The article in one paragraph
Article 4(4) of the EU Pay Transparency Directive runs to five sentences. It requires pay structures to be built on objective, gender-neutral criteria covering skills, effort, responsibility, and working conditions, agreed with workers' representatives where they exist, with soft skills not undervalued. The article reads like a single legislative instruction. It is in fact the codification of a methodological consensus that took seventy years to build. This article traces that journey through the law, the institutional guidance that interpreted it, the international labour standards that paralleled it, and the academic research that informed all three. The thesis is straightforward. The Directive's methodological standard is not a novel imposition created in 2023. It is the synthesis of a long tradition of European jurisprudence, ILO standards, and academic research, brought together in legally binding form.
One legal point should be made clear at the outset. Article 4(4) does not mandate any specific proprietary methodology. It establishes methodological characteristics that pay structures must satisfy. Compliance is a matter of whether the methodology used by a given organisation meets those characteristics, not a matter of which named framework has been adopted. This article is therefore about the standard, not about any particular instrument designed to meet it.
1. The constitutional foundation
Equal pay between women and men has been a binding principle of European law since the founding of the European Economic Community in 1957. Article 119 of the Treaty of Rome established the principle in its original form, requiring each Member State to ensure that men and women received equal pay for equal work. The provision survived every subsequent treaty revision, becoming Article 141 of the Treaty on European Community under the Amsterdam Treaty in 1997 and Article 157 of the Treaty on the Functioning of the European Union under the Lisbon Treaty in 2009. The wording across those revisions has remained substantively unchanged.
The principle was developed through secondary legislation. Directive 75/117/EEC, the first Equal Pay Directive, introduced the concept of equal pay for work of equal value, broadening the legal frame beyond identical jobs to comparable work. That Directive was replaced by Directive 2006/54/EC, the Recast Directive, which consolidated the equal pay framework with related provisions on equal treatment and brought job classification systems explicitly into the legal scope. Article 4 of Directive 2006/54/EC requires that where a job classification system is used for determining pay, it shall be based on the same criteria for both men and women and so drawn up as to exclude any discrimination on grounds of sex.
That single sentence, written in 2006, contains the methodological logic that would be developed across the subsequent seventeen years. Where pay is determined through a job classification system, the system itself becomes a legal object. It can be assessed, challenged, and required to meet standards. That is the constitutional foundation.
2. The jurisprudence that defined work of equal value
The Court of Justice of the European Union elaborated the meaning of the equal pay principle through a series of landmark judgments. Each one added a specific dimension to what compliance now requires of pay structures and the methodologies used to build them.
Defrenne II (1976): direct effect
In Defrenne II (Case 43/75), the Court held that the equal pay principle enshrined in Article 119 of the Treaty has direct effect. It can be invoked by any individual before any national court. The principle is not a policy aspiration. It is an actionable right. The judgment established that equal pay was a fundamental principle of Community law, not a programmatic objective.
Rummler (1986): factor selection
Rummler (Case 237/85) is one of the most consequential judgments for the methodology of job evaluation. The case concerned a German classification system that valued physical effort heavily but did not adequately capture other forms of demand. The Court held that a classification system used to determine pay must take account of criteria for which workers of each sex may show particular aptitude. A system that emphasises factors typically associated with male-dominated roles, without giving equivalent recognition to factors typically associated with female-dominated roles, will indirectly discriminate even if applied consistently. The judgment shifted the legal scrutiny from the application of the criteria to the selection of the criteria themselves. Factor selection became a legal question.
Bilka-Kaufhaus (1986): objective justification
Bilka-Kaufhaus (Case 170/84) established the test of objective justification. Where a pay practice has a disparate effect on one sex, the employer can defend it only by showing that the practice corresponds to a real business need, is appropriate to that need, and is necessary to achieve it. The test is strict. Mere assertion of business rationale is not sufficient. The employer must demonstrate the connection between the practice and the legitimate aim with evidence.
Danfoss (1989): the transparency principle
Danfoss (Case 109/88) introduced what has come to be called the transparency principle. The case concerned an employer who used a pay system in which individual supplements were determined by criteria that the employer itself could not fully explain. The Court held that where a pay system is so opaque that workers cannot establish whether they are being paid equally, the burden of proof shifts to the employer. Transparency, in other words, is not optional. It is the condition that makes the equal pay right enforceable. Opacity is its own form of breach.
Enderby (1993): prima facie evidence
Enderby (Case 127/92) extended the burden of proof reasoning. The Court held that where statistics show a significant pay difference between a predominantly female job and a predominantly male job of equal value, that statistical evidence is itself sufficient to create a prima facie case. Once that case is established, the employer must objectively justify the difference. The judgment turned statistical disparity into legal evidence in its own right.
Brunnhofer and Kenny : the contours filled in
Brunnhofer (Case C-381/99) and Kenny (Case C-427/11) developed the principle further across individual pay supplements, classification structures, and progression systems. The pattern across the line of authority is consistent. The pay system is the legal object. The methodology used to build the system is what determines whether the obligations are met. The employer is the party who must demonstrate that the methodology is sound.
3. The international labour standards that paralleled the European law
While European jurisprudence was developing the legal architecture, the International Labour Organization was developing the methodological architecture. ILO Convention No. 100, the Equal Remuneration Convention, was adopted in 1951 and is now ratified by 167 countries. The Convention requires Member States to promote and ensure the application of the principle of equal remuneration for men and women workers for work of equal value. The principle is the same as the European one. The route to applying it is methodological.
The ILO operationalised the Convention through a series of guidance documents. The most significant is the 2008 step-by-step guide titled Promoting Equity: Gender-Neutral Job Evaluation for Equal Pay, authored by Professor Marie-Thérèse Chicha of the School of Industrial Relations at the University of Montreal. Chicha had earlier produced a comparative analysis of pay equity models (ILO Working Paper 49, 2006), examining how Canada, the United States, and several European jurisdictions had translated the equal pay principle into operational methodology.
The 2008 guide consolidated and operationalised the four-factor model in modern gender-neutral form: skills, effort, responsibility, and working conditions. The factors themselves originate much earlier in equal value frameworks and in ILO conventions, but the Chicha guide systematised them for contemporary application. The guide explained how each factor should be decomposed into measurable sub-factors, how weightings should be derived, how the assessment should be conducted, and what conditions had to be met to keep the methodology gender-neutral in operation. It was, and remains, the most influential single methodological document in the international pay equity field.
The Chicha guide drew on the comparable worth research programme that had developed in North America from the late 1970s onwards. That research, conducted by economists and sociologists working across legal, academic, and policy settings, had established a set of findings that the ILO guide translated into operational form. Three of those findings shaped the methodology decisively. First, the value of work cannot be assessed adequately by factors developed for one type of work and then applied to another, because factor selection itself encodes assumptions about what counts as valuable. Second, the work performed predominantly by women had been historically undervalued in commercial classification systems, not through individual prejudice but through structural features of the systems themselves. Third, methodological transparency was the precondition for any reform, because pay systems that could not be examined could not be challenged.
4. The academic research base
The academic literature that informed the ILO methodology developed across four decades. Six bodies of work matter most for understanding what the Directive's methodological standard ultimately requires.
Hochschild and the discovery of emotional labour (1983)
Arlie Hochschild's 1983 book The Managed Heart established that managing the emotions of others (customers, patients, students, colleagues, family members) is a distinct category of work with measurable mental and physical cost. The phrase emotional labour entered the research vocabulary. The implication for pay equity was substantial. Roles in which emotional labour is a primary requirement (care, nursing, teaching, customer-facing service, social work) had been classified by legacy systems as low-skill or low-demand. The research demonstrated that this classification was empirically wrong. The work was demanding. The systems used to measure it had not been designed to see what made it demanding.
Steinberg on the construction of skill (1990, 1992, 1999)
Ronnie Steinberg's research programme through the 1990s extended Hochschild's insight into the specific domain of job evaluation. Her 1990 paper in Work and Occupations argued that skill is not a neutral property of jobs but a category that reflects how organisations and the wider economy have come to value certain types of capability. The factors used in job evaluation systems reproduce those valuations. The 1992 paper applied this to widely-used commercial frameworks of the period, documenting how their factor structures embedded mid-twentieth century assumptions about work that had become increasingly disconnected from the work the economy actually performs. Her 1999 paper proposed concrete instruments for measuring emotional labour as a job evaluation dimension. The proposal sat largely unrealised in the academic literature for two decades.
Grimshaw and Rubery on undervaluation (2007)
Damian Grimshaw and Jill Rubery's 2007 framework on the institutional undervaluation of women's work, produced for the UK Equal Opportunities Commission and subsequently adopted into EU policy materials, identified the mechanisms through which female-dominated occupations are systematically underpaid. The mechanisms include the structural features of job evaluation systems, the historical anchoring of pay surveys to existing wage hierarchies, occupational segregation that limits the available comparators, and the cultural framing of certain work as less skilled. The framework provided the methodological diagnosis that the policy infrastructure could then act on.
Koskinen Sandberg on gender-neutral legitimacy (2017)
Paula Koskinen Sandberg's 2017 paper in Work, Employment and Society introduced the concept of gender-neutral legitimacy. Drawing on a detailed study of eighteen Finnish organisations, the paper documented how formal job evaluation systems can produce an appearance of objectivity that forecloses challenge to underlying pay structures, even when those formal systems are not being applied as designed. The mechanism is not deception. It is procedural formalism. A system that produces numbers, follows documented steps, and is administered by professionals carries institutional authority that the underlying decisions may not actually deserve. The paper identified opacity and procedural formalism as the levers through which historical pay structures sustain themselves through evaluation methodology.
England, Sorensen, Figart on comparable worth
Paula England, Elaine Sorensen, and Deborah Figart, among others, contributed across the 1980s, 1990s, and 2000s to the comparable worth research programme. Their work established the empirical case that occupational segregation was the dominant statistical driver of the gender pay gap, that the factors which had been used to set pay in segregated occupations were not commensurable across occupations as commonly assumed, and that the policy instruments that worked best were the ones that operated at the level of evaluation methodology rather than at the level of individual pay decisions.
How the research became operationalisable
The pattern across these bodies of work was that the diagnosis was clear, the methodological implications were specifiable, but the operational translation had not been made. The ILO Chicha guide is what made the operational translation. It took the academic findings, distilled them into a step-by-step methodology, and provided a framework that policymakers and employers could apply. Without the Chicha guide, the research would have remained academic. With it, the research became the foundation of an international standard.
5. The 2013 European Commission synthesis
In 2013, the European Commission published a Staff Working Document (SWD/2013/0512) titled Overview of CJEU case law on equal pay, with Annex 1 on Gender Neutral Job Evaluation and Classification Systems. The document is one of the most underappreciated pieces of European labour policy infrastructure of the period. It took the entire body of CJEU jurisprudence on equal pay, synthesised it into operational principles, and connected those principles to the methodological framework developed by the ILO.
Three features of the document are particularly important. First, it consolidated the case law into a structured statement of what European equal pay law actually requires of pay systems, drawing on the Commission's 1996 Code of Practice on Equal Pay for Work of Equal Value and the 1994 Memorandum on the same subject. The document is in effect a translation of decades of jurisprudence into practical guidance for employers, courts, regulators, unions, and HR practitioners. Second, the 2013 SWD explicitly states that its four-factor structure is consistent with the ILO 2008 guide. The international and European standards were thereby brought into formal alignment. Third, the document operationalised the methodological implications. It set out what gender-neutral factors should look like, how sub-factors should be constructed, how weightings should be approached, and what conditions had to be met to make the resulting evaluation defensible.
The 2013 SWD did one more important thing. It established the principle that job evaluation systems are not neutral instruments. They can themselves create or reinforce discrimination if their methodological design is inadequate. The document acknowledged openly that the discrimination-related roots of the gender pay gap have to be located in the methods used to differentiate between male-dominated and female-dominated jobs, and that such discrimination is much less conspicuous, concealed in the technicalities of determining the value of work and pay through classification systems.
6. Directive 2023/970 and the codification step
Directive (EU) 2023/970, the Pay Transparency Directive, was adopted by the European Parliament and the Council on 10 May 2023, with a transposition deadline for Member States of 7 June 2026. The Directive contains the most significant codification of methodological obligations in European equal pay law since the Recast Directive of 2006. Article 4(4) is the central methodological provision.
The five-sentence provision contains six distinct obligations that should be read against the trail set out above. The four mandatory factors (skills, effort, responsibility, working conditions) are the same four that the ILO Chicha guide formalised in 2008 and that the 2013 EC SWD adopted as the structure for European guidance. The requirement that criteria be applied in an objective gender-neutral manner is the codification of the principle developed across the Rummler, Bilka-Kaufhaus, and Danfoss line of authority. The explicit prohibition on undervaluing soft skills reflects concerns developed across the Hochschild and Steinberg research traditions, carried into policy through the comparable worth literature and the ILO operational guidance. The requirement that criteria be agreed with workers' representatives where they exist is the operationalisation of the social partner dimension of European labour law. The reversal of the burden of proof under Article 18 codifies the Danfoss and Enderby reasoning. The information rights granted to workers under Article 7 operationalise the transparency principle.
None of these elements is novel. The novelty is that they are now in binding primary legislation rather than scattered across treaties, directives, judgments, codes of practice, and international labour standards. Article 4(4) is the methodological standard that European labour law has been building toward since the Treaty of Rome. It is now law.
7. The EIGE 2026 toolkit and the continuing line
Article 4(5) of the Directive instructs the Commission to update Union-wide guidelines on gender-neutral job evaluation and classification in consultation with the European Institute for Gender Equality. In March and April 2026, EIGE and the Commission jointly published an updated step-by-step toolkit on gender-neutral job evaluation and classification. The toolkit is explicitly an update of the 2013 Commission Staff Working Document, aligned to the requirements of the Directive.
The toolkit provides operational guidance for employers of all sizes. It is voluntary. It is not a legal act. It does not, by its own terms, guarantee compliance. What it does is set out, in detailed form, what a methodology aligned to Article 4(4) looks like in practice. The four-factor structure is preserved. The sub-factor decomposition is elaborated. Templates, examples, and self-assessment tools are provided. The toolkit is the practical layer that the Directive's legislative text presupposes.
The publication of the EIGE toolkit completes a cycle that began with the ILO Convention of 1951. The international principle is now operationalised through European law, supported by binding case law, codified in a Directive, and supplemented by an authoritative practical toolkit produced by the European institution specifically responsible for gender equality. Each link in the chain has reinforced the next. The methodology is no longer being constructed. It is being applied.
8. What the trail tells us about the future of pay governance
The trail set out in this article is not just historical. It is predictive. The direction of travel is unmistakable across the law, the institutional guidance, and the research.
Pay structures are being treated as systems, not as collections of individual decisions. The legal scrutiny is shifting from the explanation of individual pay differences to the methodological soundness of the system that produced them. The question increasingly asked in proceedings and in joint pay assessments is whether the organisation can demonstrate that its system is fair, not whether each particular outcome can be justified after the fact.
What was a litigation principle under Danfoss is now an operational requirement. The Directive carries the transparency obligation into the everyday workings of pay administration, not only into the moments at which it is challenged. A pay system that the employer cannot fully explain is exposed by its opacity. A methodology whose weightings sit behind a paywall, whose factor structure cannot be examined, or whose scoring rules cannot be reproduced may struggle to satisfy the transparency and evidential standards the law increasingly expects.
Evidential governance is becoming the defining standard for pay systems. The combination of the Directive's reporting obligations, the joint pay assessment trigger under Article 10, the burden of proof reversal under Article 18, and the worker information rights under Article 7 means that pay decisions now require the kind of evidential support that other regulated business processes have long required. The methodology must be documentable. The application must be auditable. The outcomes must be statistically defensible. This is a substantial shift from the pre-Directive era, in which most organisations operated pay systems whose underlying methodology was a matter of consultancy relationship rather than legal record.
The boundaries between functions are dissolving. Equal pay was historically treated as a legal matter for legal teams, a grading matter for HR, a pay-setting matter for compensation specialists, and a structural matter for leadership. The trail of jurisprudence and policy described above collapses these boundaries. The methodology is simultaneously a regulated artefact, a grading instrument, a compensation tool, and a structural design choice. Organisations that continue to treat these as separate problems will find themselves outflanked by the scope of the Directive's obligations, which presuppose that they have been treated as one.
The shift from same-job logic to work-of-equal-value logic is the most consequential single change. Most pay systems still operate on a narrow comparison frame: the same job in the same function should be paid similarly. The Directive presupposes a broader frame. Roles of equal value, however differently described, should be paid comparably. That requires a methodology capable of producing comparable scores across roles, functions, and occupational boundaries. Narrow job matching cannot satisfy this. Analytical methodology can.
9. Conclusion
The methodological standard set out in Article 4(4) of the Pay Transparency Directive is unusually well documented in its provenance. The five-sentence provision sits on top of a seventy-year body of treaty law, jurisprudence, international labour standards, institutional guidance, and academic research. Each layer reinforced the next. The Court of Justice supplied the legal architecture. The ILO supplied the methodological architecture. The academic literature supplied the diagnostic foundation. The European Commission synthesised these into operational guidance. The Directive codified the result. The EIGE toolkit operationalised it for everyday use.
The implication for any organisation now reviewing the methodology it uses to evaluate work is that the standard the Directive requires is not novel and not contestable on principle. It is the standard that European labour law has been building toward for the entire post-war period. Methodologies designed today against the underlying research will align naturally with the legal standard, because the legal standard was built from the same research.
Methodologies designed in earlier eras, against the assumptions and survey infrastructures of the periods in which they were created, were built before the research existed and before the legal standard had taken its present form. Their architectural features reflect the conditions of their origin. Whether they meet the standard the Directive now requires is a matter for each organisation to assess. The trail set out in this article should make it easier to know what that assessment involves.
Sources and further reading
Treaty on the Functioning of the European Union, Article 157.
Council Directive 75/117/EEC of 10 February 1975 on the approximation of the laws of the Member States relating to the application of the principle of equal pay for men and women.
Directive 2006/54/EC of the European Parliament and of the Council of 5 July 2006 on the implementation of the principle of equal opportunities and equal treatment of men and women in matters of employment and occupation (recast).
Directive (EU) 2023/970 of the European Parliament and of the Council of 10 May 2023 to strengthen the application of the principle of equal pay for equal work or work of equal value between men and women through pay transparency and enforcement mechanisms.
Court of Justice of the European Union: Defrenne v Sabena (Case 43/75); Rummler v Dato-Druck (Case 237/85); Bilka-Kaufhaus v Weber von Hartz (Case 170/84); Handels- og Kontorfunktionaerernes Forbund i Danmark v Dansk Arbejdsgiverforening, acting for Danfoss (Case 109/88); Enderby v Frenchay Health Authority (Case 127/92); Brunnhofer v Bank der österreichischen Postsparkasse (Case C-381/99); Kenny and Others v Minister for Justice (Case C-427/11).
European Commission (2013). Commission Staff Working Document SWD/2013/0512: Overview of CJEU case law on equal pay, Annex 1 on Gender Neutral Job Evaluation and Classification Systems.
European Commission and European Institute for Gender Equality (2026). EU-wide guidelines on gender-neutral job evaluation and classification: Step-by-step toolkit.
International Labour Organization Equal Remuneration Convention No. 100 (1951).
Chicha, M.-T. (2006). A Comparative Analysis of Promoting Pay Equity: Models and Impacts. ILO Working Paper 49.
Chicha, M.-T. (2008). Promoting Equity: Gender-Neutral Job Evaluation for Equal Pay: A Step-by-Step Guide. International Labour Office, Geneva.
Hochschild, A. R. (1983). The Managed Heart: Commercialization of Human Feeling. University of California Press.
Steinberg, R. J. (1990). Social construction of skill: Gender, power, and comparable worth. Work and Occupations, 17(4), 449-482.
Steinberg, R. J. (1992). Gendered instructions: Cultural lag and gender bias in [proprietary system name omitted] job evaluation. Work and Occupations, 19(4), 387-423.
Steinberg, R. J. (1999). Emotional labor in job evaluation: Redesigning compensation practices. Annals of the American Academy of Political and Social Science, 561(1), 143-157.
Grimshaw, D. and Rubery, J. (2007). Undervaluing women's work. Equal Opportunities Commission Working Paper Series No. 53.
Koskinen Sandberg, P. (2017). The institutionalised undervaluation of women's work: The case of local government sector collective agreements. Work, Employment and Society, 31(4), 703-720.
