MICHEL MARTIN, HOST:
If you want a car or drive one regularly, then you probably know this. Car insurance costs more for some drivers than for others. For example, women generally pay less than men. Married men generally pay less than single men. Insurance companies say that the price differences are due to the level of risk involved in insuring a driver. According to their algorithms, the people who pay less for car insurance do so because they have a lower risk of having accidents.
But a new study by Consumer Reports and ProPublica says that on average, drivers who live in white neighborhoods pay far less for car insurance than drivers who live in predominantly nonwhite neighborhoods. Price gaps that are larger than risk alone can account for - in some cases as much as 30 percent.
Here to tell us more about all this is ProPublica reporter Julia Angwin, who co-authored this study, and she's with us now. Julia, thanks so much for speaking with us.
JULIA ANGWIN: It's great to be here.
MARTIN: Try to walk me through the research here because I'm sure that people have many, many questions - that your research focused on a particular profile of a customer, a 30-year-old woman with a safe driving history. Why this particular customer?
ANGWIN: What we wanted to do was remove all of the factors that insurers use other than geography. So they use all sorts of factors about your driving record and your credit history sometimes, and how many miles you drive, what kind of car you have. And that's part of the calculation of rates. So we asked for quotes for this one woman in all zip codes across America. And then what we did was looked at the differences and compared them to the true risk, as far as we could tell, of what insurers really were suffering in terms of losses in those zip codes.
So what we found was basically that if you took a apples-to-apples a whiter neighborhood and a more minority neighborhood and said these are two that insurers on average have paid out the same amount in losses per car over the last five years, why are the premiums 30 percent higher in the minority neighborhood than in the white neighborhood? And so what we did was we looked at each of these two kind of comparable neighborhoods across a whole state and then we came up with an average disparity. So on average, for instance, the disparity between a minority neighborhood and a non-minority neighborhood for GEICO in Chicago was about 18 percent.
MARTIN: The California Department of Insurance did respond, though. They criticized the approach saying that, quote, "the study's flawed methodology results in a flawed conclusion." How do you respond to that?
ANGWIN: So what California actually brought us like a slightly more specific objection which is a fair one. So they said, look, you took all the insurers' payouts per zip code on average and compared that as your baseline for what is risk, but each different insurer may have a different rate of losses. That's true, but we can't see that number.
However, we did find there was one company, Nationwide, in California that for whatever reason actually filed its internal losses per zip code in one of their rate filings with the insurance commission. In that one case, we could test that assumption, and we found that the disparities between minority neighborhoods and nonminority neighborhoods of the same risk - when you looked at Nationwide's own actual risk, the disparities were higher than what we had found when we did it the other way.
MARTIN: Is there any explanation for this that makes sense?
ANGWIN: The one thing that kind of sticks out to me is that consistently when you looked at the comparison between premium and the true risk, you saw that in minority - predominantly minority neighborhoods, as the risk increased, the premiums increased. And it was pretty much a linear relationship like you would expect.
What we saw was in the whiter neighborhoods that that wasn't true, as the risk increased, the premiums actually flattened out and sometimes declined. And so what you basically have here is an unexplained discount that is being given to the highest risk, white neighborhoods.
MARTIN: That was Julia Angwin. She's a senior reporter at ProPublica. She's one of the co-authors of a study spotlighting disparities in automobile insurance premiums between residents of white neighborhoods and predominantly nonwhite neighborhoods, and she was kind enough to join us from our bureau in New York. Julia, thanks so much for speaking with us.
ANGWIN: Thank you. Transcript provided by NPR, Copyright NPR.