The Cheapest Car Insurance for Good Drivers
Table of contents
Warren Buffet’s success can largely be attributed to the insurance industry, the engine that consistently propelled the expansion of Berkshire Hathaway since 1967. The premise is quite simple. People pay insurance premiums which then get set aside into a pool of money called a “float” that insurance companies can then invest. That float is a free loan that can then be used to make investments with. It’s the profit on investing with the float that reaps large rewards for insurance companies. Of course, there’s also a need for smart policy underwriting which ensures that insurance companies take in more money than they pay out in claims.
We’ve talked before about How Technology Will Affect Big Insurance Companies in both positive and negative ways. While machine learning algorithms can help underwrite smarter policies, the auto insurance cash cow will all but dry up when self-driving cars get here. In the meantime, insure-tech startups are taking in loads of cash. One of them is Root Insurance.
About Root Insurance
Founded in 2015, Ohio startup Root Insurance has taken in $527.5 million in funding with $350 million of that closing just last month in the form of a Series E round. All that money is being used to build the first car insurance company that’s founded on fairness as opposed to demographics. That rectangular-shaped supercomputer that’s glued to your face during half your waking hours contains sufficient technology to measure your driving behavior – braking, speed of turns, driving times, and route consistency – to determine just how safe a driver you actually are. Everything happens in Root’s app – customizing and purchasing your policy, finding your insurance card, making changes to your policy, or filing a claim. Since each state differs in respect to insurance rules and regulations, the app isn’t offered in all states yet – but penetration is quite impressive.
As the famous American philosopher George Carlin once said, anybody driving slower than you is an idiot, and anyone going faster than you is a maniac. Root’s business model is all about not insuring the idiots and maniacs. Since the worst 30% of drivers cause nearly 45% of all accident costs, Root Insurance does not insure high-risk drivers. Period.
Old school ways of underwriting insurance policies use intuitive factors like age, previous history of traffic offenses, credit score, and of course the value and performance capabilities of the car being driven. (Nobody’s going to pull a Paul Walker in their 1982 Chevy Chevette.) Here’s how their pricing model compares to the typical methods used by auto insurance companies:
In order to obtain the driving score which Root uses to price your insurance policy, simply download their app and carry your phone with you when you drive like you always do. (You don’t need to have the app open and it’s also able to distinguish whether or not you’re a driver or a passenger using machine learning algorithms.) After several weeks of driving (it usually takes 2-3 weeks), the app makes a decision about whether or not to insure you and what the rate looks like. Root makes an interesting comment which discloses their use of machine learning to make the whole thing work better over time.
We do take into account several standard factors that are mathematically predictive of risk or fraud—it’s the responsible thing to do—but we’re committed to eliminating as many factors as we can as technology progresses and artificial intelligence gets smarter.
We’re also able to learn more about what factors Root takes into consideration when pricing a policy by looking at their explanation for why the price of a policy might change. (Every 6 months, your policy will automatically renew and the price may change.)
Additional Factors to Consider
Driving late at night is more dangerous, so the app takes into account the times of day you drive. The longer you’re on the road driving, the more likelihood of an accident, and similarly, certain routes are considered more dangerous than others. As you may already have guessed, if you change jobs or move houses then your rates will likely change. If you move to a place where few people get into accidents and not many cars are stolen, there’s a good chance your car insurance rate will go down. Changing your vehicle affects your policy, but the Root app makes it simple to check in real-time how your rate will be affected by a different vehicle. Then, there are the changes related to your environment that you’re unable to control.
Even if you’re the best driver in the world, someone can crash into you or jack your wheels. Consequently, if the rates for accident and theft go up or down in your area, your insurance rates will go up or down, too. Weather forecasts also come into play because driving in heavy rain or snow is just more dangerous. Another interesting factor is the cost of repairing cars around you. Root cites the example of electric cars which are – apparently – more expensive to repair. Lastly, the company itself will adjust rates just based on what their machine learning algorithms uncover as they commit to constantly providing the fairest rates possible.
Root Insurance and Claims
In China, when an accident takes place, some drivers quickly whip out their smartphones and start taking photos. The world’s biggest fintech company, Ant Financial, has built an AI-powered app that uses photos – and 46 patented technologies – to process claims from accident photos in seconds making it 176,000 more efficient than just using human claims adjusters. Root appears to be moving in this direction with their claims process also involving some photo-taking after which a human gets in touch with you regarding the results of the claim.
If you’re a Root user, you’re a safe driver, which means that having to make a claim is less likely. Still, the ability to easily make a claim and receive a payment is as equally important as a fairly priced insurance policy, if not more. That’s why it’s called insurance. One can expect that Root’s claim process will only become more streamlined over time as the AI algorithms learn more about how to fairly price claims. For fender-benders, it may be possible to automate much of the process such that payments for claims can be issued in hours, not weeks.
Conclusion
The extent to which technology is transforming the insurance industry is being seen around the globe. Recently, Tesla announced their intention to start offering auto insurance because their cars are automatically generating every single bit of data needed to appropriately price policies in much the same way Root does. (Apparently, Warren Buffet doesn’t think it’s such a good idea.) Perhaps in anticipation of that, Root began offering their Tesla customers an added discount a few years back that particularly rewards highway driving. Stratifying clients into various buckets according to their risk levels makes pricing more “fair,” but also means that high-risk drivers will pay more – and they should. They’re the ones the majority of us are subsidizing at the moment.
We’re currently holding two insurance companies in our DGI portfolio that have increased dividends for 38 and 27 years respectively. To see all 30 DGI stocks we’re holding, and to learn how to build your own DGI portfolio, sign up for a Nanalyze Premium annual subscription.
Sign up to our newsletter to get more of our great research delivered straight to your inbox!
Nanalyze Weekly includes useful insights written by our team of underpaid MBAs, research on new disruptive technology stocks flying under the radar, and summaries of our recent research. Always 100% free.
There’s a major flaw with how they compute risk. A “safe” driver who drives the speed limit and brakes softly may still be a high risk because they nearly sideswipe cars in the next lane because they don’t look, or they don’t look for oncoming cars as they turn into a street. I see these circumstances often. Speeders get a bad rap but what if they haven’t had an accident for 15 years? Why are they high risk? This insurance company probably will go out of business because they will learn they only insure the highest risk drivers at a low price.
Good question, but they’re learning a lot from the 10 billion miles of driving data they’ve amassed so far.