Tackling the here and now: why we need to rethink our natural catastrophe risk models

The risk landscape is evolving very dynamically. Relying heavily on historic data is creating a blurry picture of current risks.

By Martin Bertogg

While 2020 had far more than its fair share of tragedy and disaster as a result of the pandemic, major catastrophes of other sorts were actually few and far between. Contrary to other years, 2020 didn't see any staggering catastrophe losses from earthquakes or cyclones – so-called primary perils.

Frankly, that is just down to good luck. With a record-breaking Atlantic hurricane season that featured 30 named storms – of which 11 made landfall in the United States – 2020 was more like a close escape. But 2021 could be very different.

What we did see last year was a continued trend towards more small-to-medium secondary perils with greater loss severity – that is, thunderstorms, wildfires and localised flooding, for example.

For re/insurers, these frequent, comparatively low-loss events are adding up.

Of the USD 190 billion-worth of economic losses due to natural catastrophes in 2020, costs of USD 81 billion were covered by the insurance industry. That’s the fifth-highest amount in the past half-century. And almost three-quarters of that amount was due to the growing number of smaller perils.

So, how can we as an industry ensure we're responding appropriately to this rapidly emerging risk and help build a more resilient society?

Learn from the past, but focus on the now

Our current risk models tend to mean we look in the rear-view mirror too much. The risk landscape is evolving very dynamically. Relying heavily on historic data is creating a blurry picture of current risks. We need to be wary of relying only on model results based on multi-decadal averages when more recent history gives us a very different picture. In a similar vein, too much of a focus on the far future.

In a similar vein, too much of a focus on the far future can also be unhelpful. Understanding what will be happening in the long term as a result of climate change is important, but we need to narrow our lens of vision to look at what the events of the past few years can tell us about the risks we are facing right now and in the next few years.

Natural catastrophe modelling in the insurance industry has come a long way in the last ten years, both for underwriting and portfolio risk management. While we've established a number of helpful principles, we need to continuously recalibrate our models to help the industry better understand the risks it faces in the near-term. For example, to sustainably underwrite a property catastrophe portfolio with a cover duration of one or maybe just a few years, we have to be able to correctly assess the growing risk faced in the next underwriting year.

The risk landscape is shifting

 It is important to bear in mind that the risk posed by both primary and secondary perils is not static. Climate change is a factor, but so too are several man-made drivers: land use planning and its impact on urban settlements, the housing stock and its vulnerability as well as changes to claims settlements. With more urban sprawl into wildland or urban development in high-risk places such as along coastlines, there are more people and assets at risk, and more potential costs to the re/insurance industry.

Wildfires are an alarming example of this. Climate change is affecting their intensity and frequency, but also there are far more people and houses in their way than ever before. As a case in point: from 1990 to 2010, the number of homes in the US in the wildland-urban-interface grew by approximately 40%. While dry climate conditions fuelled the wildfire seasons of 2017, 2018 and 2020, it has been the countless houses standing in harm's way that actually translated into significant insured losses.   

Similarly, severe convective storms are hitting us harder. With more urban areas in their path, these storms have a greater chance of causing massive losses than they have done in the past, with a whopping USD 36 billion of insured damage in 2020 alone. The probability of extreme weather turning into a catastrophe for the insurance industry has been rising steadily for the past 50 years, and has become very acute in the past 10 years.

Clearly, as an industry we have more homework to do to help us better understand the sensitivity of quickly urbanising areas to climate change-influenced risks.

Putting a price tag on risk

Insurance is a key part of financial resilience – and can be key to getting economies back on their feet after catastrophes. But the sector also has an important role in putting a price tag on risk.

Risk modelling and adequate insurance premiums are a crucial part of helping businesses and societies understand the impact of their actions and the choices they face to cope with climate and natural hazard risk. It can be an important part of risk mitigation – steering development away from high-risk areas, for example.

Governments, too, have a role in taking a longer-term, evolving view of climate and natural catastrophe risk. They can take measures to maintain the insurability of assets, for example by ensuring considerate land use planning or fostering risk adequate insurance schemes. They can help to close the perception and protection gap by creating more public awareness of the risks. And they may need to focus on investments to be made now, to prepare for a world of more extreme climate events.

Both primary and secondary natural catastrophes are predicted to rise in both intensity and frequency as climate change, urbanisation and other factors increasingly come into play.

Risk modelling and ever more data has become crucial to helping insurers, and society more broadly, better understand how to manage and mitigate the risks and losses associated with them. But realigning our focus to the here and now, and not yesterday, will be a crucial part of understanding the absolute risks we face, as the impacts of a rapidly shifting environment are increasingly felt.