Insurers increased their use of catastrophe models, drones and mobile apps during the COVID-19 pandemic, and they anticipate growth in such technology to continue once the health crisis passes.
Some growth in what is known as "insurtech" was driven by the social distancing measures and quarantines that accompanied the pandemic, while adoption of other technologies came despite it.
Edin Imsirovic, an associate director at insurance rating firm AM Best, said market pressure from the pandemic advanced innovation by a couple of years or so.
"Digitization in the insurance space sort of really accelerated this year due to COVID," Imsirovic said in an interview.
That might accurately describe some advances such as communicating virtually with policyholders, said Serge Gagarin, manager of segment marketing at catastrophe risk modeling firm AIR Worldwide. But "large-scale systems integration projects," big projects insurers might be undertaking, are developing independently of the pandemic, he said.
Technology growth hasn't always been a priority in the insurance industry. "We're a slow industry to adopt things sometimes, I dare say," said Don Griffin, vice president of personal lines for the American Property Casualty Insurance Association.
The industry has embraced catastrophe modeling for predicting the severity of events and for claims handling. Use of the technology was increasing before the pandemic, but it accelerated during quarantines for claims handling, experts say. This modeling gives agents an idea of how strong damage was at a location, said Tom Larsen, principal of industry solutions at CoreLogic. "Weather forensics" determine what happened at a specific location rather than half a mile away.
"The influence of [COVID-19] has been to accelerate this technology, which was sort of moving like a glacier," Larsen said. Using catastrophe modeling for claims is better for policyholders because insurers can handle claims more efficiently and at a lower cost, getting money to clients faster, he said.
Because insurers don't interact with policyholders often, they can differentiate themselves from other companies by how they respond to customers at times of need, Larsen said.
Karen Clark, CEO and co-founder of catastrophe modeling firm Karen Clark & Co., said catastrophe models can be used for projecting average claim severity in the two days before a hurricane makes landfall. Her company's data, which insurers use for planning, shows damage by zip code and updates twice per day. After a catastrophe, insurers need extra adjusters, so planning for the severity of claims can help them decide where to put them, she added.
Models are also used for fraud detection to see if claims for hail damage, for example, are coming from areas where storms didn't hit. Hail is "the main type of weather that causes insurance claims," Clark said, estimating it causes insurers more than $15 billion in damage on average each year, about a third of which is for commercial property and personal auto coverage and the rest for homeowners' coverage.
Modeling for catastrophes
The use of modeling for predicting claims stems from damage modeling, which was surging before the pandemic. The technology involves insurers submitting information to modeling firms that can run millions of scenarios to predict potential impacts from destructive events.
In the past several years, hurricanes have been more frequent and sometimes more severe. A record 30 named storms, 12 of which made landfall in the U.S., formed during last year's Atlantic hurricane season, according to the National Oceanic and Atmospheric Administration.
Despite the number of storms, insured losses were consistent with the long-term average based on catastrophe modeling, so using the models helped insurers be prepared, Gagarin said.
Models for flooding and wildfires that predict where damage might occur are on the rise.
The California Department of Insurance is considering how catastrophe models predict wildfire risk and whether they can be used in rate-making, the subject of a virtual public hearing last month.
Models are also a tool for insurers to help policyholders understand mitigation techniques and their impacts, said Jeff Waters, senior product manager for North Atlantic hurricane solutions at Risk Management Solutions Inc.
"The insurer could perform some sensitivity tests and get an idea of just how much of a benefit different mitigation strategies could have at the policyholder level," Waters said. His firm used event response tools to help insurers during last year's historic hurricane season.
A tool called HWind uses weather forecasting data and conveys the uncertainty about where a storm could go and how strong it could be, he said.
Along with catastrophe modeling, drone use is on the rise. It's a technology whose use increased as a result of social distancing.
Karen Collins, assistant vice president of personal lines at the American Property Casualty Insurance Association, said drones came into widespread use after Hurricane Harvey in 2017. According to Clark, they were even more valuable during the pandemic to help avoid in-person inspections, such as when adjusters climb on roofs to see damage.
"Certainly drone technology has been used a lot more," Clark said. "That was coming along anyway, but I know a lot of companies are subscribing to that so they can quickly survey particular houses in the impacted areas to see how many of their policyholders could have roof damage," which is the most common in a hurricane. She anticipates that trend will continue after the COVID-19 pandemic.
Looking forward, insurers see more technology growth.
One area is the expansion of phone apps, such as those for initial underwriting, Collins said. Some carriers were using them before the pandemic, and some adapted because of it, she said based on her own observations.
"The insurance industry is very open to embracing new technologies," Collins said. "Anything that is bringing efficiencies into play, and might even be cost savers that they can, in turn, pass savings on to policyholders to reduce rates, are certainly technologies that the industry's going to be very receptive to and not just turn off ... when the pandemic finally concludes."