London Underground Ltd. v Susan Patricia Edwards

This Court held that a tribunal must adopt a flexible, fact-sensitive approach when deciding whether the proportion of women who can comply with a neutral requirement is "considerably smaller" than the proportion of men.

The determination is not confined to a mechanical percentage comparison. Tribunals may take account of the appropriate pool, the size and composition of that pool, contextual industrial factors and reliable common-knowledge statistics where relevant.

The claimant alleged indirect sex discrimination under the Sex Discrimination Act 1975 after new rostering arrangements made it effectively impossible for her, a lone mother and train operator, to continue in employment. The Industrial Tribunal found for the claimant but was criticised on appeal for selecting an incorrect comparative pool. The Employment Appeal Tribunal upheld the tribunal's ultimate finding and this Court was asked to review whether a finding that 95.2% of women could comply versus 100% of men was enough to be "considerably smaller" within s.1(1)(b)(i). The central question was the correct legal test for disparate impact and the role of contextual factors in small or unbalanced workforces.

1. Disposition. Appeal dismissed. The Court upheld the EAT and Industrial Tribunal that indirect discrimination had been established on the facts.
2. Pool identification. The proper comparative pool is the sector of the workforce affected by the requirement. In this case the pool was all existing train operators to whom the new rostering applied, not all LU employees or external applicants (Potter LJ, Conclusions).
3. Meaning of "considerably smaller". The phrase requires a flexible, fact-sensitive inquiry rather than a fixed percentage rule. It denotes a substantial and not merely marginal disparity, but context matters (Potter LJ, Conclusions).
4. Relevant contextual factors. A tribunal may assess the reliability of small-sample figures by reference to: the absolute size and sex composition of the pool; the industrial background; any employer measures that recognise a likely adverse effect; and accepted common-knowledge statistics relevant to the issue. Such factors may show a small percentage difference nonetheless amounts to a substantial disparate impact (Potter LJ; Simon Brown LJ).
5. Application to the facts. Although 20 of 21 women (95.2%) could comply and all 2,023 men (100%) could comply, the disparity was properly found to be "considerably smaller". The small number of women in the pool, the absence of any affected men, and the broader context of lone parenthood justified the tribunal's finding.
6. Standard of review. Appellate intervention is limited. A court will only overturn a tribunal's assessment if the conclusion is contrary to the statute or no reasonable tribunal could have reached it. That standard was not met here.
7. Practical guidance. Courts should avoid imposing rigid numerical thresholds. Tribunals must identify the correct pool, assess percentages in context and may rely on industrial expertise and reliable background statistics when evaluating reliability and significance of small differences.

• Employment Appeal Tribunal: Appeal dismissed; decision of the Industrial Tribunal was remitted for reconsideration on pool issue (see London Underground Ltd -v- Edwards 1995 IRLR 355).
• Industrial Tribunal (first instance): Found unlawful indirect sex discrimination (decision dated 17 November 1995).