The European Court of Human Rights has made no such ruling. You are confusing it with the European Courts of Justice.

The ECHR is a court established after WWII to rule on human rights in countries that are signatories to the European Declaration of Human Rights.

The ECJ is a court established as part of the European Union to adjudicate on legal matters that relate to EU law.

The ruling you refer to was a ruling of the ECJ, not the ECHR. Further, the ruling was over whether member states of the EU can have an indefinite derogation from applying the 2004 directive that gender discrimination in the provision of financial services should be forbidden. The ECJ ruled that an indefinate derogation was unlawful, and member states must actually apply the directive that they have agreed to apply.

It is the directive that requires gender discrimination be outlawed in the provision of financial services. It was agreed by the decision making bodies of the EU way back in 2004.

As a moral point, we abhore gender discrimination in all circumstances save where it is necessary. So it should only permitted in financial services if it is truly necessary. It isn't necessary, and so it should not be allowed.

Richard is correct about the legals - and it's worth pointing out that while you write "Insurance companies’ differential premiums are based on evidence about quantified risks, not prejudice.", insurance companies are very good at hiding exactly how they calculate their premiums; it is difficult to say that the huge gulf of differences between the premium for a 17 year old male and a 17 year old female is solely down to a similar gulf in the risk.

It's also a problem that insurance companies could (and should) base their premiums on much better data rather than using sex as a shorthand.

Also, the shoe size idea would be illegal as it is indirect discrimination.

In terms of insurance premiums, gender is simply a convenient surrogate for a complex combination of social and physiological issues that affect some people more than others. Rather than trying to identify these issues and incorporate them into their models, insurers have been allowed to rely instead on the inappropriately blunt instrument that is gender simply because it’s far easier to do so. In the case of car insurance premiums for the under-25’s you can think of there being two groups of young people: those that are high-risk and those that are not. The problem is that a greater proportion of young men are in the high risk group and so gender presents itself as a very convenient surrogate for the high-risk category. However, because of their gender, perfectly sensible young male drivers are being tarred with the same brush as the higher risk group whilst high risk young women continue to enjoy lower premiums at the expense of their male counterparts. There is absolutely no statistical evidence to suggest that being male actually causes a driver to have more accidents and it would be ludicrous to suggest such a thing. Yet this is what the insurers would have had to prove if they wanted to retain their exemption.

By making it illegal to set premiums on the basis of gender, the European Court of Justice forces insurers to look again at their pricing policies and to identify what the real factors are that determine whether a young individual is high risk or not. In the meantime, the premiums for young women will increase, but at least sensible young male drivers will get a fairer deal and, for the first time, the inequity that arises from the existence of the relatively small group of high-risk drivers will be shared equally across everyone regardless of their gender.

Far from being anti-science, this ruling is a challenge to insurers to improve their models for the benefit of all and, as a statistician, I welcome the challenges that this decision throws up. How will we identify the real factors that affect claim-risk and will we even be allowed to ask the relevant questions?

----
Steve Brooks is Executive Director of ATASS Select (www.atass-select.co.uk), former professor of statistics at the University of Cambridge and a chartered statistician. ATASS Select welcomes enquiries from the insurance industry and the media on this and any other related issue.

I cannot see how the ruling is anti-science.

The Court tested the validity of an article of a directive against rights guaranteed under the European Convention for the Protection of Human Rights and Fundamental Freedoms. It found the article wanting.

The ruling is, if "anti" anything, anti-discrimination. Each of us has these human rights as individuals. The question then is whether or not there are circumstances in which an individual's rights may be diminished solely because of the statistical behaviour of a group to which they might belong.

Yes, it's a bad mistake to misidentify the court. It suggests that you haven't actually read the judgement on which you are commenting - hardly a "scientific" approach.
The task of ther European Court of Justice ("ECJ") is to say how the law of the European Union applies to the case in front of it. There is no European Union law that requires the court to be "pro-science" (whatever "antiscientific" or "proscientific" means in this context). In the Belgian Consumers Association case (the one you are discussing), the court ruled that Article 5(2) of the 2004 EU Directive on equal treatment in goods and services (which allows insurers to charge premiums based on gender-related risk) must lapse on 21st December 2012. The court based its conclusion – which is not supported anywhere in the express wording of Article 5(2) – on the need to comply with Articles 21 and 23 of the Charter of Fundamental Rights of the European Union, which require equality between men and women in all areas ("areas" meaning fileds of activity, not geographical areas).
However, the Lisbon Treaty of 2009 gave the UK an opt-out from the Charter of Fundamental Rights to prevent any court – either the ECJ or a UK court – from declaring UK law inconsistent with the Charter. It is, therefore, at least arguable that the judgement in the Belgian Consumers Association case – which was referred to the ECJ by the Belgian administrative court - does not apply in the UK.
Furthertmore, if the judgement is unpalatable to the UK, it must surely be unpalatable also to Belgium, France, Germany, etc. - the ECJ acknowledged that gender-related premiums are common practice. There is an obvious solution: make common cause with our partners and amend the Lisbon Treaty at the next opportunity, in any event before 21st December 2012.

Is there any evidence that women are better drivers, ie less accidents per mile driven, or do the statistics reflect the fact that a higher proportion of women drivers drive less frequently than men?

A more transparent insurance industry would give us more confidence about whether the differential insurance policies were warranted. There will be a number of factors that determine how risky a driver is and it won't be practical for the insurance companies to capture all of them. There policies must be based on an impartial assessment of the information at their disposal.

Setting aside the detail of which court is which, would I be right in stating that in some EU countries it is the insurance practice to insure the car and NOT the person driving it ?
Would I also be right in saying insurance companies in the UK have two licks of the spoon in that firstly they assess the car for Insurance Grouping and then asses the driver's age and nothing else, ie almost ignoring clean driving record.
Would I also be right in that UK insurance companies deliiberately default on claims made agsint them ? For instance, a woman reverses into a parked car. Her contract is for her insurance company to pay for any damage she does to a third party, ie the pasrked car. The insurance company refuses to pay for the repair and instead 'writes off' the 3rd party's car at the DVLA, for a token amount, forcing the the 3rd party to spend more money than the compensation in buying a replacement car.
That's not insurance - thats a rip off ! !.