Be wary of jumping to easy assumptions about how people behave.

There have been some emotive headlines recently about people with a poor credit history having to pay more for their motor insurance. Insurers using credit scoring to determine motor premiums are working to the assumption that those with poor credit scores are less likely to maintain their vehicle than those with good credit scores. There could be a nugget of truth in that, but just how big a nugget? Assumptions about how people behave when faced with difficult choices are not always reliable.

I was once spent four years in charge of a sizeable UK motor portfolio. When I joined, the key measure of customer service was the number of seconds it took to answer telephone calls from the policyholder. This was based upon the assumption that a quick response to a policyholder’s call delivered a better service. There was a nugget of truth in that assumption, but, as it turned out, only a small nugget, for this portfolio had some unique characteristics. A survey was undertaken of policyholders’ priorities for customer service. Time to answer telephone calls came in as priority 21. Priorities 1 and 2 involved performance measures that had never previously been tracked, yet when people thought about it, those two priorities made perfect sense.

The lesson to learn is to only draw conclusions from firm premises. One seemingly logical connection does not lead to one seemingly sensible conclusion, for there could well be other factors involved, some of which may not be obvious at first glance. For example, just as some people with poor credit scores are likely to reduce maintenance spend on their vehicles, other such people are likely to reduce their use of the vehicle, which would improve the risk they present to motor insurers.

So which set of behaviours by people with poor credit scores should insurers take to be most prevalent? I’ve no idea, but I know someone who does know! Policyholders with poor credit scores of course. Ask them: conduct a survey and get some definitive evidence of overall trends.

Yet with combined ratios for motor portfolios not looking that healthy, could such an expense be justified? Why not rely on research that others may have already conducted into the behaviours of such policyholders? A word of warning. Be wary of credit agencies bearing gifts of previously conducted research. How arms length was such research conducted? How objectively was the research framed? There could well be some conflicts of interest lurking around, about as material I suspect as research by turkeys into Christmas eating habits.

As insurers gather together an increasing number of external datasets about their policyholders, they need to develop more sophisticated ways of assessing them than what appeats to be the present day preference for causal assumptions. Successful businesses are able to apply critical thinking to their product pricing and service performance, in order to avoid dead ends and identify the substantive opportunities . The motor sector’s combined ratio would be the better for it.