to volontary investment into sustainable assets and additional conceptual and modelling aspects.
About
Prof. Dr. Ralf Korn was born on 20. Mai 1963 in Eltville/Rhein (Germany). He studied Mathematics and Business Administration at the Johannes Gutenberg-University Mainz from 1983 to 1989, got his PhD in 1993 and his habilitation in 1997, both in Mathematics. Since 1999, Prof. Dr. Korn holds a professorship (associate professor until 2001, full professor since 2001) at the Dept. Mathematics of RPTU Kaiserslautern-Landau (until 2022, TU Kaiserslautern).
He has been Dean of the Mathematics Department for six years, was a member of the University Senate for the same time and a long term member of the commission for system accreditation and of the steering commitee for science. At the Fraunhofer Institute for Industrial Mathematics (ITWM) in Kaiserslautern, he founded the Department of Financial Mathematics in 1999, was heading the department for 10 years and still is a scientific consultant. Since 2010, he is the CEO of the European Institute for Quality Management of Financial Products and Methods (EI-QFM) in Kaiserslautern.
He has published 10 monographies and more than 100 refereed scientific articles. Further, he successfully supervised 60 PhD students.
Since 2003 (with a break of two years), Prof. Korn is a member of the board of the Deutsche Gesellschaft für Versicherungs- und Finanzmathematik (DGVFM).
Invited plenary talk: date/time t.b.a.
Frank Schiller (Munich RE, Germany)
Climate Change might not only impact the insurance risks – social aspects need also to be considered and the whole business model needs a thorough review
There may be conflicts of interest at various points within the ESG (environmental, social, governance) sustainability goals but also between social goals and the economic viability of the insurance solutions.
It is important for companies to evaluate and weigh up which options exist and how these affect the various target dimensions (particularly relevant are ecological, social and economic). With the advice of actuaries, the management can make fact-based, well-informed and balanced decisions that also take the interests of the policy-holders into account appropriately.
In the context of designing solutions for society as a whole, actuaries can also advise political decision-makers and evaluate the various solution options on the basis of facts.
We provide some examples for the pension, life, health and general insurance sectors, which are currently facing specific challenges in Europe and have a material impact on the business model of insurers.
- How can environmental risk for property insurance be appropriately assessed in order to ensure affordable premiums and a stable collective but also incentivise the owners to increase the resilience against climate change on the one hand, but on the other hand not to make access to insurance cover unjustifiably more expensive or restrict it?
- How can individual personal risk including social aspects be appropriately assessed in life insurance in order to ensure a stable collective on the one hand, but on the other hand not to unjustifiably marginalise individuals and make access to insurance cover unjustifiably more expensive or restrict it?
- How can pension provision be modernised to reflect ESG targets and prevent poverty in old age despite volatile capital markets and demographic change?
Solutions will often depend on the social legislation of the individual country and can therefore only be meaningfully discussed on a market-specific basis. We will discuss what needs to be considered for each of the three examples and how business models of insurers might need to be adapted accordingly.
About
Dr. Frank Schiller has been working at Munich Re as Chief Actuary in life and health reinsurance since 2015 and is responsible for the markets in Europe and Middle East. In this position he is in charge of pricing and business development. Before that he was Chief Risk Officer at Swiss Life from 2011 to 2015, first for the Swiss and later for the German market. From 2001 to 2011 he worked in various actuarial, risk management and product development positions at ERGO and Munich Re.
As an active member in the German Actuarial Association (DAV) Dr. Frank Schiller is member of the executive board and Chair of the Enterprise Risk Management Committee. In the Actuarial Association of Europa (AAE) he is member of the board and was Vice Chair of the Risk Management Committee and Chair of the Sustainability and Climate related Risks Working Group before.
He was born 21 November 1972, studied Mathematics and finished his Ph.D. in 2002. Since 2004 Frank is a certified Actuary (DAV) and since 2013 Certified Enterprise Risk Actuary (CERA).
Contributed talks
Contributed talk: date/time t.b.a.
Aleksandar Arandjelović (WU Wien, Austria)
Solving recursive stochastic climate-economy models using a deep least-squares Monte Carlo method
Uncertainty is critical for modelling climate change policy. Stochastic versions of recursive integrated climate-economy assessment models become more commonly used for studying and quantifying the policy decisions under uncertainty. These are solved as a dynamic programming problem using deterministic grid methods that become computationally infeasible for more than few stochastic shocks added to the model variables and simulation methods are needed. The least-squares Monte Carlo method (LSMC) has become a popular simulation method for solving optimal stochastic control problems in quantitative finance over the last decade. Using the stochastic DICE model as an illustration, we present the application of the least-squares Monte Carlo method for solving the model with parameter uncertainties in equilibrium temperature sensitivity, the damage function and carbon cycle coefficients, and process uncertainties in productivity growth and the rate of decarbonization. Typically, polynomial regression with respect to state and control variables is used in LSMC to approximate corresponding conditional expectations that leads to too many covariates in the case of polynomials of the 3rd and higher orders. To overcome this problem, in this study we use deep neural network approximations instead of polynomial regression. Solving the models using LSMC requires significant experience with the method and we discuss the pros and cons of the method.
Joint work with Pavel V. Shevchenko, Daisuke Murakami, Tomoko Matsui, and Tor A. Myrvoll.
Contributed talk: date/time t.b.a.
Michaela Bundschuh (d-fine Austria GmbH, Austria)
Biodiversity in a nutshell – requirements, challenges and solutions
Financial institutions play a crucial role in preventing the ongoing environmental crisis and especially the biodiversity loss. The role of banks, insurers, asset managers and investors is becoming increasingly relevant due to pressures such as strengthening liability regimes, intensifying scrutiny of supply chain practices and shifting consumer preferences.
The EU has launched the Biodiversity Strategy for 2030 as an important component of the EU Green. Legal measures are currently being developed, including a directive on soil monitoring. Besides, biodiversity is a key component of the CSRD and the Taxonomy Regulation.
At a global level, an international treaty was concluded with the Kunming-Montreal Biodiversity Framework. The framework is primarily concerned with global biodiversity and contains four central goals and 23 sub-goals, which the signatory states have signed.
The activities and investments of the financial sector are affected by biodiversity loss and have an impact on biodiversity. This means, biodiversity loss could lead to here types of risks for insurances, asset manager and other financial institutions: physical risk, transition risk and reputational risk.
By allocating capital the financial industry has the chance to finance a positive change and work against the loss of biodiversity. The first step of taking action is to assess and disclose the natural impact and the dependencies of their activities and to start the integration in the decision-making and target setting process.
Contributed talk: date/time t.b.a.
Roberto Carcache Flores (Vitalis, Portugal)
Mitigating flood risk with CAT bonds: a New Orleans case study
Floods cause billions of dollars in losses each year and involve different natural disasters like hurricanes, mudslides, and prolonged rainfall.
Data from the Federal Emergency Management Agency (FEMA) indicates that flood losses in the U.S. have increased in severity and frequency over the years, stemming from climate change and a greater number of extreme weather events. This paper presents a case study of how catastrophe (CAT) bonds can be used to manage the financial risk of flooding in Orleans Parish, an area with high exposure to flooding, according to data retrieved from FEMA.
We present a multi-period model for the valuation of a CAT bond with an indemnity trigger, that aims to provide coverage for extreme flood losses.
This valuation method incorporates Extreme Value Theory to model flood losses. The price of the CAT bond is obtained through Monte Carlo simulations with stochastic rates. Different assumptions are then tested, to show the sensitivity of the CAT bond’s price to the coverage provided and the model parameters.
Joint work with Abraham Hernández-Pacheco.
Contributed talk: date/time t.b.a.
Sascha Desmettre (Johannes Kepler University Linz, Austria)
Comparing factor models for different periods of the electricity spot price market
Due to major shifts in European energy supply, a structural change can be observed in Austrian electricity spot price data starting from the second quarter of the year 2021 onward. In this work we study the performance of two different factor models for the electricity spot price in three different time periods. To this end, we consider three samples of EEX data for the Austrian base load electricity spot price, one from the pre-crises from 2018 to 2021, the second from the time of the crisis from 2021 to 2023 and the whole data from 2018 to 2023. For each of these samples, we investigate the fit of a classical 3-factor model with a Gaussian base signal and one positive and one negative jump signal and compare it with a 4-factor model to assess the effect of adding a second Gaussian base signal to the model. For the calibration of the models we develop a tailor-made Markov Chain Monte Carlo method based on Gibbs sampling. To evaluate the model adequacy, we provide simulations of the spot price as well as a posterior predictive check for the 3- and the 4-factor model. We find that the 4-factor model outperforms the 3-factor model in times of non-crises. In times of crises, the second Gaussian base signal does not lead to a better fit of the model. To the best of our knowledge, this is the first study regarding stochastic electricity spot price models in this new market environment.
Joint work with Florian Aichinger and Christian Laudagé.
Contributed talk: date/time t.b.a.
Michelle Dong (Australian National University, Australia)
Actuarial perspectives on climate-related risks
Emerging climate risks around the world are bringing about pressing challenges for governments, industry, and actuaries working in insurance, as they face unprecedented complexities in qualifying climate impacts from physical and transition risks. These risks can be long-term (and sometimes irreversible) in nature, such as sea level and average temperature rises from climate-related physical risks, changes in the economy as countries transition and adapt to mitigate climate related risks; and could vary in their severity over time, as is the case with climate-related disasters and events. Based on a series of interviews conducted with leading experts in industry and academia, this talk explores the diverse perspectives on the issue emerging climate risks, and the role actuaries can play in helping to shape the industry’s understanding of and subsequent response to climate-related risks: ranging from advisor, to educator, to interested third party. We discuss the collective growing understanding of climate-related physical and transition risk impacts through the lens of climate modelling and scenario testing. The evolving landscape around risk management in the context of regulatory changes is also discussed, including requirements for companies to better understand and disclose their exposures to climate related risks, with a focus on insurers. We conclude by considering these perspectives in the context of our ongoing research into climate change impacts on mortality and scenario analysis.
Joint work with Aaron Bruhn and Francis K.C. Hui.
Contributed talk: date/time t.b.a.
John Ery (Signal Iduna Reinsurance, Switzerland)
A parametric-modeled loss type CAT bond
We present a new type of catastrophe (CAT) bond addressing the known trade-off between moral hazard and basis risk. As our main contributions we propose a trigger mechanism which is entirely transparent and simpler to evaluate compared to indemnity modeling techniques, and a methodology to compute the probability of exceeding the bond’s attachment level, which ultimately appears in the pricing formula. This study is particularly relevant for insurers and public authorities in a world where natural disasters are occurring with increasing frequency and severity due to climate change, but also for players willing to enter the CAT bond market and who have considered the lack of transparency of this asset class to be a significant obstacle. One foreseeable advantage would be the possibility to enable risk transfer when information on losses is not available or trusted. Moreover, e.g., for World Bank sponsored transactions, it is of utmost importance that risk transfer products pay out very fast.
According to the UNDP, insurance-linked securities can securitize resilience and provide financial aid to the countries most vulnerable to climate shocks due to their unpreparedness and low insurance protection. Our structuring mechanism could encourage more issuance of CAT bonds in countries which have been traditionally underserved by the insurance market and in this way contribute towards the reduction of the protection gap, which corresponds to the difference between economic and insured losses. This type of insurance coverage could play an important role among wider public-private partnerships which are urgently needed to boost societal resilience in a world with increasing uncertainty. Indeed, benefiting from a rapid payout after an event provides citizens with a psychological and financial safety net to face such uncertainties. Such triggers could thus benefit not only private insurer portfolios but also governmental or supra-national institutions with assets exposed to natural catastrophes. Our trigger is derived from a cost random field which separates the physical hazard, a vulnerability function, and the exposure. This allows the trigger to take a flexible form between parametric and modeled loss in case industry exposure is considered. We present a case study based on historical windstorm events in Germany. This approach can naturally be extended to any region subject to the availability of physical hazard and exposure data. On top of measuring the basis risk associated with our trigger, we perform a full model assessment and discuss numerical results.
Joint work with Erwan Koch.
Contributed talk: date/time t.b.a.
Anselm Fleischmann (BELTIOS, Austria)
How to deal with spurious accuracy in actuarial calculations targeting climate change
The sociology of professions explains, that societies entrust so-defined professions with the solution of problems that are per se unsolvable. This holds true for the classic professions like doctors and lawyers, neither of which are liable for the success of their services.
The actuarial profession arriving late at the scene and by using advanced mathematical models supplanting earlier types or foretellers, is not exempted. Actuaries make professional predictions by following appropriate methods and adhering to a code of conduct, but they are not liable, that the numbers predicted will become empirical reality.
Spurious accuracy in actuarial calculations refers to the presentation or reliance on overly precise results that give a false sense of certainty.
This may have disconcerting consequences:
- Presenting results with too many decimal places can suggest a level of accuracy that is not justified by the underlying data or models.
- Stakeholders may develop unwarranted (over)confidence in the results.
- Spurious accuracy can cause actuaries and decision-makers to overlook inherent uncertainty and variability in the data.
With respect to actuarial best estimates used in supervisory reporting and accounting systems tied to fair values (such as IFRS 17), it is proposed to differentiate reported results according to the distance, how far ahead estimated cashflows are in the future and present each of these results with adequate decimal places.
Mathematical measures assessing the instability of results from initial conditions are studied to derive a heuristic for the appropriate the number of decimal places in disclosed actuarial results.
Contributed talk: date/time t.b.a.
Ivan Alexis Fonseca Diaz (University of Lausanne, Switzerland)
An intergenerational RICE model approach to assess the impact of climate change on social security systems under shared socioeconomic pathways
Climate change is increasingly drawing the attention of research communities due to its long-term effects on natural and social systems worldwide. The channels through which climate change can affect societies are diverse. While physical effects can be translated into economic and life losses, the trends of societies to adapt and mitigate such physical effects can also have broader socioeconomic implications. Economic and financial instabilities, increasing poverty and inequality, and the collapse of health systems have been the most relevant effects studied in academic literature. Since vulnerable populations, such as the elderly and low-income households, heavily rely on social security systems, it is critical to comprehensively analyze future socioeconomic pathways to anticipate potential deficiencies of social security coverage and financing mechanisms. However, despite its notable relevance, there is a noticeable lack of research on how climate change can affect social security systems.
This research aims to be the first approach in developing an Integrated Assessment Model (IAM) scenario analysis to derive quantitative plausible pathways on the effects of climate change on social security systems. To this end, we extend the Regional Integrated model of Climate and the Economy (RICE) to allow for intergenerational consumption and saving dynamics, thereby depicting a consistent model representation of social security systems. Using the existing framework on alternative socio-economic development narratives described by the Shared Socioeconomic Pathways (SSPs), as well as mortality projections derived from the International Futures (IFs) IAM, we extend the SSP baseline scenario representation by consistently determining consumption, savings, contributions, and retirement benefit trajectories of future generations. Our representation of social security systems for different regions around the world, allows us to calculate optimal and actuarially fair consumption and saving allocations for active and retired populations. We use these reference paths to globally assess the adequacy and equity of contribution rates, as well as their pertinence to ensure the sustainability of social security systems under SSPs scenarios.
Joint work with Séverine Arnold.
Contributed talk: date/time t.b.a.
Marcello Galeotti (University of Florence, Italy)
A quantitative model of green transition resilience bonds
Climate change risk encompasses a series of extreme weather events. Since the climate is changing fast, climate risk is considered un-insurable, unless alternative risk transfer techniques are adopted. One possible strategy is issuing financial instruments, such as the Catastrophe (CAT) bonds, which allow the company to transfer the risk to financial investors. To this end, we propose a new model for supporting the green transition. We consider three agents: a government entity, a set of firms and insurance companies. We assume that the set of firms are exposed to possibly catastrophic risks, which could be mitigated through the implementation of green technologies. Hence, the insurance companies periodically issue so called resilience bonds, which are designed as CAT bonds, but, in case the risk decreases by the adoption of green technologies, they contribute to finance the green transition. This way a dynamical interaction takes place, involving the bonds interest rate, the share of virtuous firms and the threshold of the state coverage. We choose to model such interaction through a discrete dynamical model. By global analysis, we detect two main scenarios. In one all the trajectories converge to the optimal equilibrium, where all the firms adopt green technologies and the rate of the resilience bonds is minimum. In the other scenario, instead, a "poverty trap" appears, meaning that the trajectories starting in that region (poverty trap) converge to a sub-optimal equilibrium, where the share of virtuous firms is lower than 1 and the resilience bonds interest rate is not minimum.
In conclusion, we present a quantitative model for a green transition supported by a combination of financial instruments and public intervention. Finally, we investigate how our model can be implemented to mitigate the flood risk in Italy.
Joint work with Giovanni Rabitti and Emanuele Vannucci.
Contributed talk: date/time t.b.a.
Alexander Krauskopf (Deloitte, Germany)
Global warming and health insurance claims
In my presentation, I will be dealing with the effects of climate change on health insurance in Germany. Therefore, projections of future mortality and morbidity are created for various climate change scenarios. The methods and data basis used for this are explained. The effects on the development of claims cost are then presented. On this basis, the consequences for the development of health insurance premiums will then be shown and discussed in the context of demographic development.
Contributed talk: date/time t.b.a.
Christoph Krischanitz (Profi-Aktuar, Austria)
Under the radar screen: creeping developments in claim behaviour
Climate change is not only about catastrophic events. The changing weather conditions have slowly emerging impacts on people’s behaviour, on fauna, flora and the whole environment and on business activities in all industries. Actuaries need to measure these effects for many purposes. Product development, underwriting and pricing are only some of them, being enabled to define scenarios for risk management and ORSA are others. We discuss in this presentation what actuaries can actually do, what data is available and how to incorporate climate related data into standard actuarial models.
Contributed talk: date/time t.b.a.
Matteo Malavasi (University of New South Wales, UK)
The impact of extreme weather events on pro-environmental beliefs and actions in Australia
This study investigates the influence of extreme weather events, such as floods and wild-fires, on pro-environmental beliefs and behaviors in Australia, with a focus on actuarial implications. It examines how these events affect climate risk mitigation, adaptation strategies, and sustainable investments. By integrating decision-making, psychology, climate science, and actuarial studies, we aim to disentangle the effects of climate change on various decision-making processes. Using longitudinal datasets, this research assesses whether extreme weather events impact preference formation and translate into climate-conscious choices. This project promotes interdisciplinary collaboration and identifies key datasets, enhancing actuarial models and strategies for managing climate risk and promoting sustainable investments. We utilize data on voting preferences, political alignments, insurance claims, and climate risk mitigation strategies to evaluate the long-term impact of climate risk.
The results provide valuable insights for actuaries, policymakers, and community leaders to foster resilient and climate-conscious communities.
Joint work with Omid Ghasemi, Ben Newell, Charlie Ransom and Jiawei Zhang.
Contributed talk: date/time t.b.a.
Alaric Jules Antoine Müller (University of Lausanne, Switzerland)
Pluvial flood risk modelling using flood risk maps and an advanced stochastic weather generator: the case of Austria
Floods account for a significant portion of insurance losses resulting from natural hazards, and modeling these losses is challenged by complex spatial dependence patterns. While existing literature on flood risk often focuses on fluvial floods, floods triggered by heavy rainfall can also result in substantial insurance losses. In this paper, our objective is to model pluvial flood losses stemming from extreme rainfall events.
We utilize the advanced stochastic weather generator for simulating 2-D high-resolution climate variables (AWE-GEN-2d) developed by Peleg N., Fatichi S., Paschalis A., Molnar P. and Burlando P. (2017), to simulate high-resolution rainfall random fields. Focusing on one or more spatial regions, we develop monthly models to capture spatially and temporally dependent rainstorms through random fields. The AWE-GEN-2d model offers a robust framework for simulating such storms, enabling assessment of both the severity and frequency of pluvial flood losses in the regions under consideration. In the spirit of Albrecher H., Kortschak D. and Prettenthaler F. (2020), we demonstrate this methodology using Austria as a case study. Following the approach of Bernard E., Naveau P., Vrac M. and Mestre O. (2013), the country is divided into distinct regions based on clustering techniques applied to daily rainfall maxima. Several weather regions emerge as independent clusters, for which we estimate expected losses due to pluvial floods. To achieve this, simulations exceeding a given precipitation return level, such as the one in one-hundred years return level, are employed to identify locations prone to pluvial floods. To estimate losses due to pluvial floods in these locations, we propose a functional relationship between simulated extreme rainfall and resulting flood levels for different return levels. The simulated water depth in flooded locations is then utilized to estimate insurance losses through a damage function approach. The latter enables modelling of flood losses associated with extreme rainstorms for each region considered, while also providing insights into spatial diversification to enhance flood risk resilience.
Joint work with Hansjörg Albrecher, Erwan Koch, and Nadav Peleg.
Contributed talk: date/time t.b.a.
Fallou Niakh (CREST-ENSAE IP Paris, France)
Risk sharing and taxation
We consider an economy composed of regions that wish to be hedged against a disaster risk by using multi-region catastrophe insurance. This method disperses risk among high- and low-risk regions and ensures better risk mitigation. However, for natural risks, the insurer has a non-zero probability of insolvency. In this study, we introduce a public-private partnership between the government and the insurer that protects the regions against the risk of the insurer’s default. When a disaster occurs and the losses are below the coverage limit, the insurer fully compensates the regions. Furthermore, if they exceed the coverage limit, we consider that the central government decides to share the residual claims among the regions in the form of taxation. In this study, we propose a theoretical framework for regional participation in collective risk-sharing through tax revenues by accounting for their disaster risk profiles and their economic status.
Joint work with Arthur Charpentier, Philipp Ratz and Caroline Hillairet.
Contributed talk: date/time t.b.a.
Corinna Perchtold (Johannes Kepler University Linz, Austria)
A non-stationary spatio-temporal precipitation model for Austria
The talk illustrates the main results of a non-stationary spatio-temporal precipitation model interpolation process of three different precipitation scenarios distributed throughout Austria for the years 1973-1982 and 2013-2022. We model mean and maximum precipitation as well as dry spells with a Gamma, blended generalized extreme value and negative Binomial distribution. A generalized additive model accounts for influencing covariates as elevation and the coordinates of the monitoring stations which is then rewritten in a Bayesian hierarchical form. The spatial component of the model is represented through the stochastic partial differential equation (SPDE) approach and the temporal one through an AR(1) process. Inference is performed through integrated nested Laplace approximation (INLA) which comes along with a user friendly R-INLA package. The model outputs are visualised and give insights into changes in precipitation scenarios over two different time periods.
Contributed talk: date/time t.b.a.
Jens Robben (KU Leuven, Belgium)
The association between environmental variables and mortality: evidence from Europe
Using fine-grained, publicly available data, this paper studies the association between environmental variables, i.e., variables capturing weather and air pollution characteristics, and weekly mortality rates in small geographical regions in Europe. Hereto, we develop a mortality modelling framework where a baseline captures a region-specific, seasonal historical trend observed within the weekly mortality rates. Using a machine learning algorithm, we then explain deviations from this baseline using anomalies and extreme indices constructed from the environmental data. We illustrate our proposed modelling framework through a case study on more than 550 NUTS 3 regions located in 20 different European countries. Through interpretation tools, we unravel insights into which environmental features are most important when estimating excess or deficit mortality with respect to the baseline and explore how these features interact. Moreover, we investigate harvesting effects of the environmental features through our constructed weekly mortality modelling framework. Our findings show that temperature-related features exert the most significant influence in explaining deviations in mortality from the baseline. Furthermore, we find that environmental features prove particularly beneficial in southern regions for estimating these mortality deviations from the baseline.
Joint work with Katrien Antonio and Torsten Kleinow.
Contributed talk: date/time t.b.a.
Tomer Shushi (Ben-Gurion University, Israel)
Integrating socially responsible criteria in modern portfolio theory via a tradeoff between focusing on the data of stock returns and their ESG scoring
Socially responsible portfolio selection has recently gained much attention. Environmental, social, and governance (ESG) scores offer a means to quantify a company’s contributions to both the environment and society. We propose novel frameworks for optimal portfolio selection that provide a tradeoff between entirely focusing on the data of each of the stocks in the portfolio and entirely focusing on their ESG scores. To obtain the optimal weights as part of the portfolio selection process, a multivariate constrained optimization problem needs to be solved. We propose an explicit solution for such a problem and examine it using numerical illustrations for the mean-variance and tail-value-at-risk measures. These illustrations allow us to gain insights into the models. For example, under the proposed model, the investor’s higher risk aversion parameter implies a lower contribution of socially responsible preferences to the computed optimal weights.
Contributed talk: date/time t.b.a.
Johanna Suppan (FAM @ TU Wien, Austria)
On the correlation between extreme weather events and hospital admissions in Austria
In this talk, we look at the relationship between hospitalization rates and weather data in the period 2016 to 2023, using the data of an Austrian health insurance company. With the help of Generalized Additive Models (GAMs), the dependence of hospitalization rates on weather data is analysed per age group, region, gender, and diagnosis. The objective is to assess the risk connected to extreme weather events for a pool of insured individuals.
Contributed talk: date/time t.b.a.
Davide Trevisani (University of A Coruña, Spain)
Scope 3 capital design for carbon-emissions-facilitation tax risk
Among the efforts to address climate change, governments have introduced over seventy carbon pricing instruments (CPIs). Banks finance a significant fraction of global emissions, and many of them have committed to reducing their facilitated, or Scope 3, emissions to (net) zero by 2050. However, governments may introduce a new CPI impacting banks directly on their Scope 3 before 2050, providing limited notice of this.
Here, we present a carbon policy capital to make banks resilient against the possibility that governments could introduce this kind of CPI. For this purpose, we present a suitable market model to assess losses due to uncertain carbon policies. Moreover, we consider stylized examples based on interest rate swaps, which are financially relevant for counterparties with high levels of emissions. The proposed computational methodology is based on analytical formulas and Monte Carlo techniques. The obtained numerical results suggest that, for long maturities and emission intensities, an efficient capital requirement would be larger than the one currently used for counterparty credit risk. To ensure the resiliency of banks, regulators have the mandate to set adequate capital requirements. Therefore, a Scope 3 capital introduction needs urgent consideration by regulators.
Joint work with José Germán López, Chris Kenyon, Carlos Vázquez, and Mourad Berrahoui.
Contributed talk: date/time t.b.a.
George Tzougas (Heriot Watt University, UK)
Investigating the effect of climate-related hazards on claim frequency prediction in motor insurance
Quantifying the meteorological event-related risks has become increasingly important in general insurance because extreme climate events may trigger excess claims that may harm the insurer’s portfolio. On the other hand, it is challenging to model the relation between climate events and claim frequencies since climate events are often not fully recorded. Motivated by the above issues, we introduce a new class of compound frequency models for joint modelling of storm and storm-triggered claim frequencies observed by a Greek insurance company. The proposed models uncover the joint distribution of the actual storm and claim processes even if the observed data are incomplete. Additionally, geospatial covariates are included to assess their impacts on the storm and claim frequencies.
Finally, we discover a negative actual intrinsic dependence between the storm and per-storm claim frequencies, which implies that an insurance company may indeed enjoy diversification benefits from climate change that causes more meteorological events.
Joint work with Tsz Chai Fung and Himchan Jeong.
Contributed talk: date/time t.b.a.
Stefan Wrzaczek (International Institute for Applied Systems Analysis (IIASA), Austria)
A model-based measure for the resilience of resource use under the risk of disruption
While useful conceptualizations and measures have been developed that allow to assess the resilience to shocks of eco systems, socioeconomic systems or individuals, these are often context-specific, somewhat ad-hoc, focused on only certain dimensions of resilience, and not always easy to apply to the assessment of concrete policies. In this paper, we build on a simple yet general model of resource use under the risk of potentially catastrophic disruptions, and derive a rigorous, comprehensive, and policy-oriented measure of socio-economic resilience. By way of numerical analysis, we illustrate how the measure captures the key aspects of resistance, recovery and robustness; how it embraces current resilience and the continuation resilience following future shocks; and how it can be employed to study the resilience of different harvesting policies both from an ex-ante perspective and following a cascade of shocks.
Joint work with Michael Kuhn.
Contributed talk: date/time t.b.a.
Cem Yavrum (METU, Turkey)
Assessing the applicability of the Actuaries Climate Index within weather derivatives framework
The global climate change has emerged as one of the most complex and pressing challenges confronting humanity. With the impact of climate change intensifying, a growing number of industries, such as agriculture, insurance, energy, and tourism, find themselves increasingly vulnerable to weather-related risks. In response to this challenge, industries, organizations, and even individuals have the opportunity to utilize weather derivatives as a means to manage or mitigate these risks. Weather derivatives are financial instruments designed to hedge against adverse weather conditions, with their value contingent upon specific weather variables. Since 2016 the Actuaries Climate IndexTM (ACI) has integrated diverse climate variables indicative of extreme weather conditions. The ACI serves to enhance the comprehension of climate trends and their potential impacts among actuaries, insurance companies, and policymakers through a monitoring tool focused on climate change indices. Drawing on data from twelve subregions across the United States and Canada, the ACI assesses six climate indicators, including high and low-temperature extremes, precipitation, drought, extreme wind, and sea level rise. Weather derivatives with underlying variables sourced from the components of the ACI, this study seeks to evaluate the effectiveness of the ACI in managing weather-related risks for market participants. Additionally, it aims to highlight the ACI’s potential as a dependable benchmark for pricing weather-related financial instruments, potentially encouraging wider adoption of the index in future risk management strategies.
Joint work with A. Sevtap Selcuk-Kestel.
