FSRA’s plans for auto coverage regulation

Ontario’s auto insurers shouldn’t expect changes to the ‘take all comers’ or Unfair or Deceptive Practices Act rules, Michelle Dodokin, head of auto insurance supervision at Ontario’s Financial Services Regulatory Authority (FSRA), told a breakout session at Monday’s 2024 FSRA Exchange.

“These are rules that will remain in place, and we’ll continue to supervise against them,” she said in response to an audience-member question at the end of the ‘Auto rate and underwriting regulation reform’ session.

A panel discussion preceding those comments focused on a planned shift from prescriptive to principles-based regulation for Ontario’s property and casualty insurers, and their likely impact on auto coverage.

Planned changes include an accreditation process for insurers that will require companies to demonstrate robust and consistent risk management practices. Accreditation also will include a testing process for firms that opt to apply.

“The idea here is that we’ll increase speed to market,” Dodokin said. “So accredited insurers, having demonstrated that they’re already operating in line with our principles and they’re serving target consumer outcomes, will have a streamlined process for applying for [things like] rate changes, product changes and model changes.”

Standing on principles

A shift to principles-based regulation will help define roles and responsibilities for various stakeholders in the P&C ecosystem, noted panellist Michelle Huang, director of model validation and model risk management at Discover Financial Services.

Once those are established, employees will have clearer direction about who’s tasked with what, she added, and new hires will have a better sense of their job responsibilities.

Principles-based environments can also foster innovation, added Roosevelt Mosley, principal and consulting actuary at Pinnacle Actuarial Services. Under prescriptive structures, he said, “the rules can’t develop fast enough to allow for innovation, so companies try to innovate within a set of rules that don’t apply in the current environment.”

For example, he pointed to vast differences in automobiles produced today, compared with those rolling off assembly lines 20 years ago. While new cars offer superior safety features and other enhancements, underlying regulatory structures governing auto insurance are in some cases a half-century old.

“There were a lot of insurtech companies that…were going to come in and innovate all of the existing companies out of business,” he said. “And what they realized is that when they came to insurance they had to deal with sometimes archaic, sometimes old rules that didn’t allow for the kind of innovation that they desired to do.”

Another goal for principles-based regulation is to improve consumer outcomes. One way to help that along is for data scientists and actuaries at P&C companies to determine whether their risk modelling includes biases.

“Bias can be introduced at any point in the predictive modelling lifecycle. So, there’s very important considerations that have to be made that might be specific to certain parts of your model risk management or your predictive modeling plan,” said Elizabeth Bellefleur-MacCaul, a senior advanced analytics professional at Definity Financial Corp.

“There might be considerations that you need to make at the planning phase, versus your data extraction or your preparation phase, and understanding, ‘Is my data inherently biased and if the inputs are biased, what are my model results going to look like?’”

Feature image by iStock.com/utah778