Indirect discrimination and the gendered realities of part-time work. Review of the Joined Cases IK (C-184/22) and CM (C-185/22) v KfH Kuratorium für Dialyse und Nierentransplantation eV,EU:C:2022:1011

3.1. The nature of the disadvantage

Concerning the disadvantageous treatment of part-time workers in terms of pay, the KfH Kuratorium case builds on the principles established in Lufthansa CityLine, where the facts of the case were very similar: the national court asked whether national legislation that requires both part-time and comparable full-time workers to exceed the same number of working hours in a given activity, such as a pilot's flight duty, to qualify for additional remuneration. ⁶ The CJEU ruled that such treatment must be considered as providing less favourable treatment to part-time workers under the meaning of Clause 4(1) of the Framework Agreement.

In the case at hand, part-time care assistants had to work the same number of hours as full-time care assistants to qualify for overtime pay, regardless of the normal working hours agreed in their individual employment contracts. This meant that part-time care assistants faced a significant disadvantage, as they were either unable to reach the required threshold or had a much lower probability of doing so compared to their full-time counterparts.

Although, on the surface, overtime pay appeared to be available equally to both part-time and full-time care assistants—since the right to additional pay only applied beyond the 38.5-hour weekly threshold—this uniform threshold placed a greater burden on part-time workers. Due to the shorter normal working hours in their contracts, any extra hours worked above their contracted hours but below the 38.5-hour threshold did not entitle them to an overtime supplement, even though they were still required to work beyond their agreed schedule.

In contrast, full-time care assistants received overtime pay immediately for every hour worked beyond their 38.5-hour weekly limit, whereas part-time care assistants received no additional compensation for exceeding their individually agreed working hours unless they surpassed the full-time threshold. As a result, the Court concluded that part-time care assistants who work extra hours without additional pay are treated less favourably than full-time care assistants, who are compensated for every hour beyond their standard workweek. This inequitable treatment echoes the Court's reasoning in Elsner-Lakeberg, ⁷ which recognised that applying the same overtime threshold to both full-time and part-time workers creates structural disadvantage for the latter.

3.2. Objective justification and its economic aspects

The concept of ‘objective grounds’ within Clause 4(1) of the Framework Agreement requires that any difference in treatment be justified by precise and specific factors related to the employment condition in question. These factors must be assessed within the relevant context and based on objective and transparent criteria to ensure that the differential treatment responds to a legitimate need, is appropriate for achieving the intended goal, and is necessary for that purpose. Justifications may arise from the particular nature of the work performed under part-time contracts, the inherent characteristics of those tasks, or the pursuit of a legitimate social policy objective by a Member State.

In the KfH Kuratorium case, the referring court questioned whether two specific objectives could serve as justification for differential treatment: first, the aim of discouraging employers from requiring excessive overtime beyond the agreed individual working hours, and second, the prevention of any perceived disadvantage to full-time workers in comparison to their part-time counterparts. However, the Court assessed these justifications purely from an economic standpoint—meaning that part-time workers encounter a financial loss—while failing to consider the broader social implications and the influence of gender dynamics on part-time employment. Moreover, it neglected to apply the ‘greater burden test’, as established in Elsner-Lakeberg and reaffirmed in Lufthansa CityLine, which evaluates whether a measure disproportionately impacts part-time workers.

Women are overrepresented among part-time workers in the European Union, with many working reduced hours to balance professional and family responsibilities. The structure of part-time employment reflects deep-seated gender roles, where caregiving responsibilities predominantly fall on women. Although women represent 46.6% of the workforce in the European Union, they do not have the same prestige on the labour market as men. Women are overrepresented among part-time workers and other atypical forms of work, and they are more likely than men to be hired for non-managerial jobs. Mothers of young children are even less valued in the labour market: only 62.7% of mothers with children aged six years and younger were employed in the EU-28 in 2019, compared to 89.1% of fathers with young children. ⁸ Gender-specific divisions of labour persist, with women taking primary responsibility for household duties: in 2018, more than a quarter of employed workers in the EU-28 adjusted their work hours to accommodate childcare responsibilities, with more than 80% of these workers being women. ⁹

However, the Court's decision in the KfH Kuratorium case failed to acknowledge these underlying social realities. For a part-time worker with caregiving responsibilities, working additional hours is significantly more burdensome than for a full-time employee. Many employees opt for part-time work precisely because they need to accommodate non-negotiable caregiving duties, such as picking up children from daycare at a specific time. An extra hour of work is also proportionately more disruptive for someone working only four or six hours a day than for a full-time employee working eight hours a day. Moreover, full-time employees are more likely to rely on external help, making additional work hours easier to absorb. Instead, the Court limited its reasoning to the financial implications of the overtime policy, focusing on the income loss of part-time workers, without acknowledging how indirect discrimination perpetuates traditional gender roles. While in Elsner-Lakeberg the Court recognised the structural disadvantage for part-time workers, the KfH Kuratorium judgment stopped short of explicitly linking this issue to gender disparities in employment.

Thus, the Court failed to recognise the intersectional nature of discrimination faced by female part-time workers. This oversight is particularly troubling in light of the recently adopted Pay Transparency Directive, which explicitly emphasises the need to account for intersectionality in cases of gender-based pay discrimination. ¹⁰ The Directive encourages national courts, equality bodies, and other authorities to fully consider intersectional disadvantages and integrate this approach into their practice, including recognising such forms of discrimination, assessing proportionality, and determining compensation or penalties.

4.1. Implementing the principle of non-discrimination in equal pay cases

On the other hand, the Court added that for an apparently neutral measure to be considered indirect discrimination under these provisions, it must, in practice, result in a particular disadvantage for individuals of one sex compared to the other. In this context, Recital 30 of Directive 2006/54 and the Court's case law establish that assessing whether indirect discrimination has occurred falls within the competence of national judicial authorities. This assessment must be carried out in accordance with national laws or practices, which may allow indirect discrimination to be demonstrated through various means, including statistical evidence, as referred to in earlier judgments like Schuch-Ghannadan ¹¹ and the case law cited therein.

This brings back into the spotlight the issue identified by Barnard as the tension between the ‘discrimination model’ and the ‘equal pay model’. ¹² Barnard argues that one line of the case law follows the discrimination model, suggesting that the claimant needs to show not only the three main criteria for equal pay claims, such as a comparator who is engaged in same or like work in the same establishment or service, but also that the adverse measure in question affects significantly higher percentage of women than men. The Court stated in Jämo that to determine whether the lower pay for midwives violates Article [157] of the Treaty and Directive [2006/54], the national court must assess whether statistical data show that a significantly higher proportion of midwives are women compared to men. If this is the case, the measure constitutes indirect sex discrimination, unless it can be objectively justified by factors unrelated to gender-based discrimination. ¹³

Another line of the case law, argues Barnard, implies that it is sufficient to trigger the application of Article 157 TFEU if the claimant demonstrates that a man and a woman are engaged factually in the same work but paid differently. ¹⁴ This approach was applied in Elsner-Lakeberg, where the CJEU stated that in line with its earlier case law, the principle of equal pay, as established in Article [157] and Directive [2006/54], prohibits any sex-based discrimination in remuneration for the same work or work of equal value, unless the difference in treatment is objectively justified by factors unrelated to sex or is necessary to achieve a legitimate objective. This position aligned with previous rulings, including Brunnhofer. ¹⁵

The different approaches of these models imply that the burden of proof on the employer varies. Under the equality model, employers must demonstrate that the pay difference is based on factors unrelated to sex, effectively allowing them to justify any pay differential without maintaining a clear distinction between direct and indirect discrimination. This can be problematic, as it blurs the line between direct discrimination, where justification is only permitted in cases of express derogations—which do not exist under Article 157—and indirect discrimination, which can be objectively justified. Consequently, the equality model suggests that both forms of discrimination can be objectively justified, thereby weakening the strict prohibition against direct discrimination.

4.2. Applying the notion of indirect discrimination in the case at hand

The Court in KfH Kuratorium concurred with the Advocate General's position that neither the definition of ‘indirect discrimination’ in Directive 2006/54 nor any other provisions of the Directive reference quantitative criteria in assessing indirect discrimination. The definition instead adopts a qualitative approach, focusing on whether the national measure in question is inherently liable to place individuals of one sex ‘at a particular disadvantage’ compared to those of the opposite sex. As a result, the national court must evaluate all relevant qualitative factors to determine whether such a disadvantage exists, considering all workers affected by the national legislation underpinning the differential treatment.

In this context, statistical evidence is merely one of several factors available to the national court. Where statistical data is available, the Court may refer to it to establish the existence of indirect discrimination within the framework of ensuring equal treatment between men and women. Furthermore, as established in the Court's case law, such a particular disadvantage may be demonstrated, for instance, by showing that the national legislation disproportionately affects a significantly larger proportion of one sex compared to the other. ¹⁶

Regarding the use of statistical evidence in cases of indirect discrimination, the most significant aspect of the KfH Kuratorium judgment is the CJEU's explicit clarification that to establish indirect sex discrimination in cases, it is not necessary for full-time employees to include significantly more men than women. Instead, it is sufficient that the disadvantaged group—such as part-time workers—comprises a significantly higher proportion of women than men. ¹⁷

Despite its novelty in clarifying the need for statistical evidence, the judgment also reinforces the presumption that, in cases of pay inequality, women must experience the disparity as a group, rather than merely as individuals. While the interpretation of Article 2(1)(b) of 2006/54 in sex equality cases extends beyond the literal text of the Directive, requiring evidence that a seemingly neutral measure disproportionately affects one sex compared to the other, this standard is not consistently applied across other grounds of discrimination. In fact, when it comes to nationality-based discrimination, there is a long-standing tendency in CJEU jurisprudence to adopt a more lenient approach.

5.1. The issue of the notion ‘considerably higher’

The next question to be answered was how large the comparative groups must be to establish indirect sex discrimination. When the alleged discrimination stems from an employer's decisions or practices, the Court has maintained that statistical evidence must reflect the actual conditions at the specific workplace. In the Bilka-Kaufhaus case, the statistical evidence was limited to illustrate how many women and men worked part-time at the store in question. But as cases of indirect sex discrimination have increased, the Court has progressively tightened the requirements for statistical evidence. In the Enderby case, for instance, it ruled that statistical data must include a sufficiently large sample to avoid reflecting mere random or short-term fluctuations, and the data must hold general significance. ²⁰ Moreover, even though the Court stated that if the pay of speech therapists is significantly lower than that of pharmacists and if the former are almost exclusively women while the latter are predominantly men, there is a prima facie case of sex discrimination if the two jobs in question are of equal value, but still insisted that the statistical evidence must be robust. ²¹

The Royal Copenhagen case further specified these requirements. It established that all individuals in the comparative groups—men and women—who are in similar situations must be included in the analysis. To ensure that wage disparities are not due to chance or individual productivity, the comparison must include a sufficiently large number of workers to rule out short-term fluctuations or variations in individual output. ²² Given such stringent criteria, proving indirect sex discrimination in workplaces with few employees has become particularly challenging. For smaller workplaces, it has remained difficult to provide substantial statistical evidence demonstrating that significantly more women than men are disadvantaged by an apparently sex-neutral provision. Another example is the Kirshammer-Hack case, where despite evidence that 90% of part-time workers in the German labour market were women, the Court and the Advocate-General dismissed the data as insufficient to prove that women were disproportionately affected by employment security regulations in small firms too. ²³ These cases prove that a lack of publicly available data on gender distribution among part-time workers can effectively prevent plaintiffs from establishing a presumption of indirect discrimination, because excessive requirements limit their ability to demonstrate systemic discrimination.

However, the Court has never explicitly defined what proportion constitutes a ‘substantially higher’ threshold in cases of indirect discrimination. Research indicates that in successful claims of indirect sex discrimination, the disadvantaged group has typically comprised 80% or more of those affected. ²⁴

In Seymour-Smith and Perez, the plaintiff argued that even a small difference in proportions should be sufficient to establish a breach of equal treatment, provided that statistics demonstrate long-term, non-random sex differences. ²⁵ However, in that case, 77.49% of men and 68.9% of women met the set conditions, and the Court concluded that this statistical disparity was not significant enough to constitute indirect discrimination.

Nonetheless, numerical data alone are not sufficient to determine whether discrimination is present. Importantly, the KfH Kuratorium decision leaves open the possibility of proving discrimination through means other than statistical evidence, allowing for a broader interpretation of what constitutes indirect discrimination.

5.2. A more lenient approach in nationality-based discrimination

The requirement for statistical evidence in sex discrimination cases ensures that the disadvantage is directly linked to sex and not influenced by other factors. However, when we compare the evidential requirements in cases relating to other grounds of discrimination, most notably, discrimination based on nationality, we can observe that the burden on the plaintiff is much lighter in those cases. In cases of nationality-based discrimination, the CJEU has interpreted the law in line with legislative intent, extending its scope beyond strict nationality-related discrimination. This shift has led the Court to move from a focus on discrimination to a broader discussion of obstacles, making it unnecessary to prove actual discrimination in nationality cases.

The KfH Kuratorium decision shows that to establish indirect sex discrimination, plaintiffs must present statistical evidence demonstrating that a provision disproportionately disadvantages a significantly higher percentage of women compared to men. In contrast, proving indirect discrimination based on nationality requires only that a measure is shown to pose a potential risk of disadvantaging migrant workers, without the need for statistical proof.

Already in the 1990s showcasing a quantifiable disproportionate impact was not a requirement. In O’Flynn, the Court formulated a broad test, emphasising that a provision of national law, even if applicable irrespective of nationality, must be regarded as indirectly discriminatory, and hence not complying with equality of treatment, if it is simply intrinsically liable to affect migrant workers more than national workers and if there is a consequent risk that it will place the former at a particular disadvantage. ²⁶ In Masgio, in order to establish indirect discrimination, it was sufficient to demonstrate that the consequence of a certain measure could deter workers from exercising their rights and would therefore constitute an obstacle to the exercise of their freedom of movement. ²⁷

Overall, these cases suggest that in nationality-based discrimination claims, the Court prioritises logical implication and risk assessment over exact statistical thresholds, making it easier for claimants to prove indirect discrimination without extensive empirical data. Achieving gender equality within the European Communities remains limited to prohibiting sex-based discrimination, making statistical evidence crucial in indirect sex discrimination cases. While nationality-based indirect discrimination can be demonstrated on an individual level, proving indirect sex discrimination requires plaintiffs to demonstrate a collective disadvantage.