Recent figures from the Office for National Statistics show that the gender pay gap - the gap between average hourly earnings for full time male and female employees - was 19.5% in 2003.

The decision by the employment appeal tribunal (EAT) in the equal pay case of Home Office v Bailey & ors (2004, IRLR 921) will do little to reduce it. The women were backed by their union (PCS) who instructed Thompsons.

What were the facts?

About 2,000 administrative staff in the prison service decided to pursue equal pay claims with prison officers, or industrial or non-industrial support staff. The ongoing equal value claims for most claimants are being heard at tribunal level.

However, in this appeal - brought by two higher executive officers (HEOs), Ms Clemens and Mrs Pollak - the women tried to rely on a prison service job evaluation scheme dating back to 1996.

This had rated their jobs as equivalent with governor 4, governor 5 and principal officers. In 2000, the proportion of women HEOs was 50.7%. The composition of the comparator groups was overwhelmingly male.

The Home Office defended the women's applications by claiming that the difference in pay had nothing to do with sex discrimination (direct or indirect), so that they were not required, under the Equal Pay Act, to objectively justify any difference between the two groups.

The employment tribunal decided, on the evidence available (and in particular the gender breakdown of the two groups), that there was sex discrimination which required the Home Office to justify the difference in pay.

What was the issue for the EAT?

The EAT had to decide whether the tribunal had relied on the right evidence, and if not, to decide what evidence should have been used.

In the speech therapists' case of Enderby v Frenchay Health Authority (1993, IRLR 591), the European Court of Justice (ECJ) said, among other things, that if the disadvantaged group is predominantly female whereas the advantaged group is predominantly male, then there is a prima facie case of discrimination.

In R v Secretary of State for Employment ex parte Seymour-Smith and anor (1999, IRLR 253), the ECJ said that courts had to ascertain whether a 'considerably smaller' percentage of women than men were able to satisfy the disputed rule. It was ultimately up to the courts and tribunals to decide what evidence was relevant.

The employment tribunal had relied on the approach in Seymour-Smith because it had become the leading authority for the ECJ in deciding disparate impact on men and women.

As a result, the tribunal found that the proportion of women in the pool who were disadvantaged was 26%, whereas the proportion of disadvantaged men was 3.4%. It decided that there was discrimination and the employer therefore had to objectively justify that outcome.

What did the EAT decide?

The EAT, however, disagreed. It felt there was a clear difference between a situation in which there was a practice that presented a barrier to women becoming a member of a particular work group; and a situation in which the disparity in pay had arisen because of different arrangements for collective bargaining.

In the former it made sense to compare the different 'pools' of men and women who can satisfy the provision (as in Seymour-Smith). In the latter, however, what mattered was whether the claimant group was disproportionately female, as in the speech therapists' case.

The EAT concluded, therefore, that, where the advantaged group is predominately male and the composition of the disadvantaged group is neutral in gender terms, then the situation may not be fair but it is not automatically discriminatory on grounds of sex. 
The union has lodged an appeal against this decision to the Court of Appeal, which will be heard in March 2005.