Posted On: July 23, 2007, 1:02 pm CDT

Posted By: DWT

Having worked with scoring, I can tell you that there are definitely two sides to this story.

First, Scoring is a valid predictor. But it is just that, a predictor. Not everyone with eight moving violations is going to have an accident. The same is true for those individuals with low scores… not everyone is going to have an accident. And while we can easily lump those people with multiple violations into the "Bad Risk" group, the same can not be said for those with bad credit scores.

Part of the problem is that we understand moving violations. Those individuals got caught by a law enforcement agency and got a ticket. We don't understand credit scores. We don't know what goes into a credit score, we don't know how to improve our credit score and we sure don't understand how there can be a relationship between credit scores and driving patterns.

To make matters worse, there are three primary algorithms used to generate a credit score and each of these algorithms look at the information in the credit history differently. It makes it even more difficult to explain credit scoring when a person scores 667 in one company and 845 in another (which isn't as far fetched as it may sound).

So yes, credit scoring is a valid tool that helps lower premiums for more people than it raises premiums for. However I believe that even though we can prove the benefits of scoring, unless we as an industry begin to use the same scoring algorithms and can help our insured's understand what helps and hurts their scores, it will be a tool that ultimately will be eliminated.