In this second of three posts about the ethical dimensions of telematics, I’m going to sketch out an ethically positive development that should emerge from the widespread adoption of telematics. To do so, I’ll need to revisit some of the basics about insurance.
Insurance underwriting involves putting risks into categories and collecting information relevant to each category so that it can be properly underwritten. How such categories are set depends on the type of cover being provided and the availability, and cost, of the information needed to underwrite the risks. So back in the late 1980s when I managed a household scheme, having a universal buildings rate reduced the cost of collecting the information and created one overall ‘buildings’ risk category.
This process of categorisation is not always a neutral one when it comes to risk pricing. Insurers may favour category A over category B because it’s cheaper to collect information for A, even though B may be a better representation of the risk. This can lead to the true nature of some risks priced using category A being misrepresented and paying an unfair premium as a result. One of the ethical obligations of insurers is to keep the costs (social as well as financial) associated with such cases of mispricing to a minimum.
Over time, the cost of collecting information tends to fall and so in consequence can the size of categories. Sometimes these changes are smooth, sometimes not – the European Court of Justice ruling last year on gender pricing was rather a jolt to the pricing of motor insurance and resulted in gender categories having to be abandoned. In effect, motor insurers had relied on gender as a risk pricing category for too long, with the costs borne by policyholders who risks were being mispriced deemed to be no longer fair.
How are things like ‘too long’ and ‘fairness’ judged? Certainly in relation to the prevaling sentiments of society, but also in relation to the opportunities available to insurers to collect more and better data and revisit its pricing categories. In other words, if insurers can collect new data more cheaply than before, then the old categories have to go.
Clearly, back when motor quotes were done by hand, rating factors like gender, while still unfair and discriminatory, were tolerated because there were no better alternatives. But as alternatives did emerge, then that ‘no better alternatives’ defence began to look a bit threadbare.
It’s important to bear in mind though that there will always be some degree of unfairness in the pricing of risks. That’s because insurance will always have to rely to some degree on risk categories, which will always tend to fit some policyholders better than others. For example, while telematics will influence the rating of the damage and liability aspects of motor insurance, it will be less influential with regard to theft (due to jammers and the like). At the same time, categories are a natural adjunct of risk pooling, upon which insurance is founded. So while telematics could cut a swathe through many motor risk categories (for example gender, age and occupation), it will not remove them altogether.
A word of warning however. While telematics will help make motor underwriting fairer for more people, it will continue, quite rightly, to raise questions about privacy. Like many innovations, it has an upside as well as a downside. How that up and downside are ultimately balanced will largely depend on how the debate around its introduction is managed. The danger for insurers is that they keep that debate within the sector. The policyholder public will only listen to the ethical upside of telematics if they feel that their concerns about the ethical downsides are being listened to as well. Open debate allows the full story to emerge.