The EAT has ruled that an employer's use of length of service as a factor determining pay was not indirectly discriminatory on the facts of the case. A distribution analysis showing that women were clustered at the lower end of the relevant pay scales was not sufficient to require the employer to objectively justify its approach. The length of service factor did not put women at a particular disadvantage on the facts since the average pay of men and women within each payscale was not significantly different. 

The EAT also confirmed that the Armstrong defence is no longer available in an equal pay claim following the Supreme Court's decision in Essop/Naeem, although this part of the decision was not binding on future tribunals.

Background

n an equal pay claim, once a claimant establishes that their comparator does equal work, the employer must show what material factors explain the difference in pay. If the claimant shows that any factor relied upon puts women at a particular disadvantage, the employer must show that reliance on that factor is objectively justified.

Facts

In McNeil v HMRC the claimants were employed by HMRC within Grades 7 or 6 of HMRC's pay scale and claimed equal pay with longer-serving, higher-paid men in the same grade. HMRC accepted that the claimants did work which was equal to that of their comparators (it had been rated as equivalent under a job evaluation scheme). However, HMRC denied a breach of equal pay law.

A number of factors accounted for the pay differences between the claimants and their comparators but the dispute centred on length of service, which was one of the factors. HMRC argued that the impact of length of service on women within Grades 7 and 6 was marginal. The mean average differentials in pay varied between 1.2% and 3.3%. The claimants accepted that average pay statistics did not show significant differences, but relied on a particular type of distribution analysis which revealed a clustering of women towards the lower end of the pay scales and men towards the upper end. They argued that this was sufficient to demonstrate that length of service caused particular disadvantage to women, and therefore that HMRC had to show objective justification for using length of service. They originally also argued that women were disadvantaged by the length of service factor due to late career starts and career breaks, but by the time of the ET and EAT hearings they had abandoned this claim and were relying purely on the statistical disparity revealed by the distribution analysis.

EAT decision

The EAT ruled that the distribution analysis on its own might indicate a problem to be investigated, but was not enough to establish that women were particularly disadvantaged by the length of service factor. The claimants needed to show disparity in pay terms, not in terms of abstract distribution. The length of service factor did not put women at a particular disadvantage on the facts of this case since the average basic pay figures showed no significant long-term differences between the basic pay of men and women in either of the two grades.

Although this point was academic given the finding about particular disadvantage, the EAT considered that whether HMRC could have relied on the so-called Armstrong defence if the claimants had shown that length of service caused particular disadvantage. According to this defence, an employer can avoid the need to objectively justify the use of a particular factor if it can show a non-discriminatory underlying reason for any disparity. However, the EAT took the view that Armstrong was no longer good law following Essop/Naeem where the Supreme Court ruled that there is no need to establish why a protected group is disadvantaged.

Comment

The EAT's view on the Armstrong defence is technically not binding on future tribunals as the decision that particular disadvantage had not been shown was sufficient to decide the case. If the EAT's view is followed in other cases, however, it means that, in a case where a pay factor can be shown to put women at a particular disadvantage, the employer cannot undermine the claim by establishing a non-discriminatory reason for the disadvantage. Instead, employers must be ready to show objective justification for any statistically significant disparity.

However, this decision means that employers do have to show objective justification for pay policies or practices purely on the basis of the type of distribution analysis relied on in this case and that, if relying on statistics, claimants need to compare the average pay of men with the average pay of women who are doing equal work. Even then, it will be open to the employer to argue that the difference is not sufficiently marked or is co-incidental.

HMRC may well have been able to show objective justification for using length of service as a factor, since the courts normally accept that it serves a legitimate purpose. However, there can be issues with demonstrating that its usage is proportionate, especially when used to progress employees through long pay scales.