Does Canada have enough hurricane data for parametric solutions?

Thermal image of a large hurricane atop a digital map of the world. In the middle of the digital map, there is a warning on the digital interface which reads "extreme weather on earth".

When it comes to assessing hurricane risk exposure, there’s a data gap between the U.S and Canada, said a panellist in a Canadian Underwriter webinar. 

“There’s definitely a gap between the U.S. and Canada,” said Carl Bernier, NatCat specialist at Swiss Re. “I guess the fact that the U.S., which has been more affected historically by storms, might play a role in there, while in Canada it’s a little bit newer.” 

However, climate change is fuelling stronger and more frequent storms, as sea levels rise and ocean surface temperatures warm. “That also means storms that can travel further north, and that can maintain their intensity,” Bernier added.  

Cat modelling in Canada may also be underutilized, he suggested. 

“If we want to offer good coverage, if we want to close the gap, we first need to acknowledge that hurricane risk is material for Canada,” he said. “We need to model it, we need to consider it in risk assessments. These cat models exist for Canada. They are there, they’re not necessarily used, but they are there.” 

The lower frequency of hurricanes in Canada, and consequently less hurricane data, means that in this market certain insurance solutions that rely on historical data remain in relative infancy. 

For example, parametric insurance (a type of policy where payout is automatically triggered by a parameter or set of parameters, and tailored by an underwriter to an insured’s risks following a major event) is more widely available in the U.S. because their hurricane data dates back farther.  

See also  Evolution of Data Extraction: From Optical Character Recognition (OCR) to Intelligent Document Processing (IDP)

Parametric insurance works in cases where there are physically quantifiable events, said Megan Linkin, senior parametric NatCat structurer at Swiss Re Corporate Solutions. 

“There needs to be a sufficient historical time series, so going back multiple years to multiple decades of the aforementioned event, to be able to perform a probability analysis and determine…the likelihood of a given event of a given intensity,” she said. 

When creating a parametric policy for a client, insurers and insureds decide on a reputable third party who provides sufficient data in order to agree on the payout triggers. 

“Because you are settling on event intensity, that is reported by a reputable third party. In the case of Canadian hurricanes, and if [the policy] was covering the Maritimes, it would be the National Hurricane Centre who monitor storms as they move up into Canada,” said Linkin.

“The intensity that the storm has to reach before a payout begins is very clearly outlined in the policy, and both the insured and the insurer agreed to these terms prior to entering the policy.” 

Compared to the U.S., Canada does not have as robust a national flood map system, although Bernier said the system isn’t perfect in the U.S. either.  

“In the U.S., we have FEMA [Federal Emergency Management Agency] mandated by the government to develop these,” he said. “So I think there’s a little bit more awareness with respect to flood risk.” 

 

Feature image by iStock.com/da-kuk