Auto Insurance Growth Opportunity #4: Turn On All Driver Classifications
Many insurers turn off certain driver class segments. While this practice was effective in the past, it may limit future growth. With machine learning, the risk selection paradigm has changed, creating an opportunity to capture more profitable business within every driver class.
Driver Classifications Don’t Tell the Whole Story
Men, teenagers, singles and the elderly have all been characterized as risky drivers, but do these assumptions hold up under scrutiny?
Consider the following two drivers:
- Peter is an 18-year-old single male who has been driving for two years with no crashes or moving violations. He drives distraction free, stays within the speed limit, wears his seatbelt, and ensures passengers are also wearing their seatbelts. He avoids driving at night or when it’s raining.
- Becky is a 50-year-old married woman. She typically drives home late at night after a long shift at work and is often barely able to stay awake. On the weekends, she likes to meet up with friends for drinks, after which she drives home. She typically goes 10 to 15 miles over the speed limit, but she’s only received one speeding ticket.
Becky sounds like the riskier driver. However, many insurers might charge her less than Peter due to assumptions about her gender, age, and marital status.
Are Certain Groups Really Less Proficient at Driving?
There are statistics to support the theory that certain groups of people are more likely to get into collisions. Consider the following four groups, often associated with risky driving habits:
- Teen Drivers. According to the CDC, drivers between the ages of 16 and 19 have a fatal crash rate nearly three times as high as the fatal crash rate of older drivers per mile driven.
- Male Drivers. Among 16- to 19-year-olds, the CDC says male drivers are three times more likely to suffer fatal crashes than female drivers. Older male drivers are also more likely to be involved in fatal crashes. According to IIHS, the number of male crash deaths was more than twice as high as the number of female crash deaths nearly every year between 1975 and 2022.
- Elderly Drivers. The IIHS says the fatal crash rate per mile increases at around age 70. However, it’s important to note that older people generally drive fewer miles, meaning they’re involved in fewer police-reported crashes.
- Single Drivers. Research published in Injury Prevention found that people who have never been married have a significantly higher risk of driver injury than married people.
Do Segment-Based Insurance Decisions Still Make Sense?
According to Bankrate, male 18-year-olds pay an average of $6,712 a year for full auto insurance coverage. The average 50-year-old female only pays $2,116. Although the prices vary from one auto insurance company to another, it’s clear that insurers treat these groups very differently.
Based on generalized data, this may seem justified. In some cases, it’s easy to see why certain groups may be more prone to collisions. Teen drivers are a prime example: since they’re inexperienced behind the wheel, it’s no surprise they are involved in more crashes. On the other hand, older drivers have ample experience, but they may also have slower reflexes and poor vision. It’s harder to pinpoint why male and single drivers tend to be riskier, but, regardless of the reason, the data is pretty clear. For insurers, that’s often enough to justify group-based insurance rates.
However, it’s important to keep in mind that individuals may perform differently than the average of their group. In general, teen male drivers tend to be riskier than other drivers, but a particular teen male may be a safe driver. Likewise, 50-year-old married women tend to be safe drivers, but a specific 50-year-old married woman could be a menace behind the wheel.
If insurers could find a way to assess risk at an individual level, it would surely result in a much fairer process. It could also result in more premium volume. When insurers file super high rates for a segment, or worse yet, they turn a segment off, they miss out on all the policies within the segment that would have been rate-adequate.
Taking A More Personalized Risk Selection Approach
Until now, auto insurance risk selection has been segmentized, and based on just a handful of criteria. A person’s driving history is informative, but it may not tell the whole story. Driver classification based on demographics may point to a higher or lower level of risk, but this doesn’t hold true for every individual. As a result, the filed rates for each segment are only really fair and competitive for part of the drivers within the segment. If your rates are high, you lose out on a lot of potentially profitable business. If you shut the segment off, you miss out on even more.
How can auto insurers grow?
Soteris uses machine learning to evaluate dozens of risk characteristics – not just the driver classifications – for each applicant. The predictive platform then compares the likelihood of a claim to the insurer’s filed rate for the risk class to produce a personalized rate adequacy score for each individual policy, at point of sale in less than one second.
If the risk score falls within the insurer’s risk threshold, the application may qualify for straight-through processing. If the score is not within the risk threshold, the application is flagged for further review. Learn more.
Interested in Other Auto Insurance Growth Opportunities?
