Home > credit scores, discrimination, economics, rant > Predatory credit score-based insurance fees

Predatory credit score-based insurance fees

April 13, 2015

I’ve been looking into who uses credit scores – FICO scores or other alternative scores – and I’ve found that the insurance industry is a major user.

Homeowners insurance rates, for example, varies wildly by state depending on what kind of credit score you have, often more than doubling for people with poor credit versus people with excellent credit. This is in spite of the fact that homeowners insurance applies not to the payments of mortgages but rather to the contents of an apartment or home.

Similarly, auto insurance rates vary by credit score, even though someone with a poor credit score isn’t obviously a bad driver. For example, in Maryland, people with bad credit scores can be charged 40% more just for having bad credit scores.

Statistics like this make me wonder, how much of this price discrimination comes from the insurance companies trying to understand and account for actual risk, and how much comes from their understanding that poorer people have fewer options and will simply pay predatory rates?

And just in case you’re a believer in free markets and fair competition, and think such predatory behavior would be whisked away in a competitive market, insurance companies actually target people who don’t shop around and charge them more. In other words, it’s not a free market if not everyone actually has good information.

Tell me if you have more examples like this, I’m a collector!

  1. Randy Nye
    April 13, 2015 at 7:38 am

    It’s probably worse than merely targeting the desperate and or gullible. Consider that the business of insurance enjoys broad exemption from antitrust laws.



  2. fxflinn
    April 13, 2015 at 11:52 am

    I have some experience in the homeowners space. The data shows there is a direct correlation between people’s credit scores and claims history. We were working towards a far more detailed method of rating policies until it became evident that credit scores were an effective proxy for a lot of other factors. The principals of bounded rationality have some application in this circumstance — adding more factors doesn’t improve the decision-making process. In the end we instituted a whole-home inspection (instead of a drive-by), and that allowed the underwriters enough information to decide the marginal cases effectively. Avoiding one big claim a year easily paid for the more rigorous inspections, and allowed us to offer more insurance for lower premiums. Of course, because we did not provide a market for low credit score owners with properties that inspected badly, we helped contribute to their higher costs of home insurance.


  3. captain obvious
    April 23, 2015 at 6:45 pm

    Whether using credit scores or not, the same sort of equilibrium economics apply.

    1. the whole industry will converge on the same predictors of risk (since all companies operate in the same legal environment, and there are a limited number of things they can ask about)

    2. the whole industry will converge on very similar models (since they are modeling almost the same data) using those predictors

    3. therefore any company can assess, using either their own model (or their own mock-up of “whatever everyone else uses”), how desirable a customer is to the competition. Everyone will know approximately which customers are likelier to be desperate, and can jack up the rates accordingly if they wish to.

    Maybe some companies will give up the premium of screwing the poorer customers, but if as a result they get too many risky accounts they will have to adjust the rates anyway, so it comes to the same thing as trying to charge because they can (but arriving at the equilibrium slower).

    This doesn’t explain the variation between states, but the dynamics in any large enough state market should show the above features.


  1. No trackbacks yet.
Comments are closed.
%d bloggers like this: