Insurance logic in risk rating & pricing should be used for green finance

French insurers have reported paying out a total of over 16 bn euros in 2022 and 2023 due to climate disasters. In the budget allocated by the Caisse Centrale de Réassurance (CCR), this corresponds to once-in-50 years maximum expected loss. After this kind of shocks, insurance companies should raise the bill to make sure there is enough funds available for coming years. All insurance policies should be readjusted for high risk, including the areas not affected in 2022 and 2023. Readjusted for what? What are the main climate risks that require special attention?

According to 30-year statistics, in France, on average, 60% of insured losses are caused by floods. Additionally, flooding is the primary climate risk in terms of the number of municipalities affected.

It is worth to mention that the average budget allocated for flooding by the CCR is less than one billion per year. Clearly, the actual losses are incomparable to the dedicated budgets.

Here is another scientific figure. In France, on average, 45% of flood claims, almost half, are outside the official flood zones. 45% every year! This indicates that official flood maps need to be updated more frequently than every ten years, using different methods, not just historical observations but also future predictions. This is how insurance companies map flood risk internally, they process more data, more frequently and in more detail. This is the core of their business: assymetric knowledge of climate risks. Insurance companies don't ever use any gouvernmental official flood maps for the pricing.

What is the "secret sauce", how do they manage to assess future flood risks years ahead? Floods have very similar patterns, including the rainfall amount, the distance to river, soil type, land use and geography. Predictive modeling of floods in one area can be done using the observations for another area. AI-based predictive models enable detailed mapping of past-&-future floods and the corresponding physical damage for each municipality projecting the flood risk far beyond weather forecasts: to 5, 10, 20 and 50 years ahead.

Estimating future economic losses would allow assessing the necessary budget for municipalities to upgrade flood defenses and, if needed, build new protections.

Today, the total debt of all French municipalities is around 70 billion euros, or, on average, 1,000 euros per inhabitant. This debt continues to grow with each natural disaster, each more severe than the last.

Finding new sources of funding, other than loans, is urgent to meet the needs of municipalities facing flood risk.

I would be delighted to present the concept for a novel funding opportunity for municipalities.

This solution could emerge thanks to technologies based on the predictive flood risk modeling for the future, built with artificial intelligence and the existing economic modeling methods used by insurers.

It is a matter of fact that insurers have the most advanced methodologies to evaluate the expected losses for future climate risks over the next 5, 10, and 50 years.

Today, the solution we need to support the municipalities is a new financing instrument for anticipation and prevention projects. An instrument other than debt or insurance for climate perils. With this new financing, municipalities will no longer need to borrow to protect us!

As a result, not only will the cost of insurance will decrease thanks to investments in infrastructure, but also new economic opportunities will be possible in well-protected municipalities.

Another positive outcome: thanks to the new scientific methods for AI-powered flood risk mapping, municipalities will be better informed for all any new real estate development projects.

For the safety of the population, anticipation and prevention of risk must be the focus of sustainable finance.