Four Architectures of Pay Transparency

The OECD case studies show why pay-transparency systems should be compared as architectures rather than ranked by one attractive feature. This chapter places Belgium, France, Portugal and Spain against the same questions—what is compared, who is covered, who sees the result, who must act, and who checks the process—so the reader can judge causal usability instead of formal ambition.

The four countries are not controlled experiments, and their legal systems were not built from a common template. They are useful precisely because each makes a different link in the accountability chain more visible. Belgium relies heavily on sectoral bargaining and representative institutions. France combines an index with a public database and targeted enforcement. Portugal uses administrative data to calculate information for almost the entire private sector. Spain links broad registries to preventive audits and equality plans. None solves every problem.

The same test produces different national designs

A useful comparison begins with five shared dimensions. Comparison quality asks whether the system can identify unequal pay for the same work or work of equal value. Reach asks which employers and workers are included. Visibility asks whether information stays with management, reaches representatives, or becomes public and comparable. Action asks whether the employer must investigate, explain or correct a gap. Verification asks whether an authority checks filing, data quality and follow-up. Administrative burden cuts across all five: a system that cannot be completed reliably will have less practical reach than its legal coverage suggests.

flowchart LR T["Shared test: comparison, reach, visibility, action, verification"] --> B["Belgium"] T --> F["France"] T --> P["Portugal"] T --> S["Spain"] B --> B1["Sectoral bargaining and classification review; confidential reporting"] F --> F1["Index and public database; targeted follow-up"] P --> P1["Administrative calculation; reactive evaluation plan"] S --> S1["Registry and preventive audit; equality plan"] B1 --> O["Different strengths and failure points"] F1 --> O P1 --> O S1 --> O

The point of the diagram is not to create a scorecard. It is to keep the comparison at the same level. A public database may solve a visibility problem while leaving a classification problem. Administrative calculation may solve employer burden while leaving public pressure weak. A preventive audit may identify causes while worker access remains limited. The question is what each architecture makes possible and what it leaves to another institution.

Belgium: strong collective institutions, limited public visibility

Belgium’s private-sector system is anchored in sectoral collective bargaining. Wages are commonly set through sectoral agreements, and those agreements usually establish job classifications as well as minimum wage scales. Sectoral coverage is broad, so classifications negotiated by social partners reach much of the private sector. Belgian law requires classifications used in collective bargaining to be gender neutral and subjects them to public review.

That review is a distinctive strength. Sectoral classifications are submitted to the Federal Public Service for Employment, Labour and Social Dialogue and assessed with a statutory control instrument. The instrument checks more than the wording of job titles: it considers whether the method is analytical, whether governance is gender balanced, whether job descriptions and evaluation criteria are transparent, and whether the system covers dimensions such as skills, effort, responsibility and working conditions. A classification with gendered job titles fails the decisive neutrality test. A joint committee then has time to correct it; if it does not, an explanation and an eventual “name and shame” escalation can follow.

The arrangement gives Belgium a way to scrutinise the classification layer without imposing one national method on every sector. It also illustrates the value of social partners when pay structures are negotiated collectively. The limitation is that public control focuses primarily on formal and procedural gender neutrality. It does not systematically establish that the resulting job groupings accurately compare work of equal value or that they have been translated into fair pay outcomes.

Belgium’s pay-gap reporting reaches private organisations with 50 or more employees, which report every two years. The system is relatively detailed. Larger companies report pay by job function, seniority and education, as well as benefits and workforce information. Yet the reporting categories may be too broad for equal-value comparisons. Companies can use their own classifications, or default categories such as execution, executive and management staff. Those categories do not automatically reveal whether differently titled or differently situated jobs make comparable demands.