Big Data, Better Rates: Why Current Car Insurance Rate Calculations are Unfair

Ryan Steward
6 min readJan 5, 2017

--

Shopping for auto insurance can be a lengthy process. Comparing rates from insurers and evaluating whether you’re getting a good price can be more complicated than it should be. Different companies may seem to be offering the same thing at vastly different prices, but why? And even if you think you’ve found a deal, traditional insurance companies may simply raise their rates over time so you ultimately end up paying more. How can you be sure you’re actually getting a fair rate?

Traditional car insurance prices all start at the same place: risk pools. Customers are categorized by demographics such as age and sex and then the company makes guesses based on statistics about their predicted risk of future accidents. That means the rates you get are only as good as whatever information the company chooses to include in your “profile”. And that data usually has very little to do with how you actually drive.

The price you pay is mostly based on the historical risk of other people who have similar profiles. Some of the risk factors make sense — like if you have previously filed several claims, you’re probably more likely to have one again in the future. Others, however, such as education or whether or not you drive an imported car, have less to do with your driving habits but still count toward your price. And the most useful measure of risk — your actual driving behavior — that’s hardly used by anyone!

So how has car insurance evolved over time? How have premiums for good drivers improved over the years, and what’s next for policy buyers in the auto insurance industry? Here are just a few areas where traditional insurance carriers are getting it wrong, and how modern, mobile-first carriers like Root are paving the way in offering drivers the best — and most fair — possible insurance and pricing.

Ways Traditional Car Insurance Fails: Redlining by Another Name

You probably wouldn’t be surprised to find out that the history of car insurance hasn’t always been one of fairness and good deals for consumers. But the ways in which consumers have often been let down by ineffective use of data and just plain unfair practices are still striking.

While the pricing of car insurance based on a person’s race is prohibited by law, a 2007 study of Los Angeles residents found that many black and Latino neighborhoods and neighborhoods with high levels of poverty were being overcharged on car insurance premiums, disproportionate to their actual risk factors. In one case, only 3% of the gap in insurance premiums offered between neighborhoods could be attributed to real differences in risk. In other words, many policy buyers were being charged unfair and egregious rates based on who they are and where they live, rather than how well they drive.

Where The Old Way Falls Short: Under-utilizing Big Data

Major car insurance carriers have access to vast quantities of information and computational power, used for the purpose of determining risk, coverage, and premiums. It’s a complex affair, involving figuring out how to condense a dozen or more different factors into a single price. But you might not realize just how much Big Data goes essentially unused — representing a missed opportunity for traditional carriers.

These providers use a number of different methods to compute rates for the various forms of coverage sold to drivers, but not all of these systems perform as optimally as they could. A recent study of the uses of data mining in auto insurance found that there are more accurate possible methods of identifying high-risk drivers and separating them from low-risk drivers, including one recent model incorporating 16 risk factors to provide extremely high accuracy in risk appraisal.

These mathematical techniques can help ensure that good drivers aren’t overpaying due to being mistakenly grouped with more dangerous ones.

Why Dongles Don’t Work

Some “cutting-edge”, digitally-enabled insurance companies are starting to take advantage of what modern technology can offer. These carriers are using telematics, sensor data recorded during driving, to try to incorporate actual driving behavior in their risk profiles. Many of them are offering their customers plug-in car devices known as dongles in return for discounts for better driving. The feedback offered by on-board dongles, with reports of data on a person’s safe driving behaviors, has been shown to reduce speeding, aggressive driving, and traffic accidents involving these drivers. More than that, the wider use of telematics and behavioral reports could lead to 12–18% fewer road accidents overall.

Yet these traditional insurance companies are still not making use of the full potential of telematics. Not all of these carriers’ customers will choose to use telematics devices, meaning you’re still paying for the risk of drivers profiled using lesser-quality data. Including non-participating drivers means that even users of telematics won’t fully benefit from their own better driving. With traditional dongles, the best drivers only save 5–15% off the average rate. If their low risk were fully taken into account, these drivers could actually deserve up to 40% off — yet they’re missing out on these savings because they’re forced to shoulder the burden of both bad drivers and the rest who simply aren’t using a dongle. With Root, you won’t miss out on the discount you deserve. Telematics is where every policy begins — meaning their customers receive all the savings they’ve earned from their good driving.

Where Root Does It Better: Savings for Good Drivers With Ultra-Personalized Premiums

The shortcomings of traditional car insurance aren’t inevitable. Today, there are easier ways for good drivers to get the premiums they deserve. Root is the first fully mobile car insurance company designed to fit your on-the-go lifestyle, insuring only good drivers in order to ensure the best rates. Our mobile-first approach fits your on-the-go lifestyle — no clunky proprietary devices are needed, only the app. Drive with Root onboard for two to three weeks, and we’ll deliver you a personalized quote based on your unique driving data. You can easily select the coverage you want to buy and pay immediately via your phone. It’s that simple.

Root is taking Big Data in car insurance to the next level. Download the Root — Car Insurance app on iTunes or visit JoinRoot.com to try us out for free today.

--

--