This is an extended version of the article that appeared in the September 2015 issue of Investment Advisor.
Insurance is acknowledged to provide not just a valuable bulwark in case of loss, but also to allow clients to sleep at night. Peace of mind is worth a premium — but just how much of a premium? That depends on what kind of information is being used to determine how much the consumer pays.
If you’ve been following the headlines, you’ve no doubt noticed that Big Data is playing an increasing role in all sorts of things, from targeted ads on your Web browser to bargain — and maybe less-than-bargain — prices on everything from airplane tickets and hotels to tools and home furnishings.
Big Data also plays a role in how your clients’ car insurance policies are priced. Thanks to price optimization software, consumers’ shopping habits, marital status and other factors unrelated to their driving and safety record are now taken into account by a number of insurers in multiple states when pricing auto insurance premiums.
While the rumble over price optimization has been growing over the past couple of years, a fresh flurry of reports at the end of July highlighted a study by the Consumer Federation of America (CFA) that found that premiums on state-mandated liability coverage for single, separated, divorced and widowed individuals increased by an average of 20% at four of six major insurers.
Some might argue, as the Insurance Information Institute (III) does, that there has been an “absence of consumer complaints” about the use of price optimization, which relies on myriad data to predict not just how an individual might drive but also how much he or she might pay for insurance before reaching the tipping point and deciding to look elsewhere.
That absence of complaints could be simply because the public hasn’t quite realized how much data unrelated to their driving skills is being used to determine the premiums they’ll pay. According to a survey commissioned by CFA, there was “overwhelming consumer support” for insurers to emphasize “driving-related factors such as accidents, traffic violations and miles driven” when determining pricing.
People don’t like to be taken advantage of, and numerous reports have pointed out that if Big Data indicates a certain demographic is more inclined to be loyal customers (perhaps senior citizens who aren’t Web-savvy, or those singles, widows and divorcees who are busy dealing with a bunch of other tasks?), they’re most likely to get hit with rate increases simply because they don’t routinely go hunting for bargains. (Not for nothing is the use of marital status in pricing known as the “widow penalty.”)
Those consumers probably don’t think that’s fair — and neither does Bob Hunter, insurance director at CFA, who said in a statement, “It is terribly unfair and entirely illegal for insurance companies to vary premiums based on whether or not [individuals] are statistically likely to shop around.” In the wake of Florida’s decision to ban the practice, Hunter termed the use of price optimization “price gouging,” while Birny Birnbaum, executive director of the Center for Economic Justice, called the practice “Big Data run amok.”
Price optimization, the insurance industry hastens to point out, has no standard definition, so it’s difficult to pin down exactly what it is. The National Association of Insurance Commissioners (NAIC) said in an email to Investment Advisor that the lack of a “uniform definition of price optimization” is an issue it is dealing with, although it added, “It is expected that the NAIC will provide a definition of price optimization and adopt a recommendation to states about what is considered to be unfairly discriminatory.”
But when Indiana became the seventh state to ban the use of price optimization in determining premiums (the other six are California, Florida, Maryland, Ohio, Vermont and Washington; New York is investigating the practice), the Indiana Insurance Department (IID) described it as including the use of “[...] data collection and analysis to predict which consumers will accept higher rates without changing insurers and/or varying premiums based upon factors that are unrelated to risk of loss so that each insured is charged the highest price that the market will bear.”
In fact, one of CFA’s criticisms of price optimization is that until it came under scrutiny by consumer advocates, it was marketed by firms providing the software as a means of maximizing profits, not improving underwriting. After a presentation by software firm Earnix before NAIC last year about the firm’s price optimization tool, CFA submitted a “rebuttal” to NAIC that called the Earnix presentation “misleading” and said that it was “astonishing in its divergence from statements made in advertisements and documents produced by Earnix prior to March 17, 2014,” the date on which the presentation was made to NAIC.
In its rebuttal, CFA cited earlier Earnix ads and documentation and said, “Earnix has repeatedly touted its product as an advanced predictor of a customer’s likely reaction to price increases — what economists call ‘price elasticity of demand.’ Earnix even referred to price optimization as an ‘elasticity model.’ [...] When the company presented its product to regulators, however, elasticity was never mentioned. In the presentation the word ‘elasticity’ was replaced by the word ‘competitive,’ as if raising the rate for some increases competition.”
The insurance industry defends the practice. Robert Hartwig, president of III, said that the use of “price optimization [...] is completely consistent with actuarial ratemaking guidelines, with practices insurers have used historically, and [has] been approved by regulators and [is] consistent with promotion of rate stability over time.”
Indeed, the NAIC said in its email: “Insurers are saying they use ‘constrained’ price optimization, meaning the insurer caps the rate change for a customer so the price doesn’t change as much as it otherwise would. Constrained means the insurer changes the rate from the current rate toward the indicated rate, but not all the way to the indicated rate at once. So the rate change is phased in over time. The price optimization in this case is a measurement of how large of a change the policyholder would likely accept at one time without leaving the insurer. Regulators are still debating whether the use of price optimization to cap rate changes provides more benefit than harm to consumers.”
III’s Hartwig said that abolishing price optimization could threaten such consumer benefits as loyalty discounts. But in an NPR report, Hunter of CFA said, “They’ll give you a discount for loyalty. But they’ll give you a 10% discount after they’ve raised your rate 25%.”
Two class action suits have been filed over the issue, and it’s clear that the discussion is far from over. Whichever side of the matter you’re on, both insurance professionals and consumer advocates advise that if you’re not happy with your insurer, shop around. Advisors, you might want to see not only how long your clients have been with their present insurer, but also what kind of premium increases they’ve experienced, and make sure that they’re getting the best possible price for the coverage they need.