Using Credit Data to Predict Insurance Loss
Individuals who take better care of their finances probably also take better care of their cars and homes.
One objection to the use of insurance scores in underwriting decisions goes something like this: “I can see why a score based on credit data can forecast credit performance, but I don't see how it can forecast insurance claim performance.” If you look at it from the perspective of responsibility, however, it is perhaps not surprising that poor credit management reflects a greater degree of claims risk, and vice versa. The analysis of millions of consumers’ credit histories and insurance losses shows that individuals who use credit wisely result in lower losses for insurance companies. This makes sense — individuals who take better care of their finances probably also take better care of their cars and homes.
That said, it is important to note that Fair Isaac does not claim that the predictive power of
the scores is based on a causal relationship (we’re not claiming a less than stellar credit
history causes you to file insurance claims), but rather that it is based on empirical correlation
(that people with below standard credit histories on average file more claims than people with good
credit history), and this correlation can be validated.
The consistently high performance of the
insurance score models has been validated many times, in fact, both by Fair Isaac and by independent entities.
The real proof, however, is that hundreds of leading insurers in the US and Canada who use Fair Isaac insurance scores continue to see improved results. Insurers wouldn’t use insurance scores if they didn’t work — and the vast majority of insurance companies are using credit-based insurance scores as one of the factors they evaluate in the underwriting decision.