3 ms·
The more acurately the insurance can predict risk, the better. And if they calulate that some profession/gender/lifestyle/whatever carries a higher risk, it's o
by Silhouette 8y ago
The more acurately the insurance can predict risk, the better. And if they calulate that some profession/gender/lifestyle/whatever carries a higher risk, it's only fair that those people have to pay more, after all they also cost the insurance more.
This is the classic correlation/causation fallacy that is why sadly we have needed explicit anti-discrimination laws in other contexts.
When I added my partner to my car insurance policy, at a time before prohibiting the use of gender to determine premiums, our payments went down quite significantly despite now covering two drivers. When I asked why, they told me that statistically women are safer drivers than men, and since my wife had a good track record with no accidents, that meant our risk together was now lower.
The thing is, at that point I had been driving regularly for many years, while my partner had also passed a test many years earlier but had hardly driven since. We both had clean sheets, but as you'd expect given our vastly different levels of experience, she was not as safe a driver as I was and would have been the first to admit it. However, the insurer's questions hadn't identified any of this, and had reached an obviously absurd conclusion.
Even after the policy change, in practice I was almost always still driving anyway, so clearly whatever my level of risk was before, our combined level of risk was still similar afterwards. But again, nothing asked when we adjusted the cover or since would have picked that up.
This is the trouble with almost any profiling based on personal data, from insurance calculations to targeted police actions to screening job applicants: in principle, it might be a reasonable thing to do, but if your model doesn't properly incorporate all relevant facts, it can actually be worse than nothing because not only does it give an incorrect assessment, it also instils false confidence in that assessment.
- TheCoelacanth 8y agoYou are missing out on the additional information about you that they got when you added your partner, i.e. that you are in a long-term relationship. Married men are well-known to be lower-risk drivers than single men, presumably the insurer also knows that men in a relationship significant enough to buy insurance together are also lower risk. Also, someone who hardly ever drives is very safe from the insurer's perspective. They may not be very skilled, but they also don't have many opportunities to get in accidents.
- Silhouette 8y agoEven if that first part were true, we'd been together for a considerable time before she was interested enough in driving the car that we changed the insurance, so their risk analysis would still have been miles off for years.
- TheCoelacanth 8y agoBut they can't base insurance premiums off of information that they don't know. They can't just take your word for it that you are in a relationship, because that is too easy to lie about. Buying insurance together is a clear indicator that would take significant effort to game.
- Silhouette 8y agoBut they can't base insurance premiums off of information that they don't know. Of course. But as long as they can't form a sufficiently complete picture to make fair decisions -- that is, pricing based on actual risk -- discriminating on easier grounds that are correlated with risk but also happen to be incorrect in many cases isn't fair, so we make laws that stop them doing that.