The reinsurance market in the face of change

Reinsurance provides a back-bone for socio-economic recovery in times of disaster that has gained in importance given the increasing occurrence and severity of destructive weather events.¹  Yet, the everyday activities that enable the reinsurance market to work remain somewhat a mystery. In ‘Making a Market for Acts of God’², our theory of market making offers a glimpse into the norms, rituals and practices of the reinsurance industry. Over three years (2009-2012), we lived and breathed reinsurance having gained access to 25 reinsurance and broking firms and over 33 insurers globally. As we observed this market we saw rapid changes to long-standing practices and norms that underpinned how reinsurance is traded. These changes may have long-lasting effects specifically to the pricing of reinsurance deals and the longevity of capital availability.

Insights from a fly-on-the-wall study

Gaining the trust and goodwill of hundreds of underwriters, brokers and insurers, we were able to get behind the scenes to observe reinsurance trading at the world’s most influential trading hubs, such as Lloyd’s of London, Bermuda, several centres of trading in Continental Europe and Asia-Pacific hubs such as Singapore and Japan. We observed the norms, rituals and practices that sustained the evaluation of deals of this 300-year-old market. At annual conferences, we sipped champagne and drank overpriced coffees whilst listening to the regular catch-up, information sharing and positioning tactics between reinsurers, brokers and insurers (clients). Returning to their desks, we shadowed underwriters as they evaluated and priced reinsurance deals. As they did so, underwriters drew on both catastrophe modelling (when available and appropriate) and also the information exchanged with clients and brokers within long-term relationships, including the most recent updates at industry conferences and site visits of the risks themselves (everything from churches, farms to space shuttles). In doing so, they priced the unknowable: the event that might not happen, and the associated losses that they cannot know in advance. In the final year of our study it appeared that this industry, characterised by long-standing professional and personal relationships and cultural norms, was at the crossroads³. In our study we demonstrate that these changes are potentially profound, undermining current practices for evaluating and pricing deals.

Squeezing the margins while playing with science and nature

Trading reinsurance is tightly linked to two factors: the measurability of risk and the destruction caused by nature. Post hurricanes Andrew (1992) and particularly Katrina, Rita and Wilma (2005), catastrophe modelling has become more sophisticated providing industry standards for the pricing of risks. Yet, the evaluation of risk remains a blend of art and science. While underwriters draw on the output of catastrophe models (science), they balance these values with their understanding of the underlying risk and particular characteristics of a client (art) that impact the potential size of a claim, such as the speed of claims handling, regular maintenance assessments etc. The tendency towards ‘art’ or expertise based judgement over science and models increases with those risks where information is either highly variable, such as marine risk, or simply unavailable, as with many emerging markets. Expert judgment has thus remained a key element in evaluating reinsurance risk.

However, there are several forces that shift the balance towards a dominance of science, resulting in an increasing emphasis on modelling in determining the price of a risk. On the one hand, insurance firms foster greater reliance on catastrophe modelling in order to determine the level and size of risks to cede to the reinsurance market. Due to industry consolidation, these insurers have continued to grow into global players with a diversified portfolio of policies⁴. Global diversification and financial strength offer the opportunity to offset risks from different classes of business and geographical regions within the insurer’s own portfolio, so retaining more risk in-house. Yet, the occurrence of multiple, significant events challenges decisions to offset risks based on diversification across regions and classes of business, as demonstrated in 2011 which, due to multiple catastrophic events around the world, turned out as ‘the costliest year ever in terms of natural catastrophe losses’⁵. Based on the latest IPCC report⁶, such multiple, significant events within a year may be on the rise.

On the other hand, new players entering the reinsurance market further the reliance on catastrophe modelling to evaluate reinsurance risks. To offset other financial risks, hedge funds and pension funds invest in Alternative Risk Transfer products such as catastrophe bonds or other insurance-linked securities. New reinsurance products pose particular challenges. Firstly, they are specified on a narrower set of parameters than a traditional reinsurance deal, so disposing greater reliance on models not judgment⁷. Secondly, these products remain largely untested in terms of losses triggered, or response to large claims, so raising questions about the longevity of alternative capital available for reinsurance. It is thus vital that they remain a supplement to, rather than a substitute for, the traditional reinsurance deals. Otherwise, if traditional reinsurance erodes insurers may face the dire outlook of who to turn to in desperate times.

Implications of change

A growing emphasis on catastrophe modelling comes with an increasing reliance on scientific assumptions about weather patterns, and leads to an erosion of valuable skill and expertise currently brought to bear in the underwriting process. Such changes may increase the vulnerability of insurers and reinsurers alike to potentially underestimate the potential losses from written risks. The growing importance of catastrophe modelling to evaluate risks may create a myth about accurate return periods based on past weather patterns. Yet, dramatic changes in weather patterns leading to more severe and frequent events demonstrate the potential fallacy of assumptions within catastrophe models. For instance, events such hurricane Sandy (2012) or the length of extreme droughts in North America (2015) demonstrate the unpredictability of weather patterns. Events are ‘unpredictable’; thus challenging some fundamental assumptions and proxies designed to accurately measure risks. The gap between a risk’s abstract properties, as expressed in models’ probabilities, and the occurrence of actual events raises questions about an overreliance on modelled assumptions when pricing risks. Such questions should emphasise the importance of including soft factors that continue to contextualise risks on an annual basis. Otherwise, valuable skills and expertise in evaluating reinsurance deals may get lost.

While we concur with others projecting that the reinsurance industry is at the crossroads, our study outlines additional factors that may exacerbate change. Changes to the way reinsurance is evaluated and traded may have long-lasting effects specifically on the pricing of reinsurance deals and the longevity of capital availability. Such changes may have profound effects on socio-economic recovery following disasters. Identifying the implications of these changes provides an opportunity to attend to the ongoing viability of the reinsurance market.

CPD: Actuaries Institute Members can claim two CPD points for every hour of reading articles on Actuaries Digital.