Automobiles are so solidly made now that they’re safer and lasting far longer than their predecessors. Their durability is delaying wider use of even better technology, however, a trend that will leave auto insurers with a broader range of risks to cover, experts concluded at a recent conference.

“We’re going to have various autonomous vehicles that can do different things in different ways,” Thomas Karol, general counsel for the National Association of Mutual Insurance Companies, said during a panel discussion at the Casualty Actuarial Society’s 2018 Ratemaking and Product Management seminar in Chicago. “That’s probably going be the biggest hassle in terms of underwriting.”

11.7 and Counting

“Vehicles right now are about 11.7 years old,” said Chris Mullen, director of State Farm’s technology research division.

With vehicles lasting that long, “it takes a long time for the vehicle fleet to turn over, even in a good economy,” Mullen said. “It takes a long time for technologies to permeate the fleet, and when they perform as designed, the vehicles are leaving the fleet less quickly.”

“Cars are more robust than they used to be,” said panelist Matt Moore, senior vice president for the Highway Loss Data Institute (HLDI). “Cars don’t rust very much anymore, and we don’t have problems with them wearing out.”

“Advanced technologies decelerate turnover in vehicle fleets,” he said. “We did a study of vehicles with electronic stability control and we determined they crash out of the fleet much less frequently than vehicles not equipped with electronic stability control.”

Moore added that turnover of America’s vehicle fleet is also impeded by “a ton of debt” carried by car owners.

“Folks have car loans in excess of six years,” he said. “People’s current debt load hampers their ability to roll over debt to purchase incrementally more expensive vehicles.”

(Dis)Connected vehicles

For auto insurers, a prolonged turnover in the nation’s vehicle stock means insurers will be faced with a widening range of risks with vastly different levels of automated safety equipment and monitoring devices. At the very least, the variety of vehicle capabilities on the road will complicate the task of synchronizing traffic flow.

“We talk about these trends as if there would be one type of vehicle,” said panelist Thomas Karol, general counsel for the National Association of Mutual Insurance Companies (NAMIC). “In fact, we’re going to have multiple models with various qualities and abilities.

“It’s not going to be one-size-fits-all,” he continued.

Mullen, of State Farm, said that the different vehicle systems offer some unprecedented challenges for drivers.

“[A motorist] might have one operating system on his or her car, but the guy next to them has a different system,” Mullen said. “[A household] might have one brand new car, but the car next to it in the garage may be four years old, and may not have the same system. Then suppose they rent a car and have yet another system to use.

“Are motorists ready for differences in systems and vehicle performance?”

Moore of HLDI added that “when you’ve got different cars doing different things, it creates a real problem. There’s no standardization in terms of how people get warned” about road hazards.

Arizona Fatality

The panel met just days after an automated vehicle operated by Uber in a test run struck and killed a pedestrian in Arizona, an event that left panelists wondering if the fatality would stall progress toward automated vehicles.

“This event may be something that slows things down,” said Karol of NAMIC. “In the short run, six months to maybe a year, people are going to have second thoughts.

“The fact that [the automated Uber car] didn’t respond any better than a person is the question,” Karol added later in the session. “The whole idea behind autonomous or assisted driving systems is that they are supposed to be an improvement over human drivers.

“If they’re not an improvement, why invest all this money into it? The fact that it didn’t do any worse than a human is not the standard we want.”

Moore of HLDI noted that injury to pedestrians was a particular concern for autonomous vehicles.

“Pedestrians are vulnerable, hard for cars to detect, and not necessarily compatible with the current suite of vehicle sensors,” he said. “From a [public] policy perspective, we really need to think about protecting pedestrians from autonomous vehicles.”

The public will probably see more accidents involving autonomous vehicles as the technology is developed and tested, said Alex Timm, CEO and co-founder of Root, a startup auto insurer that rates coverage solely through driving performance recorded on a smart phone app.

“When we launch new technologies, these sorts of events (the Arizona fatality) are almost expected and should be planned for,” he said. “Honestly, I think this is probably the first of more to come.”

But Timm said that improvements in autonomous vehicle safety “is probably going to move faster than what most people think.”

Moving On

As challenging as the transition to driverless vehicles may be, panelists had little doubt that it would happen and be well-advanced by the year 2028.

“Look at Silicon Valley and the amount of money [invested] in new start-ups and new technology firms,” Timm said. “Some of the best minds in the world are [moving] into this space, and I don’t know how it doesn’t get a lot better real quickly.”

As a result, Timm believes that “within 10 years, you’re going to see the cost of auto insurance decline, and you may see the [auto insurance] industry contract somewhat.

“I’m already seeing a lot of insurance companies, especially multi-line insurance companies, shift more into other lines of business,” he added. “I don’t think that auto insurance is going away, but I do think the average price [of a policy] will go down, which will put further pressure on expense ratios, agency distribution models, and billion-dollar-plus advertising budgets for mandatory products.”

Personal auto insurance may shrink, but it won’t disappear.

“As long as a human driver has some sort of control [over a vehicle], there will probably be some attendant liability,” said NAMIC’s Karol. “As long as there’s some attendant liability, there probably should be some personal insurance for that.”