New underwriting models could mean new revenue streams for auto insurers.
Reports of the demise of the auto insurance business due to driverless vehicles may have been premature. Nevertheless, as more autonomous cars speed along U.S. roadways, the insurance industry will need to quickly adjust its underwriting practices for auto insurance.
A 2015 white paper by KMPG predicts that the personal auto insurance sector could shrink by 40% over the next 25 years, with loss costs potentially plummeting to $50 billion by 2040. The forecast reflects the dilemma that the industry confronts: while these vehicles will purportedly result in fewer accidents and claims, this also means that insurers will need to write fewer premiums, losing revenue in the process.
Accenture, meanwhile, takes a less dire view of driverless car’s impact on the auto insurance industry, estimating that autonomous cars will create $81 billion in new premiums by 2025. It adds total premium losses will not outpace anticipated gains until 2050, even as auto insurance premiums start to decline after 2026. Insurers who adapt to the oncoming reality of autonomous vehicles stand to benefit the most.
New Risks to Underwrite
Traditionally, auto insurers have based rates on driver demographics (age, education) and behavior (driving record, claim history). More recently, insurers have honed this system by using data on individual driver patterns collected on telematic devices placed in cars.
Driverless vehicles completely flip that model. Instead of writing policies hinging on driver performance and personal statistics, insurers will switch to a product liability framework, insuring the complex hardware and software that literally drive autonomous vehicles. If any of those systems fail, original equipment manufacturers (OEMs) and other participants in the supply chain will require liability coverage, which could foster collaboration between OEMs and insurers. Accenture suggests that revenues from product liability coverage could reach as much as $2.5 billion.
As with regular cars, autonomous vehicles will need insurance against theft, vandalism, and weather-related damage. Unlike them, however, they’ll also need protection from hackers, who could take control of the car and request ransom. Covering those exposures with cyber security insurance could expand the insurance industry’s coffers by $12 billion, Accenture estimates.
Autonomous vehicles run on an interconnected, cloud-based infrastructure to operate the car and manage traffic flow via external sensors and signals. Unfortunately, if any of those systems malfunction, so does the car. Insurers have another opportunity to step in here and insure those hazards, leading to $500 million in revenues, according to Accenture.
Insurers can also translate their personal auto insurance expertise to the commercial vehicle segment. On-demand ride-sharing services such as Uber and Lyft need to insure their fleets, as well as provide per-trip insurance to riders.
It’s All About the Data
In 2015, KPMG surveyed insurers regarding the future effects of driverless vehicles, with a majority saying they would be felt as soon as the next decade. More than a third of respondents expected to increase their focus on personal auto lines, while 48% said their commercial auto business would be unchanged.
One UK-based insurance startup has already ventured into driverless car coverage. Adrian Flux Insurance Services currently offers autonomous car owners policies to cover mechanical failures in addition to cyber attacks.
However, the insurance industry needs more data to correctly underwrite all the potential risks associated with driverless autos on a large scale. Without a long claims history or plentiful performance data, insurers lack sufficient information to forecast — and underwrite — future risks.
As Bryant Walker Smith, law professor at the University of South Carolina, told NPR.org, there currently isn’t enough case law being decided to determine rates and the technology is still new. “Insurance is a data-based effort to really predict the future based on the past,” he stated, “and when you have dramatically different technologies and new applications for automated driving, it makes predicting the future much harder because you don't have those reliable data about the past and present.”
Yet once insurers collect that data and create new underwriting models, the industry will be able to race into all the opportunities driverless cars generate. A new revenue stream will ultimately replace the outdated model — just as autonomous cars will soon overtake driver-operated vehicles.