Dutch positive action measures in higher education in the light of EU law

In June 2020, the Netherlands Institute for Human Rights delivered an opinion on a positive action program run by the Eindhoven University of Technology. In this legally non-binding opinion the Institute declared the measures incompatible with the Dutch Equal Treatment Act and several EU provisions and CJEU judgments. The program reserved all vacant positions exclusively for women for a number of years. Consequently the Eindhoven University of Technology revised the program and the Netherlands Institute for Human Rights approved this revised program in March 2021 in another opinion. These opinions raise a number of interesting questions concerning EU non-discrimination law that are worth investigating further. Foremost is the question of how to interpret EU law on positive discrimination after years of silence on this issues from the CJEU.


Introduction
On 1 March 2021 the Netherlands Institute for Human Rights (the Institute) delivered its second opinion on the Irène Curie fellowship program of the Eindhoven University of Technology, finding the revised program compatible with Dutch and European Union (EU) non-discrimination law. 1 This Institute can be called upon by individuals or organizations. 2 The opinion came after the Eindhoven University of Technology revised the original program that the Institute had declared to be in breach of Dutch and EU non-discrimination law nine months earlier. 3 In its evaluations the Institute mainly used the Dutch Equal Treatment Act to judge the lawfulness of the program, however, it also drew heavily on the relevant EU law and rulings of the CJEU on positive discrimination. The program under scrutiny was a positive action program for women run by the Eindhoven University of Technology, which originally reserved all vacant academic positions for a number of years only for women in order to get more women into academic positions from Assistant to Full Professor. On the scale of positive action measures, the reservation of positions for one gender is a heavyweight. With measures ranging from consciousness-raising and targeting underrepresented groups with job advertisements to offering special career trainings and, as in our case, quotasystems, positive action measures have many faces. They not only play a role in the field of gender equality but they are also explicitly allowed when combating discrimination based on racial or ethnic origin, religion or belief, disability, age and sexual orientation. 4 There is no obligation in EU law to introduce such measures, it is left to the discretion of the Member States or employer whether to establish such measures or not. However, when adopted, positive action measures have to comply with EU non-discrimination law. Most CJEU cases on positive action measures concern quota systems, like the one under scrutiny here, reserving a certain number of job positions for women or giving preference to women in the application or promotion procedure.
The Irène Curie fellowship program is not the only positive action program of this nature in higher education in Europe. Similar programs exist(ed) at other Dutch universities and in other EU member states. In 2020, Ireland even introduced such program on a nation-wide scale for higher education after consulting the Delft University of Technology and the Attorney General of Ireland. 5 The careful way the Irish Ministry for Higher Education prepared their women-only program and the media attention it received stand for the legal and the social relevance of this issue. Moreover, just by looking through internet advertisements for academic positions it is possible to find a number of calls for women-only positions, for example as part of the Leibnitz 1. Opinion of Netherlands Institute for Human Rights in Case 2021-19 Het herziene Irène Curie Fellowship-programma van de TU Eindhoven voldoet aan de eisen van voorkeursbeleid en is in overeenstemming met de gelijkebehandelingswetgeving (The revised Irène Curie Fellowship program at Eindhoven University of Technology meets the requirements of preferential policies and is in line with equal treatment legislation.), https://mensenrechten.nl/nl/oordeel/2021-19. 2. The opinions of the institute are not legally binding, they are rather tools to exert pressure to change certain practices. This institute is financed by the Dutch government. In a not very specific statistical analysis found on the Institute's website it says that 80% of its opinions are followed by some form of action, listing an apology and a change of policy as possible reactions (https://mensenrechten.nl/nl/college-voor-jou Programme for Women Professors of the Leibnitz Association. 6 The questions addressed are thus of relevance beyond the case of the Irène Curie fellowship program. The evaluation of the Netherlands Institute for Human Rights highlights many of the uncertainties concerning positive discrimination measures. I will take this evaluation as an opportunity to investigate these uncertainties. In order to do this, I will look at the two most important rulings on positive action measures in the field of gender equality of the CJEU, Kalanke and Marschall. 7 The main controversies surrounding positive discrimination measures can be distilled from these two rather old rulings. In addition, a number of later rulings of the CJEU on positive action measures confirming the conclusions from the Marschall ruling will be discussed in brief. The evolution of EU law since these rulings and the lack of recent case law of the CJEU on positive discrimination, made it, in its own words, difficult for the Institute to determine the state of EU law on this matter. A quick look at primary and secondary EU law will provide the background for the analyses of the case law and the opinion of the Institute. However, I will first describe the set-up of the Irène Curie fellowship program based on the facts found in the opinion.

The national context
The Irène Curie fellowship program corresponds to the Dutch national strategy to increase the percentage of women in top academic positions. 8 Sex has long been the main focus of the Dutch nondiscrimination policy in higher education and the most prominent and controversial initiatives concern women. Only lately have more grounds, such as ethnicity, sexuality and religion, been added to the mix with the goal to increase diversity at Dutch universities in general. 9 For female academics, a number of programmes exist or existed, both at national and university level. In 2018, 100 female professors were appointed in the Netherlands with financial support from the Ministry of Education, Culture and Science in addition to the usual appointments, the Westerdijk Impulse initiative. Moreover, the Dutch Research Council (NWO), the Royal Netherlands Academy of Arts and Sciences (KNAW) and the Association of Universities in the Netherlands (VSNU) set up the Aspasia grant, which supports the promotion of women who had a very good or excellent Vedi/Vici application to associate or full professor by financially supporting such promotions. Moreover, all Dutch universities publish target numbers for female academic staff (the University of Amsterdam with a 50% target being the most ambitious one) and have or are working on gender equality plans required to receive funds from the Horizon Europe research funding programme of the EU. 10 The Eindhoven University of Technology is not the only university in the Netherlands that has a rather far-reaching positive action program in place. Tilburg University has the Philip Eijlander Diversity Program, the Free University of Amsterdam has the Fenna Diemer-Lindeboom chairs, the University of Groningen the Aletta Jacobs chairs and the  11 The latter two will be addressed again below.

The Irène Currie fellowship program
Starting in 2019, the Irène Curie fellowship program of the Eindhoven University of Technology reserved all academic positions, from Assistant to Full Professor, for female applicants for the duration of five years, adding up to approximately 150 positions in total. 12 If after six months no interview with a female candidate had taken place, the position would open for all applicants. However, half of the nominated candidates had to still be female. The program was supposed to be evaluated after a period of one and a half years, with the idea that, if successful, some positions could again be opened for men. The program knows two more exceptions. First, if a male candidate is extraordinarily qualified, he can be appointed. Second, if a female Irène Curie Fellow has a talented male partner, a maximum of 10% of all positions can be given to male partners. It is unclear if the opening for applicants of all gender after six months and the consideration of extraordinarily qualified male candidates were designed as saving clausesclauses that save a positive action measures from being in breach of EU lawbut these details will play a role in the analysis below. This program was meant to combat the quite extreme underrepresentation of women at the Eindhoven University of Technology. 13 The facts of the case hint at differences concerning the underrepresentation of women between the nine faculties without specifying them. The Eindhoven University of Technology argued that the program was necessary as its earlier attempts to tackle the underrepresentation of women did not achieve the desired result. These earlier attempts were apparently limited to verbal commitments by the University and thus, unsparingly, failed to bring about the change intended. The long-term goal of the Irène Curie program was to have at least 35% female Associate Professors and 50% Assistant Professors. 14 The University argued that if no drastic measures were taken now, the gender imbalance would persist for years to come. The University also claimed, probably meant as a form of defence, that its attempt is in line with government policy, which aims to increase the number of female scientists. An antidiscrimination civil society organization called radar had received around 50 complaints about the Irène Curie fellowship program and decided to ask the Institute for an opinion. 15 The university itself could have asked the institute for an opinion on the program before it started it but did not do so. In order to avoid unnecessary grievances, to ensure legality of the program and for the sake of procedural fairness, they should have requested the opinion because of the striking resemblance of the measure to other schemes which had previously failed legal scrutiny. The complaints received by civil society organization are not described in the opinion but it can be safely assumed that while referring both to Dutch and EU law they contained accusations of direct discrimination based on sex.

Primary and secondary EU law
Discrimination on the grounds of sex was already prohibited by the founding treaty of the European Economic Community (Article 141 EEC). In 1976, the Equal Treatment Directive introduced a provision allowing for positive action measures meant to remove existing inequalities, thereby explicitly mentioning women's opportunities concerning access to employment, including promotion, vocational training, working conditions and social security (Article 2(4) in conjunction with Article 1(1)). 16 In the early 1980s the discussion on positive action measures started in earnest. The main actors agreed that positive action measures were an appropriate tool. However, they disagreed on the degree of obligation with which the Member States should have to implement such measures. 17 The CJEU addressed positive action twice in the 1980s without adding much to the debate. 18 There were no provisions in the Treaty concerning positive action before the Treaty of Amsterdam (ToA). With the ToA, a paragraph was inserted into Article 141 EC that allowed measures that provide 'for specific advantages in order to make it easier for the under-represented sex to pursue a vocational activity or to prevent or compensate for disadvantages in professional careers.' Declaration 28 attached to the ToA made it clear how the new paragraph was to be understood, namely, that the measures referred to in this paragraph '(…) should, in the first instance, aim at improving the situation of women in working life'. After the ToA entered into force, the European Commission considered Article 2(4) of the Equal Treatment Directive to be redundant. In order to ensure coherence between the ToA and the Directive, the European Commission suggested amending the Directive. The proposal included changes to almost all the other Articles of the Directive as well. The relevant provision that was now in Article 2(8) of the new Directive (Directive 2002/73/EC) merely pointed to Article 141(4) ToA. In the latest consolidated version (Directive 2006/54/EC), which combines existing Directives on equal pay, equal treatment, occupational social security and the burden of proof, the reference to the treaty article can be found in Article 3: 'Member States may maintain or adopt measures within the meaning of Article 141(4) of the Treaty with a view to ensuring full equality in practice between men and women in working life'.
With the Lisbon Treaty, Article 141 EC became Article 157 TFEU but its wording remained unchanged. In addition, the Charter of Fundamental Rights of the EU (CFREU), binding since the same year, also allows for the possibility of positive discrimination. Article 21 CFREU forbids discrimination based on sex and 23 CFREU contains the same formulation on positive action measures as Article 157 TFEU but instead of saying that these measures should compensate 'for disadvantages in professional careers' it says 'for disadvantages of the under-represented sex', thus arguably moving beyond the field of employment.

The CJEU's case law
The discussion concerning positive discrimination can best be captured by analysing what the Court said in Kalanke and Marschall. In the Kalanke case, the city of Bremen was looking for a section manager of the city's Parks Department. Two applicants were considered by the relevant board to be 16 equally qualified for the job: Mr Kalanke and Ms Glissmann. The board decided that the position should be given to Ms Glissmann because women were underrepresented in the pay bracket in question. The board based its decision on the Land Law on Equal Treatment. The Federal Labour Court found nothing wrong with the decision of the board concerning the equal qualification of Mr Kalanke and Ms Glissmann. It also saw no conflict with any national provisions. It even added its own saving clause to the provision by saying that the provision has to be interpreted in accordance with the Constitution, which means that 'exceptions [to the priority for women] must be made in appropriate cases'. 19 What remained were concerns that the Land Law on Equal Treatment was incompatible with Directive 76/207. Thus, the Federal Labour Court asked the CJEU via the preliminary ruling procedure for an interpretation of Article 2 (1) and (4) of the Directive mentioned above. The CJEU argued in the grounds of the judgment that: (…)in so far as it [a system of national rules] seeks to achieve equal representation of men and women in all grades and levels within a department, such a system substitutes for equality of opportunity as envisaged in Article 2 (4) the result which is only to be arrived at by providing such equality of opportunity. 20 The Court then followed the proposal of the Advocate General (AG) Tesauro and ruled that provisions that automatically give priority to women are not allowed by the Directive. In short, the Bremen rule was declared incompatible with European Community law.
At the time, the ruling triggered fierce responses from legal experts, the European Parliament (EP) and the European Commission. Legal experts criticized the ruling because of its lack of clarity and/or its contents. The lack of clarity is seen as a result of the lack of reasoning found in the ruling. 21 How unclear the ruling was can be illustrated by two cases concerning a provision of the Beamtengesetz für das Land Nordrhein-Westfalen. The Oberverwaltungsgericht Münster found unlawful a similar provision as the one at issue in the Kalanke case. It based its decision explicitly on Kalanke. 22 Only two days later the Verwaltungsgericht Gelsenkirchen asked the CJEU for a preliminary ruling in another case concerning the same provision. 23 This preliminary ruling reference will become known as the Marschall case, to which I will return later. During the Kalanke proceedings, the EP Committee on Women's Rights had already criticized the opinion of AG Tesauro before the CJEU delivered its ruling because AG Tesauro had called upon the CJEU to 'resist the temptation to follow the dominant current'. 24 About a week after the ruling, the judgment was debated by the EP in plenary session. 25 A majority of the speakers criticised the ruling as a step back in the endeavour of achieving equality between women and men. Many of these speakers advocated a change of the Treaty at the upcoming Intergovernmental Conference (IGC) that would make such a ruling in the future impossible and called for amending the Equal Treatment Directive. Only a few speakers welcomed the ruling. Half a year after the ruling, the European Commission formulated its interpretation of Kalanke in a communication to the EP and the Council. In this communication the European Commission stressed its view that the ruling only applied to laws that give women 'absolute and unconditional priority' like the Bremen legislation. In other words, the ruling does not forbid positive action in general but 'a rigid quota system under which there is no possibility of taking particular individual circumstances into account'. 26 An interpretation that was shared and reiterated by the CJEU in its later preliminary rulings on positive action measures, like Marschall.
Mr Marschall and a female applicant were considered to be equally qualified for the job. However, as women were underrepresented in the relevant pay and career bracket the female candidate was appointed. The decision was based on the Law on Civil Servants of the Land of North-Rhine-Westphalia, which required that women are given priority when they are equally qualified and underrepresented. However, an exception to this rule can be made for reasons specific to the male candidate. Note that this latter rule makes the law different than the one in question in Kalanke. Mr Marschall then tried to revise that decision by going to court. The German court referred the case to the CJEU for a preliminary ruling, asking if such a rule was compatible with Article 2 (1) and (4) of Directive 76/207/EEC. AG Jacobs beseeched the CJEU to resist 'any temptation (…) to tailor the results to policy, however attractive it may seem (…)'. 27 Nevertheless, the Court ruled against the Opinion of the AG and argued that the national rule in question is not precluded by Article 2 (1) and (4) of the Equal Treatment Directive under the condition that: (…) in each individual case the rule provides for male candidates who are equally as qualified as the female candidates a guarantee that the candidatures will be the subject of an objective assessment which will take account of all criteria specific to the candidates and will override the priority accorded to female candidates where one or more of those criteria tilts the balance in favour of the male candidate. 28 In other words, the Court accepted the saving clause as sufficient to distinguish this case from Kalanke. Even more interesting is that the CJEU changed its argumentation. In Kalanke it argued that equal representation should not be confused with equal opportunities which would mean that measures aiming at equal representation are not covered by Article 2(4) Directive 76/207. The CJEU seems to have taken a more substantial approach in its argumentation in Marschall by saying that because of prejudices and stereotypes 'the mere fact that a male candidate and a female candidate are equally qualified does not mean that they have the same chances'. 29 More rulings on positive action that followed were Badeck, Abrahamsson, Lommers and Briheche. 30 These will be discussed in brief. In the year 2000 the CJEU ruled twice on positive action, in the Badeck case and in the Abrahamsson case. In Badeck a German rule reserving 50% of available places for women in a state-run vocational training programme came under scrutiny. The Court decided that given that the state does not have a monopoly of training and that enough women apply, the Court declared a fixed quota of 50% compatible with the Directive, in case enough qualified women apply. It argued that an advancement plan, such as the one in question in Badeck, may provide for a 'minimum percentage of women which is at least equal to the percentage of women among graduates, holders of higher degrees and students in each discipline'. 31 It did not explicitly forbid targets that go beyond this minimum but it must be kept in mind that this ruling concerned training positions for which the state did not hold the monopoly. The ruling also stated that candidates must be 'subject of an objective assessment which takes account of the specific personal situations of all candidates' (operative part of the judgment). In Abrahamsson the CJEU had to decide again on positive action measures. 32 The CJEU repeated its reasoning found in Kalanke and Marschall, namely that automatic preference without any consideration of the 'specific personal situations of all candidates' is not covered by Article 2(4) of Directive 76/207. 33 Although Lommers did not concern any kind of quotas, the ruling shows the Court's concern with individual rights by its insistence on a saving clause. Mr Lommers' employer, a Dutch ministry, refused to grand Mr Lommers' child the use of the subsidised nursery scheme. The ministry argued that only female employees may use the nursery scheme for their children in order to tackle the under-representation of women working for the ministry, only in exceptional cases of emergency may male employers make use of the nursery. Adding its own saving clause, the CJEU ruled that these exceptions are to be 'construed as allowing those of them who take care of their children by themselves to have access to that nursery places scheme on the same conditions as female officials.' 34 In 2004 the CJEU delivered yet another, and to my knowledge the latest, 35 ruling on positive action measures in favour of women in the context of employment: Briheche. In this judgment the CJEU ruled that Directive 76/ 207 EC precludes '(…) a national provision (…) which reserves the exemption from the age limit for obtaining access to public-sector employment to widows who are not remarried who are obliged to work, excluding widowers who have not remarried who are in the same situation'. 36 However, positive action measures have also been discussed in the field of social policy. The CJEU struck down pension schemes that only applied to women because of missing saving clauses in Griesmar and WA, and a rule that reserved parental leave only for mothers in Alvarez for the same 31 reason. 37 In Leone, another case a concerning pension plan, the Court argued a little differently by saying, that the plan did not address the problems women encounter during their career and is thus not covered by the exception to equal treatment provided for in Article 141 (4). 38 In all of its rulings, the CJEU treated these positive action measures as an exception to equal treatment. This approach has been criticized from Kalanke onwards. Legal scholars, 39 the European Women's Lobby, 40 the European Commission, 41 and representatives of the European Parliament and the Council of Ministers, 42 all criticized that the CJEU saw positive action as an exception to equal treatment and not a tool to combat indirect discrimination. The later rulings of the CJEU were seen as only forbidding strict quotas without a savings clause and the critique somewhat subsided without disappearing completely. 43 It can be argued that Article 141(4) TEC and later (157(4) TFEU) call for a reclassification of positive action measure as a useful tool in achieving equality in practice and not as an exception to the principle of equal treatment. However, the CEJU would need a case to be able to revisit and revise its interpretations.

Where does the Treaty and the case law of the CJEU leave us?
The non-discrimination law of the EU is characterized by the balancing of a formal approach and a substantive approach to equality. 44 The formal, procedural approach emphasises that like cases should be treated alike. In contrast, the substantive approach pays attention to context and its aim is to achieve outcomes that are considered socially desirable. 45 EU law is somewhere in between these two poles but arguably has moved from a rather formal to a more substantive approach over time, a development visible in the re-wording of the relevant provisions and driven by the case law of the CJEU on concepts such as indirect discrimination, obligatory differential treatment and positive action measures. The approach taken by the CJEU has been described as an equality of opportunity approach. 46 From the case law of the CJEU it follows that positive action measures have to be able to stand the test of proportionality. The CJEU uses a proportionality test to determine the legality of positive action measures with a strong focus on legitimacy and less attention to effectiveness and necessity. This focus in the application of the proportionality principle also explains the lengthy reasoning of the Eindhoven University of Technology when justifying the Irène Curie fellowship program. The University seemed to have been unsure of which imbalance is big enough to warrant positive action measures and wanted to be on the safe side by supplying a vast amount of numbers. It is true, as the Institute stated, that the last rulings of the CJEU on positive action concerning gender has been almost two decades ago. What does this mean for a contemporary legal evaluation of such measures? There is no indication that the new wording of the Directive in combination with the wording of Article 157 TFEU will make a difference to the approach which the CJEU has taken so far. However, the equality principle has gained legal influence by now being part of the EU's wider human rights approach, as, for example, expressed through its CFREU. The relevant article on positive discrimination in this charter, Article 23 CFREU, is not stricter but on the contrary, seems to leave more room for positive action than the Equal Treatment Directive or Article 157 TFEU, in which positive action measures are restricted to vocational training and professional careers. As the latest wording in the relevant Directive, the TEU or the CFREU is definitely not stricter than before, there is no legal reason for the CJEU to backtrack on its previous case law should a case on positive action measures, like the one of the Eindhoven University of Technology under discussion here, be transferred to it. What might help to remedy the problem of the temporary remoteness of the last preliminary rulings delivered on the subject by the CJEU is a look at how positive action is dealt with by the CJEU in concerning other grounds of discrimination. De Vos does just that by looking for clues in the CJEU's more recent rulings on discriminations on the grounds of age and disability. 47 He argues, taking the latest case law on discrimination based on disability as an indicator, that: 'It may well inspire a rethink of the more formalistic gender case law on positive action, if and when the CJEU gets the opportunity to do so.' 48 A less formal approach taken by the CJEU can be illustrated by the Milkova case from 2017. This case concerned Directive 2000/78 which the CJEU interprets in the light of the United Nations Convention on the Rights of Persons with Disabilities. 49 The CJEU emphasized the importance of the goal of the General Framework Directive arguing that: The purpose of Article 7(2) of Directive 2000/78 is to authorise specific measures aimed at effectively eliminating or reducing actual instances of inequality affecting people with disabilities, which may exist in their social lives and, in particular, their professional lives, and to achieve substantive, rather than formal, equality by reducing those inequalities.

The two earlier opinions of the Institute
The Irène Curie Program is not the first Dutch positive action program for female academics under scrutiny. In an earlier opinion the Institute allowed a positive action program of the Delft University of Technology which reserved 10 tenure-track positions exclusively for female applicants without any saving clause. 51 In this case, the Institute argued that the extreme underrepresentation of women made the program acceptable. The Delft University of Technology had after all tried with different measures to hire more female academic staff and the fact that the program only concerned a total of 10 positions put it within the boundaries of law. The Institute explicitly acknowledged that this opinion goes against the case law of the CEJU which forbids the complete exclusion of one gender. The Institute argued that the new text of the new Directive, allowing measures 'to ensure full equality in practice between men and women', was different than at the time the CJEU delivered its rulings when it allowed measures 'removing existing inequalities which affect women's opportunities'. A more thorough explanation is not given. In line with the national trend, the number of women in higher academic positions has increased in Delft over the years but the percentage of female full professors (17%) is still the lowest of all Universities in the Netherlands (26%). 52 Only a year earlier, the predecessor of the Netherlands Institute for Human Rights considered a promotion program of the University of Groningen exclusively reserved for women to be in breach of EU law, stressing that promotion programs have to be open to men and only in case of equal qualification, can the underrepresented sex be promoted. 53

The Institute's opinions on the Irène Curie Program
The two opinions on the Irène Curie Program from 2020 and 2021 from the Institute for Human Rights fall somewhere in between these earlier opinions, the one on the University of Groningen being quite strict and the one on the Delft University of Technology being quite lenient. Throughout its first opinion on the Irène Curie Program from 2020, the Institute also refers to the case law of the CJEU extensively. In this opinion, the Institute first had no problem deciding that the program constitutes direct discrimination. The Institute then tested the program against five standards. These had been developed in the two earlier opinions discussed above and have to be fulfilled in order to justify a positive action program under Dutch and European law. First, such a program has to have as its objective to reduce de facto inequalities of women (EU law would speak of the underrepresented sex). Second, underrepresentation has to be objectively shown and related to the existing supply. Third, the advertisement has to clearly state that a position is part of a positive action program and at the same time that the position is open for all. Fourth, there has to be an objective assessment of the candidates and positive action can only occur in cases of equal qualification. An absolute preference for one gender is prohibited. Fifth, the measure has to be proportionate and justified by the level of underrepresentation. 54 The legitimacy of the objective and the underrepresentation of women was not doubted by the Institute. However, the program failed to fulfil the criteria for the advertisement phase as it stated that only women will be considered in the first six months. In addition, it failed to apply positive action only in cases of equal qualifications. The Institute then asked itself if the program might nevertheless be necessary to achieve full equality between men and women. It considered the application of the program in all faculties independent of differences not to be proportionate. From the fact that in practice only 2 out of 35 positions were given to male candidates and the remaining 33 to female candidates, the Institute concluded, that the interest of male candidates was insufficiently taken into consideration. Moreover, the Institute concluded that the University failed to show that such a far reaching measure was necessary as no other measures were tried. 55 The Institute based its guidelines on the rulings of the CJEU addressed above in the CJEU case law section and at the same time pointed out that the date of the rulings of the CJEU is a problem. After all, the wordings of the relevant laws have changed since the CJEU addressed this issue. Nevertheless, it did not connect any consequences to this observation and its opinion is solidly based on the existing, somewhat dated, case law.
As a reaction to the first opinion on the Irène Curie Program of the Eindhoven University of Technology, the University revised the program. In the words of Frank Baaijens, the rector of the University: The Institute for Human Rights acknowledges our goals and agrees that female faculty are underrepresented at the university. However, the Institute concludes that the program has been applied too broadly. It is appreciated that the Institute gives clear clues on what type of measures we can consider to reach a better gender balance. We will study the findings of the Institute, as well as their recommendations, to determine our next steps. 56 This was indeed thoroughly done. The program is now restricted to a specific level of positions in faculties in which the share of women holding such a position is below 30%. In addition, only 30% or 50% of new positions will fall under the program, depending on the evenness and unevenness of the number of positions. The opinion reproduces a calculation of the Eindhoven University of Technology stating that in 2021, 15% of open positions would fall under the program, leading to a considerable reduction of positions that fall under the program compared to the original set-up. This time the Eindhoven University of Technology itself asked the Institute for its opinion on the revised program. 57 The Institute now considered the program proportional as many positions fall outside of the program. It basically referred back to the reasoning found in its first opinion. In addition, it now suddenly referred to the Convention on the Elimination of All Forms of 54. Opinion of Netherlands Institute for Human Rights in Case 2020-53, para. 6.6. 55. Ibid., para. 6.7-7. 56. TU/e, 'Eindhoven University of Technology remains committed to gender equality', 3 July 2020. https://www.tue.nl/en/ our-university/departments/electrical-engineering/department/news/news-overview/eindhoven-university-of-technologyremains-committed-to-gender-equality/. 57. Opinion of Netherlands Institute for Human Rights in Case 2021-19. Discrimination against Women (namely, Article 4.1) and general recommendations of the UN Committee on the Elimination of Discrimination against Women on temporary special measures. These provisions consider such measures not to be discriminatory to make the point that positive action measures are not forbidden from the outset. 58 After declaring the program in line with the Dutch Equal Treatment Law, the Institute added a recommendation, saying that the new program might be too weak to reach its goal and that the percentages could be slightly raised.

An analysis of the analyses
There is no doubt that these kind of positive action programs are allowed under EU and Dutch law. Actually, there should have never been any doubt after Marschall but how to make them compatible with the principle of equal treatment is another question. It always comes down to the fact that the saving clause is too weak or that too big a share of positions is unconditionally reserved for women. It is difficult to predict what the CJEU would say today about these programs, the percentages show the persistent gap between men and women and little progress towards the goal of full equality in practice. After some early confusion, the CJEU generally accepted a saving clause including equal qualification, comparability of circumstances and a 50% goal (if there are enough applicants), so measures that demanded that in case of equal qualification the underrepresented sex must be promoted, if other circumstances are equal. A positive action measure that does not stop before equal representation is reached might be challenged on grounds of available qualified personnel. The reason I assume for the Eindhoven University of Technology to go to some lengths showing that there are women scientist available globally. However, if the goal of a positive action measure is to stimulate the underrepresented sex to work in a certain field, the 50% mark in case of equal qualification is an appropriate tool condoned by the CEJU and the actual availability should not matter. It would have several advantages: it never needs adjustment and is gender neutral; no complicated calculations are needed to determine how many percentages is the right number and for how many positions; it will arguably encourage the underrepresented sex to qualify for jobs that were traditionally dominated by one gender knowing that chances are fair; and it would work outside a university setting. Nevertheless, if not accompanied by other measures, from gender mainstreaming to awareness raising, any percentage is probably not worth the paper it is written on as the concept of 'equal qualification' leaves quite some discretion to the employer, starting with the formulation of the required qualifications to the weight given to family work, social commitment or unpaid activity. 59 Moreover, a great number of new Member States have joined the EU since the last ruling of the CJEU on positive action measures and their politicians and judges might bring new (or old) ideas to the table. The critique that the interpretation given by the CJEU is too formal and misses the will of the legislator as expressed in primary and secondary legislation has been around since its first ruling on this issue in Kalanke. what constitutes justifiable objectives that positive action can legitimately pursue, and make clear that there is a range of legitimate objectives.' 60

Conclusions
It is unclear why the Eindhoven University of Technology at first designed a programme which quite obviously violated EU law. Although the case law of the CJEU is not recent, it leaves little room for interpretation concerning the importance of a savings clause. It might nevertheless be time for a preliminary ruling of the CJEU on positive action measures in the field of gender equality to mitigate the insecurity created by the age of the CJEU rulings, an insecurity experienced by the Institute, practitioners, legal scholars and affected individuals alike. Twenty-six years after Kalanke, positive action measures seem to be as controversial as they used to be despite the CJEU giving some useful guidance over time on what are acceptable measures and what are not. The fact that it has been nearly two decades since the CJEU ruled on positive action measures in the field of gender equality does create some insecurity about how EU law should be interpreted. After all, EU law has seen a shift from a rather formal to a more substantial approach to equality which might lead the CJEU to evaluate positive action measures more sympathetically. More recent cases concerning other grounds of discrimination might contain a glimpse of how a more substantial approach could look like. Nevertheless, after pointing out that no recent case law of the CJEU on positive action measures in the field of gender equality exists, the Institute chose to apply an interpretation which was based on the dated case law of the CJEU. Correctly so, as it is the CEJU which should update its own interpretation and adjust it if necessary. Moreover, the case law of the CJEU from other fields of discriminating is not easily transferable. The essence of Marschall (but also of the interpretative communication of the European Commission after Kalanke) is that measures giving unconditional preference to the underrepresented sex are forbidden. At any rate, the old case law of the CJEU is quite clear on the matter and still allows for a wide range of different positive action measures, including preferring female candidates above male candidates if they are underrepresented and all other things being equal (the saving clause preventing unconditional preference). So the Eindhoven University of Technology could have easily gotten away with the rather strict reservation of all positions for women, if they would have added a more substantial saving clause, assuming that this would have also shown in the statistics of who was actually employed in the end. Whether the individual initiatives of universities or the Irish introduction of positive action measures stand for a wider European trend towards more positive action measures or not, is unclear to me. If they do, more national and European court cases are sure to come.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biography
Tobias