The insurance industry claims, through statiscal analysis, that there is a strong relationship between credit scores and accidents, that it is one of the strongest correlations they have. The link between both of these may not be as simple and, in fact, dependent upon each other. Why using credit scores may not make much sense is described below.
First, there is a possibility, and I'm speaking from personal experience, that filing a claim, which results from something unexpected, like an accident, may itself be a cause of a lower credit score. When I had a car accident, my $1000 car was wrecked and so I had several new, unexpected expenses like deductibles, buying a new car, registration, sales tax, renting a car, towing and storage, and having the second car I bought in a hurry break down. I found I couldn't cover all expenses for those few months, including minimum payments for revolving accounts, and as a consequence, found myself using more credit. My credit score was dramatically lower after 3 months. It was the accident that caused a lower score, not a lower score that predicted risk. In fact, I had an excellent credit score and then I had an accident. Yet that will not count toward the statistics; it will be the after-accident credit score that would be correlated.
Second, a possiblity may exist that those with higher credit scores may file fewer claims. When making the decision whether or not one should file a claim or not, what would cost more, borrowing $350 to fix a windshield, or filing a claim and having an ncrease of $400 per year for the next five years. Borrowing $350 may not exist as an option for those with lower credit scores and thus can again, be a reason for the link between credit scores and filing claims.
Studies will need to be conducted to test these relationships. For example, let's say we take a certain date, like June 2006, and take a sample of those who filed a claim during that month for the first time. If information is available, then take the credit scores of all those people, and compare the correlation between both points. Was the credit score predictive? For the other relationship, we can send out a survey to a large sample of people and ask whether a filing a valid claim was avoided, the value of the claim, and the credit scores. Is there, in fact a relationship between avoiding filing a claim and credit scores?
We can get further insight about why this correlation exists. At this time, the use of baseless, sweeping, fallacious generalizations with many exceptions, to increase premiums is unethical and cheating many people of their hard-earned money. I already have two claims so I doubt the use of credit scores will cause a further increase in my policy, but I feel such a policy is obviously, without a doubt, wrong.
Here's another analogy, There is a link between lack of intake of Vitamin C and the risk of getting scurvy. Getting scurvy (filing a claim, for example) does not cause one to take less Vitamin C (lower credit score). It's taking less vitamin C that causes scurvy. It would not make sense to prevent one from getting scurvy so they could take more Vitamin C. My point is correlations don't indicate cause and effect in both directions or even in one direction all the time. It is this point that the insurance companies are so crassly ignoring.
