Extreme weather events have cost the insurance industry billions of dollars in losses, and amid one of the busiest Atlantic hurricane seasons on record, there are fears that another major catastrophic event could send the real estate market into a tailspin.

Catastrophe models are a common scapegoat in a challenging real estate market. While they provide insurers with an analytical means of assessing risk, over-reliance on catastrophe models can lead insurers to ignore unique property-specific risk factors that these assessments may not capture.

However, one expert seeks to clear up misconceptions about transportation models. “The model is the starting point for carrier pricing,” explained Bruce Norris (pictured), executive vice president of national equity practice at GenCap Group. Norris pointed out that while models provide a basis for pricing risk, real-world conditions and constraints significantly influence final pricing decisions.

“A carrier is pooling the capacity it has in an area,” he said. “Let’s say you have $100 million to sell in a particular county or ZIP code, and it’s priced at $95 million because values ​​are going up, and the carrier will increase its price.”

Why are disaster models misunderstood?

Catastrophe models provide insurance companies with a sophisticated means of assessing risk. By simulating different catastrophe scenarios, these models enable insurance companies to more accurately estimate potential losses.

This enhanced risk assessment allows insurance companies to set premiums that more accurately reflect the risk profile of the properties they cover, ensuring financial stability and protection from bankruptcy.

But despite their complexity, disaster models are not infallible. They rely on historical data and assumptions, which may not accurately predict future events.

Uncertainty in model inputs, such as climate change and evolving land use patterns, can lead to significant discrepancies between predicted and actual losses. This uncertainty can lead to over- or underestimation of risks, affecting the pricing of insurance premiums.

But how much influence do cost models have on insurers’ pricing decisions? While the models provide a baseline, insurers adjust them based on their understanding of the market and risks, Norris said.

“Carriers can also override certain features in the model,” Norris added.

Factors affecting cat models and distinctive prices

The critical factor in this pricing equation is the balance between supply and demand. When carriers approach the limits of their capacity, they adjust prices upward, regardless of what the models indicate.

“The model doesn’t affect the price at that point, all they need to do is offer a minimum premium,” Norris said.

Other factors, such as the cost of capital and reinsurance costs, also affect the financial health and capacity of carriers, and thus affect pricing decisions.

Data quality is critical when it comes to disaster models. Norris said the industry is moving toward better data quality to assess increasingly complex and volatile disaster risks. He also highlighted the importance of secondary characteristics in the data, which can impact model results and how risks are perceived and priced.

Why should brokers care about cat models?

By understanding the model’s assumptions and outcomes, brokers can better prepare for discussions with carriers.

“If you can leverage the CAT model to understand the risks, which is kind of a ‘pre-underwriting’ before you go to the underwriting community, you’ll be in a better position to negotiate,” Norris said.

Brokers are also increasingly using models for their valuations, helping clients manage and understand their risks. Norris offered some advice: “The main way to get the most out of catastrophe modeling is to make sure your data is correct. If you send incorrect data to a carrier, it undermines trust and causes the carrier to adopt a more conservative pricing approach.

Always check your data using online tools. Additionally, including secondary characteristics in your data can greatly improve the accuracy of the model. When discussing discrepancies with the carrier, you can highlight the secondary characteristics used to explain the differences.

“Don’t send data randomly; use the model to anticipate the carrier’s point of view.”

What are your thoughts on disaster models? Have something to say about this topic? Please leave a comment below.