Text size
Insurers using a common replacement value model would have been off by about $13,000 when trying to predict the value of a 2017 Toyota RAV4.
Dreamstime
About the author: Marty Ellingsworth is executive managing director of Global Insurance Intelligence at J.D. Power.
The core measure of insurance industry profitability recently hit its all time worst level, according to industry data. Auto insurance loss ratios reached 112 in 2022, S&P Global Market Intelligence reported. It means for every dollar insurers collected in premium, they paid out $1.12 in claims and other expenses. Worse, that inauspicious milestone was hit despite chronic rate increases as profitability has continued to erode.
There are lots of reasons for insurers’ profitability problems. Inflation, persistent supply chain disruptions, increased collision severity, frequency and increased complexity of repairs, and prolonged state rate-taking approval processes are just a few of the factors at play. But the most potentially damaging change to the long-term viability of the auto insurance industry is inadequate use of data on vehicle valuation in pricing.
The central value proposition of an auto insurance provider is to sell a future promise based on careful analysis of risk and accurate forecasting of cost trends. Insurers have missed the mark wildly on the forecasting part of that equation for the better part of the past three years. And now they are paying the price in the form of outsized financial losses, declining customer loyalty and plummeting customer satisfaction scores.
Old-Fashioned Predictions Fall Flat
To put the problem in perspective, consider the historical trend in retail price depreciation for the most popular production vehicles in America. The rule of thumb is vehicles tend to lose about 20% of their value a year. For example, a 2017 Toyota RAV4 had an estimated cash value of about $30,000 when it was new. Insurers using a common replacement value model would have projected today’s cost for that vehicle at $10,000. They also would have been off by about $13,000. The real market value of a 2017 RAV4 today is roughly $23,000, and the gap between those two numbers is creating some major challenges for insurers.
Compounding the problem, these outmoded prediction methods not only missed the mark on valuation variability, they also lacked the granularity required to accurately assess vehicle costs in an increasingly individualized automobile marketplace. For the past decade or more, many brands of automakers have been expanding the number of options and special features they offer, creating incredibly wide variation in costs within a single model. Porsche is perhaps most famous for this, routinely selling cars that have upward of $20,000 to $60,000 in options beyond the base MSRP. Some vehicles—especially pick-ups and SUVs—can have a total MSRP 50% higher than a base MSRP value.
The industry’s age-old valuation methods are helping it live up to the caricature of “100 years of tradition unhampered by progress.” Standard valuation methods do not capture the level of detail needed to accurately gauge the detailed composition of each individual vehicle. That’s because they are built using “squish VIN” data, which is a shortened version of the 17-digit vehicle identification number. The squish VIN contains the baseline characteristics of the vehicle, such as make, model, year, and perhaps basic trim-level information. But for many vehicles it doesn’t provide enough detail for a truly accurate projection of a vehicle’s configuration. (Brands and models with no options available are an exception.) Today’s economic environment shows just how problematic reliance on squish VINs can be. Production has been constrained, and manufacturers skewed toward higher line trims with wider ranges of options, further ratcheting-up vehicle valuations above any base model value.
As a result, insurers have been left playing catch-up by raising premiums, cutting costs, shuttering acquisition funnels, and crossing their fingers that the last three years of retained used-vehicle values were just a fluke. Meanwhile, customer-satisfaction ratings are in a free fall. According to our recent U.S. Auto Insurance Study, overall customer satisfaction with auto insurers is down 12 points (on a 1,000-point scale) from 2022, and record numbers of auto insurance customers are currently shopping for new policies. These are big problems for an industry that lives and dies based on lifetime customer value.
Building a Variable Pricing Model
Many industry professionals thought this trend would be short-lived. That old saw about a new vehicle depreciating the moment you drive it off the lot did not become conventional wisdom by accident. As long as there has been a used-vehicle market, most vehicles depreciated in a pretty reliable pattern. Yet, here we are, three plus years past the start of the pandemic and used-vehicle prices are still defying gravity.
If there are still holdouts clinging to their spreadsheets and waiting for things to go back to normal, they may not last long enough in the business to see it happen. The auto industry is rapidly transforming, with more widespread introduction of EVs, plug-in hybrids, special limited-run models and other wild card variables coming to the table each day. Insurers need to get serious about using more precise, dynamic vehicle pricing models based on an individual VIN that includes individual features and packages, driver behavior, near-real-time replacement cost and other factors that go into the cost of the premium offered to a customer.
The industry has come a long way on personalization when it comes to digesting and analyzing customer data to create usage-based policies. We need to apply that same approach to the vehicles themselves. The alternative will be another few years of valuation guesswork followed by a long retirement spent wondering what went wrong.
Guest commentaries like this one are written by authors outside the Barron’s and MarketWatch newsroom. They reflect the perspective and opinions of the authors. Submit commentary proposals and other feedback to ideas@barrons.com.
Based in New York, Stephen Freeman is a Senior Editor at Trending Insurance News. Previously he has worked for Forbes and The Huffington Post. Steven is a graduate of Risk Management at the University of New York.