Zip code and credit are two huge factors for insurers.

The insurers don't know the race of people being insured, but those zipcodes have higher claims, that's why they pay more.

Claims, claims, claims, claims, population density, crime, tickets, rate of uninsured motorists are what the actuaries use.

But again claims are the biggest factor in rating up a zipcode.

I pay a lot for car insurance.

Fraudulent accidents and claims, carjackings, predatory ambulance chasers with garish advertising, car theft for illegal export to Latin America or the Carribean, drivers without licenses, drivers without insurance, etc. If I lived in Sioux Falls, South Dakota, I would be paying only a small fraction of the amount.

homeowners, SF, never a claim-one year $2,600.00 plus $600 a yr FEMA.

The lower the income, the more likely that there is no homeowners insurance bundled with the auto insurance. Both of these scenarios affect auto insurance rates.

It is not surprising that lower income correlates with higher premiums. Personally, I think that actuarial tables should be regulated so that lower income people get a break. Driving is not really optional for many. And if you drive the law requires you to buy insurance. So for starters, the law should ensure that auto insurance is affordable to full time workers who make the minimum wage.

The people across the street from me are black. For the same record and vehicle they would pay the same as I do. I own rental properties in poorer, higher black population, higher crime neighborhoods (higher is relative btw, we're not talking about the old Cabrini Green here). My white tenants with the same record and vehicle would pay more (this I know for a fact as I lived briefly in one of my rentals myself during a delayed move) What causes the increase is the neighborhood and its crime/claim rate, not the driver's race. We see blacks paying more because they are more likely to live in higher crime thus higher claim neighborhoods. Insurance is a brutally actuarial thus factual system. More likelihood of claims = higher rates and vice versa. They do not, and should not, care how nice or how oppressed or how melanin-endowed you are. How many cars are stolen where you live? Vandalized? Accident rate? Do you have garages or driveways? These are the factors utterly unrelated to the insured which determine rates.

That has to be the first question, and the researchers want a political sideshow after pointing out a correlation and suggesting we infer a causal connection.

I'm white. I live in a mostly black or at least mostly minority neighborhood. (Which it is depends on where they draw the line.) I pay higher insurance because of race discrimination--presumably anti-white? Nah. Unclear I even pay more. My neighborhood may be mostly black, but the correlation is that residents of mostly black neighborhoods on average pay more, not that black residents pay more or every mostly black neighborhood pays more.

At times my insurance rates have been higher. When single, they were higher. Before I bought a house, they were higher. When I was newly insured--not just as a teen, but also in my 30s after being carless for a few years--my rates were higher. Before I could easily go around to lower-cost insurance agencies, they were higher--some agencies have restrictions on who they cover based on age, insurance history, claims. When I was a teen and in college, I got lower rates because I was a good student and had good attendance--two things with a racial correlation. If I were pulled over a lot I'd have more points on my insurance record and pay higher rates. If I had more claims--this "study" includes liability only--I'd pay more.

Blacks ... more often single, fewer are house owners, more often let insurance lapse, have higher rate of stops, often don't quality because of these things for lower-cost carriers, etc., etc. However, these things aren't granted lower insurance rates because they're nice, but because they correlate statistically with fewer claims. Nobody said, "Gee, you've been uninsured and that means you have higher insurance rates--we intend for that to punish blacks, but you're white ... well, what the heck, we'll give you the lower rates." No, it was, "You've been uninsured for a few years and that puts you in a higher risk risk-pool, so you pay for that higher risk." The guy on the phone didn't see me.

(As for the reliability of this study, a recent study posted at NBER was dismissed as utterly worthly partly because it wasn't peer reviewed. Notice, this will have nobody arguing that it's meaningless because it's not peer reviewed. It fits the general confirmation bias.)