Â鶹ԭ´´

With a “silver tsunami” of skilled underwriters eligible to retire in coming years and a shift in employee expectations, recruitment and retention is a critical focus for the insurance industry right now, said experts during a November webinar hosted by Reuters Events.

“Talent development, recruitment and retention is probably the No. 1 issue for the industry,” said Anthony Shapella, deputy chief underwriting officer at SiriusPoint. (Shapella was recently promoted to group chief underwriting officer, effective Jan. 1, 2025.)

The insurance industry is seeing 12-15 percent turnover—with double-digit turnover in underwriting for most organizations—and it’s taking longer to fill open positions, said Jessica Diffin, who leads talent for GRS North America at Liberty Mutual Insurance.

“It costs a whole heck of a lot more to get an underwriter today because of the tight competition,” Diffin said. “We’re chasing the same people, [and] it’s inflating salaries. If you add MGA startups to that, it’s compounding the salary inflation.”

Value proposition is key.

Diffin used the term “silver tsunami” to reference the impending retirements that have long been predicted. “We have a lot of people in underwriting that are retirement eligible. Play that against a younger workforce now dominating most of the available labor pool. They’re going to want some different things from us,” she said. “They’re thinking about what they want from work and how they want to work,” and talent strategies will need to be adapted.

“We really have to start with what is the value proposition,” Diffin said. “I believe that value proposition will be key for any employer who wants to attract, retain or build a pipeline.”

She advised that, for younger workers, “company purpose matters. They want to know who you are, what you’re about, and how they fit. They want to hear about things like social justice and climate change.”

Beyond that, Diffin said that millennial and Gen Z employees are looking for workplace flexibility and differentiated benefits such as mental health support and extended leaves.

She added that “mobility is super important to these folks… This is a generation that says, ‘I’m going to have seven, eight, nine jobs,’ and there’s no longer a stigma associated with job movement. They will go where they get mobility,” she said, noting that the average tenure is only about two-and-a-half years.

Insurance has a marketing problem.

When it comes to recruiting, Diffin believes the insurance industry has an image problem. “Younger workers don’t see insurance—let alone underwriting—as really a destination career, so that’s going to have to change,” she said.

Shapella agreed. “If you were to go out and talk to a teenager and say [the word] insurance, their eyes would glaze over. Insurance does not have a good reputation.”

He thinks younger people would be much more likely to join the industry if they knew how exciting it was. “No matter what’s going on in the world, if there’s a major event—it could be a plane crash or emerging technology with autonomous vehicles; it could be marine accidents—all these things that we’re reading about each and every day flow through the insurance industry.”

“For people that are curious and analytical, it’s a fabulous place to build a career,” he said.

Wendy Crosley, global director of underwriting automation and transformation at WTW, said that insurers also need to take a good look at their job postings and reconsider what those requirements look like. Remember that you’re trying to attract as many people as possible, she said.

Crosley noted, for example, that “women generally feel that they need to meet 100 percent of the criteria, when men usually apply after meeting only 60 percent.” When asked why, the women said that even though they felt they would be able to do the job, they thought applying was a waste of time because they wouldn’t even get a call back from the company if they didn’t meet all of the qualifications.

Another thing to consider is pay transparency, she said, as Gen Z is unlikely to apply for a job if the posting does not include an expected salary range.

What skills do underwriters need to cultivate?

More insurance companies are adopting artificial intelligence (AI) and large language models (LLM), which means underwriters need to upgrade their skill set to stay relevant.

Crosley said that one of the chief skills underwriters need to sharpen is data literacy—they need to be comfortable reading charts/graphs and interpreting the data if they’re going to gain insights.

They also need to be able to communicate effectively, she said, so they can share complex data and concepts with brokers and customers.

“It’s not just enough to say, ‘Well, the model told me that there’s a rate increase,'” she said. “I need to be able to articulate the why behind that and help my customer understand, ‘What does this mean for you?'”

Underwriters also need to be able to adapt and learn to trust the AI, Crosley said.

“I think one of the biggest skills we have to cultivate is having people be informed consumers of model output,” said Shapella. “We’ve used cat models for many years, and we know that cat models are not always accurate—in some regions they don’t model risk well at all.”

He noted that it’s also important to train people on how to use the right prompts for AI. “Sometimes it presents nonsensical results, and it often is depending on how you ask the question.” You need to know what question to ask to get the needed results, what data is needed to train the model, and then how to interpret the output, he said.

Shapella said it’s essential to train underwriters so that they understand how the model output “was derived, how to use it, what the shortcomings are, the pitfalls, so they’re able to sniff out the things that don’t make sense.”

“When you’ve been in the industry for many years and you see a model output that makes no sense, you know pretty quickly that it doesn’t pass the sniff test,” he said. But more junior employees “may not have that same nose…and we may run into the issue where people are blindly using model output.”

Shapella stressed the importance of training and transferring knowledge from senior employees down to the junior level.

Soft skills matter.

Matt Waters, head of U.S. Middle Market at AXA XL, talked about “the shift we’re seeing in the whole art and science debate. If you prove the science, then it shifts a little more onus on the EQ, on the art side of the business. And that is how you work with others, how you interpret data and analytics, and not just how you sell and market,” he said.

Justin Smith, chief economist at Applied Underwriters, agreed that soft skills are critical—especially if underwriters are going to stay ahead of actuaries impacted by AI, LLMs and other new technology that seems to be taking over the science of modeling.

“In a sense, I think actuaries have their toes closer to the impending flood than underwriters do,” Smith said. “If we can divide what actuaries and underwriters do, they’re both attempting to assess risk, but actuaries tend to do so in situations where they can find a lot of fairly homogeneous data and they can build classical models, and then they can present those results. The results are usually quite inarguable, which is why underwriters had to learn to kind of accept them and work around them.”

LLMs are going to perform similarly, he said, “except that what counts as an appreciable pool of data for an LLM is much smaller than what an actuary would consider a credible dataset, so it allows LLMs to reach into smaller, more niche questions.”

Smith believes that actuaries are more likely than underwriters to be impacted directly by AI “because an awful lot of what the actuaries do, the AI can do, too. So, I think underwriters should be looking over their shoulders at the wave of actuaries coming in to try and take over underwriting.”

He highlighted underwriting skills such as negotiation and the business of bargaining, noting that “these are skills which AI is not going to evolve anytime soon because you’re playing against a fundamentally human opponent—and one thing AI struggles to do is think reflexively.”

Check out the full webinar:

Topics Catastrophe Natural Disasters Talent Underwriting Training Development