Yesterday’s ECJ ruling outlawing discrimination on grounds of gender in insurance has been greeted with howls of protest. Â (You’ll find a sample of 48 howls, albeit from IFAs and therefore exceptionally illiterate ones, here: Â http://www.moneymarketing.co.uk/protection/ecj-rules-insurers-cannot-price-on-gender/1026905.article.)
I suppose this negative reaction isn’t massively surprising. An awful lot of people don’t like the ECJ much at the best of times; an awful lot of people just don’t like change; and obviously there is clear evidence that women and men present different levels of risk in insurance, with, for example, women having fewer car accidents and tending to live longer. The howlers say it’s ridiculous not to take these facts into account.
However, I’m here to argue that the howlers are wrong, or at least out of line with the spirit of the times – which, when it comes to questions to do with social policy, comes to the same thing.
The thing is, there are all sorts of actuarial criteria which we could take into account in evaluating and/or pricing all sorts of risks. We haven’t ever taken all of them into account, and we haven’t ever taken none of them into account – we’re in the middle, taking some of them into account. When you’re in that large, grey, middle area, logic and rationality don’t apply – the only consistent and logical positions are at the two extremes. Instead, there are two not-so-logical considerations which determine which statistics we use, and which we ignore: a) how easy they are to obtain, and b) how acceptable they are to society at large. And the thing about both of them is that they can, and will, change over time.
We needn’t concern ourselves too much with the ease-of-obtaining issue now, although in the not-too-far-distant future it will be interesting to see what happens to the whole concept of pooling risk in life assurance when people can carry their entire genetic code around with them on a chip. It’s the acceptable-to-society-at-large issue that arises today.
So. How about race? There is evidence that people of different races live for shorter or longer lengths of time, and are more or less likely to have car accidents. Would you be happy to see those factors represented in product pricing? How about if a motor insurance company decided, on the basis of those factors, to demand much higher premiums from Jews, or Pakistanis, or Greeks? Or indeed to refuse to insure some racial groups altogether?
Or how about religion? I’m not sure if there’s evidence that followers of certain faiths live unusually long. But say that there is. Would you be happy to see methodists or Mormons offered lower annuity rates?
I can think of a hundred other examples, but I’m pretty sure these are the two most obviously unacceptable – so unacceptable, in fact, that I feel quite uncomfortable even suggesting them.
I suppose, though, that at some time in the past – or indeed even today in some other parts of the world – they’d be uncontroversial. They’d be seen as entirely sensible and reasonable underwriting criteria, just like age and postcode.My point is that what may appear to be an objective matter of actuarial calculation is actually something more complicated than that – an objective matter of actuarial calculation which takes place, or doesn’t take place, within a framework of societal values. As those values change, permissions to make – or act upon – those calculations can change too.
Rejecting sex discrimination is not just about acting on the premise that there are no differences between men and women. It’s often about saying even when there are differences, we refuse to allow our decision-making to be affected by them. To take the most obvious and I guess most undeniable example, women become pregnant and men don’t – but it would be a) wrong and b) illegal to take that fact into account in making employment decisions.
The thing is, either you believe that it’s wrong to make decisions on the basis of gender or you don’t. If you think it’s wrong, then it’s always wrong. These days, most of us do think it’s wrong, and that’s why the ECJ is right to rule against it in insurance. The evidence that women drive better and live longer is absolutely undeniable, but also absolutely irrelevant.
And now we’ve cleared that one up, we can sit back and wait for the next change in the rules of the societal permissions game. I confidently predict it’ll be on an issue that makes this gender business look like a tea-party: age.