A car crash judgment

March has been a busy month at the ECJ, with several cases of note passing through '“ but there is one that stands out: the well-publicised and highly controversial decision on discrimination in insurance in Case C-236/09 Test-Achats (see solicitorsjournal.com, 1 March 2011).

It is, in one sense, a mark of distinction for a judgment to be the subject of a critical editorial in The Times, and Test-Achats earned that distinction. The ECJ held that '“ from 21 December 2012 '“ it will no longer be possible to permit insurers to take sex into account when fixing premiums. The result, it seems, will generally be to increase premiums for women and as a result the papers were full of comments from drivers whose premiums are set to increase.

We have become accustomed to criticism of European decisions; and sceptics are aware that they are not always well informed. Unfortunately, in this case, the criticisms do indeed turn out to be amply justified. For Test-Achats is a highly unsatisfactory decision, not merely in its results, but because of its meagre and inadequate reasoning.

In 2004, the EU adopted a directive 'to lay down a framework for combating discrimination based on sex in access to and supply of goods and services'. The basic principle of the directive was that there was to be no 'direct discrimination based on sex, including less favourable treatment of women for reasons of pregnancy and maternity', nor any indirect discrimination.

The directive dealt specifically with insurance premiums. It provided that member states should ensure that, from 21 December 2007, 'the use of sex as a factor in the calculation of premiums and benefits'¦ shall not result in differences in individuals' premiums and benefits'. This is insufficient reasoning: if the use of sex as a 'factor' is not to result in any 'differences', then it is not possible to use sex as a factor. What the legislation was really saying was that sex was not to be used as a factor in calculating premiums and benefits.

Crunching numbers

There was, however, an exception. Member states were allowed to permit sex to be used as a factor if doing so was 'based on relevant and accurate actuarial and statistical data'. In other words, if there was reliable data which showed '“ for instance '“ that men are more likely to have motor accidents than women, then member states could permit that to be used. They had to review their decision after five years.

What the ECJ has in effect held is that when that five-year period expires member states are obliged to force insurers to disregard sex, even when relevant and reliable actuarial and statistical data shows that it is relevant.

The key problem is that prohibition on discrimination requires that comparable situations are not treated differently. But, as the council pointed out, there is a respectable argument that if objectively reliable data shows that men and women do not present actuarially comparable situations, then non-discrimination does not require one to pretend that they do. If men are statistically less careful drivers than women, why should female drivers subsidise male drivers? The discrimination here is not truly based on sex, but on objective characteristics for which sex is '“ objectively '“ a reasonably reliable predictor: a tendency to drive carelessly.

The ECJ rejected that argument in one brusque paragraph. But it gave no coherent reason for doing so. It simply said that the directive assumed that, in principle, the position of men and women in an insurance situation was comparable, and that 'accordingly' based on its own 'premiss' the legislation invalidated itself.

Read as a whole, however, the legislation did not assume this 'premiss'. Its starting point was that sex should be irrelevant to insurance, but it recognised that there could be situations where it was legitimate to treatit as relevant. The ECJ simply cherry-picked the first principle and ignored the second.

By doing so it relieved itself of any obligation to explain why the legislator had been wrong to suppose that there could be at least some cases where men and women are not, for the purposes of insurance, in a comparable situation. It just claimed that the legislator had decided there were not such situations, as a 'premiss' when (1) no such decision had been made and (2), even if it had been, that could not have been decisive for the validity of the legislation. For one would still have to ask whether the premiss was correct; and this the ECJ did not do.

There are arguments which might support the result that was reached. Advocate General Kokott mentioned some of them in her opinion. She recognised that there was a financial price to be paid for refusing to consider differences in sex, even where they are objectively speaking relevant; she thought that on grounds of human dignity that was a price worth paying.

Such arguments raise policy issues, on which reasonable people may disagree. But whatever one's views on those issues, it is depressing that the ECJ avoided them, preferring to use a feeble technical argument to pretend that they had been answered by the legislation itself.