You disliked this video. Thanks for the feedback!
Speaker(s): Melchior Mattens
For non-life insurance companies operating in agriculture, property insurance or other insurance fields which are exposed to weather conditions, climate change is a potential threat to their businesses. In this thesis, four different weather conditions, which have a serious impact on the property insurance business, have also been recognized to be affected by climate change. For instance, the worst case weather conditions including wind, hail, rainfall and lightning might become significantly worse compared to the estimated worst case scenarios now. Moreover, the changing climate might increase the annual aggregate loss of the insurance businesses in general, as the prevalence of storms (wind, hail, rainfall and lightning) is likely to increase. Clearly, the changing underlying risk is important for actuarial valuation of the insurance policy. However, as insurance policies are sold in a highly competitive market, management pricing decisions might deviate from actuarial pricing methods given that they try to optimize the annual profit of the insurance company.
In this thesis, it’s been tried to identify the influence of relevant market conditions and of the underlying risk on future premium levels. In this thesis, a general claim size generating model for insurers is specified for the four aforementioned weather risks. This model incorporates likely trends in claim events, numbers and sizes. Next to this, a micro-economic market model, incorporating price elasticities and (marginal) costs per policy (for which actuarial valuation methods can be applied) is specified. From this, a profit optimizing strategy has been derived given the estimated (marginal) cost of a policy. Subsequently, the effect of solvency capital constraints (based on Solvency II regulation) and asymmetric information on the market premiums are considered. Based on the derived profit optimizing strategy given asymmetric information, the premium level distribution for premiums in 30 years is simulated for different initial market conditions. The relevant market conditions considered are related to the initial number of market participants and the assumed price elasticities. This gives a sensitivity analysis of the future premium levels to the chosen parameters. Besides future premium levels, the influence of initial market conditions and the influence of climate change on default probabilities are estimated.
Based on the aforementioned approach, the results presented in this thesis indicate that climate change can change the insurance market drastically. More dangerous weather events are not only likely to push up the marginal costs of insurers, but also increase the probability of default substantially. The latter might reduce the market competition incentive to insurers and as a consequence the premiums can go up even further. Estimates indicate that the average premium level in 2047 is expected to increase by an additional 9 to 17 percent (depending on the initial market conditions) solely due to the estimated climate change effect. Simulation results have shown that, due to solvency capital restrictions, the premium level in the different specified market conditions are expected to increase by 15 to 28 percent. The additional climate change effect therefore leads in some cases to an almost 45 percent increase of the premium in 30 years from now.
Likewise, default probabilities are likely to be strongly affected by climate change. In a market with many insurers, default will lead to less market competition, but more (geographical) diversification benefits for the remaining insurers. Nowadays, as many big insurers face strong profit reductions or even losses in their life-insurance portfolio due to longevity and market risk, heavy (tail) losses in non-life portfolios have to be covered by proper reinsurance. A stronger future dependence on reinsurance might make premium levels to go up even more than predicted in this thesis, as the relatively limited number of reinsurers worldwide and the high level of complexity of reinsurance optimization for normal insurers might lead to a lower level of competition in the reinsurance market.