Usage-based insurance, and more generally Insurance Telematics, have been shouldered with tremendous expectations. Most of those expectations have thus far fallen short of the hyped promises to disrupt private passenger auto insurance, bring lower rates and personalized pricing to many drivers, and encourage safer driving behaviors. Never has a technology missed so many supposed “tipping points.”
At its basic, UBI is a segmentation and pricing strategy. UBI allows for additional data variables to be considered as part of the underwriting and pricing process. Intuitively, how, when and where a car is driven should matter. And they do. They matter more than credit and all other pricing variables in place today. They have interactions with age, marital status, and vehicle type, and likely negate violations, territory, and gender. The problem with using driving data for underwriting and pricing is that (1) driving data is not readily available and is hard to collect, (2) it takes a lot of data to know how to use it, and (3) the objectives of all the players flooding into the telematics ecosystem aren’t aligned.
Unlike all other rating variables in practice today, predictive driving data can’t be collected from the consumer via the application process, can’t be ordered from the Department of Motor Vehicles, and isn’t likely to be shared between insurers. Instead insurers are left to pursue their own costly individual data collection projects. They do this through pilots with employees and volunteer customers by distributing black boxes, OBD dongles, and smartphone apps. Unfortunately, insurers’ data collection goals are often derailed by the problems TSPs have managing the logistical, compatibility, web, and communication components of pilots. Pilots inherently lack scale and expensive per vehicle costs often upset CBA assumptions.
Pilots, by design, are short in duration and have a limited number of participants. The data collected during a pilot (or even several pilots) almost never is enough to build a predictive model. Adding to the difficulty is that the pilot volunteers are actually safer drivers (more on self-selection later) resulting in limited claims history around which to model the driving data. It’s one thing to say mileage matters, it’s another to use analytics to show that not every mile is equal in terms of risk, for example. The risk associated with any particular mile varies by road type, time of day, relative speed, and the frequency and magnitude of changes in speed. Insurers need lots of data to determine these risk factors.
UBI is segmentation strategy for insurers to find and acquire new customers who are safer drivers. From a pricing perspective, UBI success depends on granularity, variables, and interactions of the driving data. Actuaries love this stuff. Yet TSPs haven’t made it easy to collect the driving data – pilots are priced by the vehicle rather than by car years of data and get structured by the TSP’s contract rather than by the insurer’s objectives to understand acquisition, risk, and retention.
It’s not all bad though. The good news is that those consumers who are attracted to UBI are better risks for insurers; this self-selection bias is helping to minimize some of the problems noted above. Self-selection lift can pay for a UBI program. TSP consolidation is also helping to clean up the ecosystem. As smartphones become viable data collection platforms, as the OEMs realize the value of the data they’re sitting on, and as consumer telematics products emerge, the problems start to go away.
Today’s version of usage-based insurance appeals to roughly 15% of the market – those safe drivers who are overpaying and are willing to share their driving data to prove they deserve a lower price. More transparency and improved Insurer-to-Insured (I2I) coaching is needed to penetrate beyond the early adopters. UBI enables a unique opportunity for insurers to connect with their insured customers. Insurers need to leverage telematics to become the advocate for safety, security and overall in-vehicle experiences. I2I connectivity is critical for the success of insurance telematics.
About the author
Dave has led all aspects of the development, pilot, launch and roll out of a number of Usage-Based Insurance products for U.S. insurers. He believes today’s offerings provide just a glimpse of the segmentation power, adoption rates and connected technologies that will drive insurance telematics in the future. He also doesn’t think it’s as complicated as it’s been made out to be.
