Since 1992, when Hurricane Andrew ravaged the state of Florida and caused the bankruptcy of numerous insurance companies who struggled to reimburse their policyholders, the insurance industry started hiring meteorologists and catastrophe modellers to build models able to better assess short- and long-term climate-related physical risks, says Geoffroy Marcassoli, ESG Audit Leader at PwC Luxembourg.

As climate change proceeded unabated since the early 1990s, exceptionally damaging weather events – such as wildfires, droughts, storms etc. – have only increased in frequency, forcing the insurance industry to face the daunting challenge of global warming. As a matter of fact, “most assets will be uninsurable” if average temperatures are raised to 4°C above pre-industrial levels, as per a recent study.

If the insurance industry were no longer able to operate, global GDP would be impacted beyond recognition. Not only do insurance companies hold a lot of corporate and governmental debt, but any negative trend within the sector impacts the banking sector as well.

Populations at large would also be left in an extremely vulnerable position, with little to no safety net against climate-inflicted damages if governments cannot foot the bill.

A gap between climate science and financial actors’ climate scenarios

While the insurance industry has relied on catastrophe models to predict upcoming natural disasters and adjust prices accordingly, these models rely on historical data. Given the unprecedented nature of global warming, such models might not be adequate, and some insurers may see the cost of insuring assets from climate risks as simply too high.

The scenario modelling and data providers used by insurers for their climate risk assessments are also used in the broader financial sector.

As a matter of fact, the financial sector has undertaken a considerable amount of effort and expense to develop climate-change scenario analyses – over 80% of financial institutions are currently performing climate-scenario analysis (up from less than 50% in 2019) and are using the results to assess climate and transition risks, according to a 2022 survey by the Global Association of Risk Professionals.

However, a worrying disconnect between the financial sector’s models and those used by climate scientists has been noted by the aforementioned IFoA-Exeter study, with some financial institutions estimating GDP loss to be identical in a disorderly transition scenario as in a failed transition scenario.

Many scenarios are optimistic because they do not take into account extreme weather events and expect governments to act the way they ideally should.

In addition, tipping points – such as glaciers melting and sea levels rising could set in motion complex feedback loops which permanently alter the earth’s climate and seriously affect forests’ health, water supplies and soil fertility. The complexities rooted in such tipping points are not commonly taken into account by all models.

Forecasts: Damned if they are good, damned if they are bad

It is thus crucial for regulators and financial market participants to scrutinise overly optimistic climate risk assessments and ensure that they are not based on incomplete or biased data.

However, in negative scenarios, the financial sector potentially faces a long-term environment where all assets may gradually become uninsurable – an untenable perspective. The sector, and society at large, is thus currently mired in uncertainties, as we still do not know exactly which policies will be implemented in the long-term, and there is still debate on how disruptive policies need to be.

As The Economist astutely highlighted in July 2023, “climate change involves the messy world of policy as well as the clarity of physics” and “there is no model that can predict whether policymakers will pull the levers that are available to them to prevent [wildfires and other extreme weather events] from happening.”

There is no one-size-fits-all approach that could take all potential policies and their impacts into account, which does explain to a certain extent the struggle financial market participants face when it comes to climate risk assessment. All we know is that if not enough is done to reduce global emissions, things will get bad before getting worse.

Technical (and policy) solutions exist

The industry has found some creative solutions to the rise of primary and secondary perils through developing Insurance Linked Securities (ILS) and catastrophe bonds: these instruments transfer the risk of providing insurance in case of natural disasters to investors in global capital markets.

Parametric catastrophe bonds, for instance, ensure a pay-out to customers based on a trigger event (e.g., a certain amount of rainfall, measure of windspeed). Demand for these instruments is growing, especially in regions exposed to increasing extreme weather events.

For regulators and financial market participants dealing with the aforementioned limits in mathematical scenario modelling, a switch to qualitative, narrative scenarios and to visualisations may be a justified move, as “quantitative scenarios are ill-equipped to effectively model the impacts of tipping points and the cascading effects of climate change,” as per the IFoA-Exeter study.

Obtaining insurance from climate change-related disasters will only get harder as temperatures rise. Investors have started building portfolios that not only hedge climate risks but also focus on industries more likely to withstand the effects of extreme weather events.

It is thus important for all stakeholders to keep in mind that focusing on data and mathematical models is not enough, and that qualitative assessments are as crucial.

Looking beyond assessing climate change-related risks, public and private stakeholders should also think about how to minimise and ultimately reduce them. Only by redesigning current economic systems through green industrial policies will we be able to reach the goals of the Paris Agreement, and hopefully reverse climate change in the long run.

The cost of inaction far exceeds the most unfavourable realistic outcome.

By Geoffroy Marcassoli, ESG Audit Leader at PwC Luxembourg