Thursday, September 22, 2022

How do insurers identify good drivers?

The previous post reminded me of the importance of Credit Scores as a screening device.  Here is a post from a couple of years ago.  And I am still curious to hear about unique uses of credit history (see question at end). 

The FTC's Bureau of Economics relased their FACTA study, which concludes that:
  1. Credit scores effectively predict ... the total cost of [auto insurance] claims.
  2. Credit scores permit insurers to evaluate risk with greater accuracy, which may make them more willing to offer insurance to higher-risk consumers ... . [note: this is why you can call up GEICO, let them look at your credit report, and get an auto insurance quote over the phone].
  3. a group, African-Americans and Hispanics tend to have lower scores than non-Hispanic whites and Asians.
  4. ...scores effectively predict risk of claims within racial and ethnic groups.
  5. The Commission could not develop an alternative scoring model that would continue to predict risk effectively, yet decrease the differences in scores among racial and ethnic groups.
So even though credit scores help insurance companies price insurance more accurately, point 3 implies that some groups pay more, on average, than others. The policy issue behind the study is whether the government ought to ban the use of credit history for anything but making loans. As point 4 implies, banning the use of credit scores would result in higher prices for good drivers, regardless of their race or ethnicity.

Theory tells us that in states which ban the use of credit scores to price insurance (California and Massassachusetts) insurance companies would find it more costly to distinguish high from low risks, so they may lump them together (called "pooling"), and price insurance at the average risk. Or they may be concerned that only high risks would be willing to buy high-priced insurance (what economists call "adverse selection") and price high or, if price controls prevent high prices, exit the market.

I would be curious if any of our readers know of novel uses of credit scores as a screening mechanism, or if they have developed better predictors (point 5) in a particular application, like pricing insurance or screening job applicants.


  1. When I started at Gibson Guitar in 1995, I met a seasoned salesman named Ben Rhodes. Ben said he always knew if a music store dealer would be approved for credit or not. I asked how he knew. He said "After I talk to the owner a bit, I ask to use the bathroom. If the bathroom is clean, the credit will be good. And if the bathroom is dirty, the credit will be bad. There are people in the world who take care of things, and people in the world who don't take care of things. Someone who keeps their bathroom clean will pay there bills."

    I later learned that his method was as good as any!

  2. And if they use "there" in place of "their", demand a full background check before granting credit!