Could superannuation help solve our healthcare funding crisis? In our latest instalment of Investment Ideas, hosts Douglas Isles and Jan Swinhoe speak with Dr Raymond Yeow, a unique voice in healthcare who combines perspectives as both a GP and an actuary. Key areas discussed include: A proposal for Health Savings Accounts within superannuation, based on […]

We’re witnessing a historic democratisation of data and AI capability. So says Zilinka Jiang in our latest instalment of Investment Ideas. In Digitisation, Zilinka sits down with hosts Jan Swinhoe and Douglas Isles to discuss the transformative potential of AI. Key areas discussed include: Practical applications in business, including how Telstra has improved customer service […]

How can retirees confidently spend their super without fear of running out? In our latest instalment of Investment Ideas, hosts Jan Swinhoe and Douglas Isles explore the challenges of the decumulation phase with Giacomo Tarantolo from Unisuper and Estelle Liu from AMP, specialists in retirement solutions. Key areas discussed include: The complexity of transitioning from […]

Are we ready for a world with fewer babies and longer lives? In the latest instalment of Investment Ideas, Demographics, Jan Swinhoe and Douglas Isles sit down with Dr David Knox AM and Associate Professor Katja Hanewald to explore the economic and investment implications of global demographic shifts. Key areas discussed include: The dual demographic […]

How do complex views on decarbonisation create investment opportunities? In the third episode of Investment Ideas, Decarbonisation, Nicolette Rubinsztein AM, incoming Chair of Greenpeace and financial services non-executive director, speaks with Jan Swinhoe about Australia’s energy transition and how diverse views on decarbonisation create opportunities for investors. Key areas discussed include: The role of gas […]

As nations continue to respond to crises by taking on more debt, are we kicking the can down the road on global debt sustainability? In the second episode of Investment Ideas, Debt, Douglas Isles sits down with Simon Warner, Head of Portfolio Management at Aware Super, to explore how investors should think about debt sustainability […]

The debut episode of“Investment Ideas”dives straight into one of the most significant forces reshaping global markets: deglobalisation. Co-hosts Jan Swinhoe and Douglas Isles speak with Belinda Cheung, Director in the Economics and Capital Markets team at Future Fund, and Annette King, non-executive director, former President of the Actuaries Institute and CEO, to explore […]

What happens when two of Australia’s most experienced investment actuaries sit down with leading experts to explore the forces shaping our financial future? The Actuaries Institute’s new podcast series“Investment Ideas”finds out. As markets grapple with profound structural changes, from a shift back to imperialism to declining birth rates globally,“Investment Ideas”pairs Jan […]

In the latest episode of ConverseAI, host Tong Zhang speaks with Estelle Liu about how data science can help address key challenges in superannuation, including large-scale projections, understanding customer lifetime value and predicting member behaviour. Exploring essential data science techniques, Tong and Estelle discuss: essential data science techniques ; the need for proper training; the […]

Traditional aggregate reserving techniques assume that each data triangle used is a homogeneous group of claims. However, the practical constraints of management and reporting requirements often hinder the adoption of a statistically optimal approach to segmentation. There is often a reliance on legacy class hierarchies, business processes or expert judgement due to the difficulty of determining an optimal segmentation. This can reduce homogeneity in triangles and affect the performance of aggregate reserving techniques. To address these challenges, we propose a framework and methodology for automated segmentation to allocate individual claims in a reserving class to homogeneous subgroupings more suitable for triangle-based projections. The proposed approach involves use of machine learning techniques to: • Evaluate claim features that can relate to claim development in a statistically significant way. • Implement an algorithm to segment claims. • Assess and compare different segmentations on their impact on claim reserve accuracy. We also explore the practicalities for constructing models in a way that allows managing segmentation changes across reporting periods. Overall, we propose a methodology and practical framework for leveraging clustering techniques as part of a reserving process to improve accuracy and reduce the need for intervention of actuaries managing insurers’ claims reserves.

The rapidly evolving landscape of insurance in the modern era brings with it a unique conundrum—how do we ensure fair pricing, especially when introducing new products with sparse data? Though long-standing, the principle of fairness and anti-discrimination is complicated by the dynamics of innovative products, evolving consumer behaviour, and the complexities of AI and emerging risks. In the first part of this presentation, Michael will share practical challenges he faced when helping relaunch the startup insurtech PassportCard Travel Insurance after the Covid pandemic forced the hibernation of the business. He will discuss the team's approach to applying established guidance and literature around anti-discrimination for protected attributes such as age and disability, especially where the apparent lack of data requires more diverse thinking. Through real-world examples, attendees will gain insight into practical strategies to navigate these challenges, ensuring product pricing that aligns with both business goals and societal expectations. In the second segment of our session, Fei will review the state of the art of research in this area with an international perspective. She will then respond to the practical challenges shared by Michael with insights and potential solutions. A list of open questions that need further work and collaboration between academia, government, and industry will be summarised and discussed in the end.

FY 2022/23 marked Australia’s cyber wake-up call – Australia experienced the largest breaches in its history, with the personal details of more than 10 million Australians impacted by one data breach alone. In response, there has been a suite of initiatives from government looking to tighten Australia’s cyber security settings, and prepare individuals, corporates and the companies for the breaches that will inevitability come. This has included strong signals from Australia’s corporate regulator that it will look to prosecute those companies who have not adequately prepared for cyber incidents, the development of a new cyber security strategy for the nation, and incorporation of cyber into new prudential standards for APRAregulated entities. Corporate Australia has responded to the ‘annus horribilis’ by increasing its investment in information security, and looked to the cyber insurance market to transform some of the risks faced. In this paper we: • Look at the major changes to the cyber landscape that have occurred over the previous year, and score how well corporate Australia has responded to the cyber wake-up call • Discuss how actuaries can use our unique skillset to address the challenges posed by cyber risk, and add value both in the boardroom and to policy makers by 1) measuring cyber risk, 2) designing solutions to address cyber risk, and 3) quantifying the impacts of cyber risk.

The compounding effects of heavy rainfalls and strong winds can lead to severe disasters, resulting in catastrophic losses. This paper presents a novel approach to modeling joint extreme climate events involving precipitation and wind speed. A case study is undertaken, utilizing data gathered from weather observatories spanning across New South Wales, Australia. In terms of methodology, we propose an automated threshold selection method and implement an automated declustering technique to address inherent challenges in applying extreme value theory to a substantial number of datasets. Remarkably, these methodological advancements are applicable to various other forms of joint climate extremes. Furthermore, we introduce a proxy derived from these joint extremes and create a heatmap to identify any significant trends of joint heavy rainfalls and strong winds. This proxy offers a refined understanding of the compounding effects of joint extremes and holds the potential to provide valuable insights for pricing insurance products.

The Federal Government’s Strengthening Medicare Report sets out an expansive agenda to improve healthcare in Australia through blended care pathways, supported by investment in better use of digital technology to share patient data. But the My Health Record system that underpins part of this plan has existed for over 10 years, has seen $2bn in investment and is still considered, to date, unsuccessful. This paper explores the reasons for those failures but also the possibilities for health and health actuaries should further investment in this system prompt the progress and growth long desired. With the Strengthening Medicare Report also putting the role of private health insurers in primary care back on the agenda, the presentation will also explore the current landscape and sentiments around private health insurers’ involvement in this area and the benefits to consumers of an appropriately safeguarded but better integrated data sharing platform.

]In recent times, the reinsurance market has been going through a hardening phase with substantial price increases (corrections), tightening coverage, and even withdrawn capacity. While it is relatively well understood that the key drivers are post-COVID recoveries, challenging state of the economic, claims activities, and uncertainty associated with climate changes and geo-political risks, the solutions are not obvious, especially for the public sectors. When it comes to natural catastrophe risks, many in the private sector speaks of the language of economic and/or regulatory capital, e.g., ICRC, 1/200, 1/1000, return period, etc. The public sector speaks the language of funding, e.g., taxations, levy, etc. Since (re)insurance is an important form of funding for the government, a hardening (re)insurance market means that the government will face even wider protection gap, delay in infrastructure projects development, bigger risk in economic development, etc. Thera are, however, other public sector solutions available which has proven to be innovative and effective. These solutions help facilitate economic development and increase insurance protection and financial inclusion. In this context, the public sector refers to governments (at sovereign and sub-sovereign levels), state-owned enterprises (SOEs), sovereign wealth funds (SWFs), development banks, export credit agencies (ECAs), multilateral organisations and 'third sector' entities/NGOs. The presentation will conclude with a few case studies. Indicatively, there will be case studies like: (To be confirmed) Catastrophe bond for Pacific Alliance sovereign earthquake protection covering four countries. Mexico Hurricane Coral Reef Cover - first insurance coverage to protect a natural asset. Sovereign catastrophe risk transfer in the Asia-Pacific region – PCRIC Livestock protection against drought for smallholder farmers Longevity swap for sovereign pension

An overview of how implementation of AASB 17 for the health insurance industry has gone. Where was the conflict? Where were the tears? Where were the happy hugs? With the standard now in-force for the entire industry, I will comment on: • The path taken to where we are now. • How did the implementation ahead of the first external reporting (APRA returns for 30 September 2023) go? How did it all go down? What went well What went badly Lessons for the future • What the public accounts have to say about what is being implemented. What level of consistency is there? Where is there divergence? What is the range of the divergence? Differences between large/small or profit/non-profit? Are differences reasonable and expected? • Where did actuaries get caught out? • Next steps for the industry

A logistic regression model was used to estimate how much more likely a diabetic with a mental health condition is to have an admission than a diabetic without a mental health condition after adjusting for a robust set of risk factors. These results indicate that diabetics with a mental health condition were 48% more likely to be admitted and 45% more likely to be admitted for a high costing admission than a diabetic without a mental health condition. In addition, these diabetic members were 14% more likely to be admitted for a diabetes-related admission and 7% less likely to be adherent to their diabetic medication. This illustrates the impact of mental health conditions on total costs, in-hospital admissions and on the management and control of diabetes. Causal inference refers to the process of drawing a conclusion that a specific treatment was the cause of the observed outcome. A causal forest model was used to determine the causative relationship between diabetics developing a mental health condition and the impact on scheme cost, adjusting for appropriate confounders. Results showed that developing a mental health condition would increase the total cost of diabetic members by 7%, increase costs in-hospital by 11% and increase total costs excluding spend on mental health related claims by 4%. Furthermore, developing a mental health condition has a greater causative impact on in-hospital costs for diabetics with progressed disease compared to those in the preliminary stages of their disease. Understanding these interactions between mental health conditions and other chronic conditions allows us to appropriately tailor disease management programmes and optimise risk management strategies with the aim of reducing scheme costs and managing premium increases.

The rapidly evolving landscape of insurance in the modern era brings with it a unique conundrum—how do we ensure fair pricing, especially when introducing new products with sparse data? Though long-standing, the principle of fairness and anti-discrimination is complicated by the dynamics of innovative products, evolving consumer behaviour, and the complexities of AI and emerging risks. In the first part of this presentation, Michael will share practical challenges he faced when helping relaunch the startup insurtech PassportCard Travel Insurance after the Covid pandemic forced the hibernation of the business. He will discuss the team's approach to applying established guidance and literature around anti-discrimination for protected attributes such as age and disability, especially where the apparent lack of data requires more diverse thinking. Through real-world examples, attendees will gain insight into practical strategies to navigate these challenges, ensuring product pricing that aligns with both business goals and societal expectations. In the second segment of our session, Fei will review the state of the art of research in this area with an international perspective. She will then respond to the practical challenges shared by Michael with insights and potential solutions. A list of open questions that need further work and collaboration between academia, government, and industry will be summarised and discussed in the end.

We traditionally quantify climate change risks as a function of the risk determinants of hazard, exposure, and vulnerability. Traditionally, insurers have been concerned with the built environment domain, focussing on assessing the damage to properties and buildings from natural disasters. Many practitioners are extending climate risk assessments to other domains to look at climate risk more holistically, as climate change does not only impact the built environment, but also the social, natural, and economics domains. But there are limitations to only relying on a quantitative assessment as climate change risk is extremely complex and interacts both internally between the determinants of risk, and externally with many other risks. This can lead to cascading and compound events with widespread systemic risk impacts. For example, vulnerabilities in one domain can stem from multiple hazards, leading to exposures in other domains, whilst exposures that change over time can create changing vulnerabilities across domains. Hazard interactions from disasters may cause feedback loops that lead to untenable capacity to support social, economic, natural and built domains. A qualitative risk assessment can help identify deficiencies in quantitative modelling and suggest improvements for future risk assessments. Our methodology for qualitative assessment refers to reports aggregated from community interviews, which involves decision-makers, local officers, emergency response officers, and other stakeholders. The methodology combs through the main conclusions from the reports to form our basis for the interactions between hazard, vulnerability, and exposure under the four domains (built, social, economic, and natural). The confidence of the interaction increases where there are multiple instances of the same interaction being noted across different reports, or sometimes across similar themes within the same report. The output from our qualitative assessment are multiple mapping diagrams that show the interactions captured within the reports within and between domains. We then draw conclusions from the mapping diagrams, which can be then used to supplement the quantitative risk assessment and identify gaps for future assessments.

Our profession's move into sophisticated data science was led by GI practitioners taking techniques developed/implemented for pricing to much wider areas and applications. Yet GI claims liability estimates are largely made using techniques and processes that were developed in the 1980s. At the time, spreadsheets were a new thing and appeared to be, and indeed were by the standards of the day, excellent tools for implementing the techniques. This situation no longer pertains. We argue that modern statistical and machine learning techniques, combined with well documented research specific to the use case, require the profession to move on from traditional approaches. The many benefits can include: • more statistically robust and accurate estimates; • more flexibility and efficiency of model specification (and variation/exploration thereof); • greater automation of and statistical basis for parameter estimation ; • separation of data from model logic; • better process control and documentation; • improved collaboration. We also discuss common arguments against moving away from spreadsheet-based artisanal processes, such as their flexibility, accessibility to non-coders and perceived adequacy. In addition to outlining the need for, and value of, wholesale change in the profession's approach to claims reserving, we explore what changes are necessary to professional standards, guidance and the education system in order to assist practitioners make this change. The work is based on the authors’ experience in designing, building, operating and reviewing reserving processes, as well as interviews with a number of senior reserving actuaries.