10. Summary of recommendations

This pilot study has explored how available official statistics can be used to measure pay differentials between women and men in work of equal value at the national level in the Nordic countries, with a particular focus on Finland, Norway and Sweden. As stated previously, this report makes no claims about how equivalence between occupations can be determined at the national level or which occupations should be considered as ones in which work of equal value is performed. The study is written as if it is possible to identify occupations in which work of equal value is performed, leaning on the Swedish expert network Lönelotsarna’s (Harriman et al., 2023) valuation of occupations based on the Swedish occupational classification. In light of the explorative analysis of national statistics carried out in this study, some tentative recommendations are provided for facilitating the collection of relevant statistics on pay differentials for work of equal value across occupations at the national level.

The first recommendations concern national occupational classifications. The authors recognise that occupational classifications serve a variety of purposes and that collecting statistics and introducing changes presents challenges. Thus, the following recommendations are to be considered a wish list based solely on the need to compare pay between jobs in which work of equal value is performed. Currently, each of the three countries surveyed already follows one or more of the recommendations, but they differ in those they follow and none follows all of them.

As the classifications are a relevant starting point for identifying work of equal value across occupations, it is important that they present information on occupations at the same level of detail, as far as possible. At present, classifications at the four-digit level are generally the most detailed with regard to defining an occupation for statistical purposes. However, the ‘map’ provided by the occupational classifications does not always coincide with reality or fulfil the required demands. An overall recommendation is therefore to:

Revise the occupational classifications

Large, female-dominated occupational groups may need to be broken down into more occupations, while smaller male-dominated occupational groups in which relatively similar work is performed could be merged. Individual countries can sometimes draw inspiration from their Nordic neighbours in terms of reviewing the occupational classifications and determining the level of detail in which different occupations are classified. 

The remaining recommendations from the pilot study concern the basic requirements that official, national statistics would need to fulfil to facilitate analysis of pay differences between women and men in work of equal value in occupations at the national level in the Nordic countries. The recommendations can be seen as complementing and expanding on what is prescribed in the EU Pay Transparency Directive 2023/970 and the reporting requirements for Agenda 2030, Target 8.5.1. They cover statistics that should be available for analyses of gender and pay for work of equal value at the national level and also which statistics would be relevant for such analyses:

It is a basic requirement that statistics on the number of men and women within occupations at the 4-digit level are available in the same table as pay.
Even if it is not possible to provide mean and median pay for reasons of privacy, in cases in which there are very few women or men within an occupation, the overall number of women and men should be available.
Data on both mean and median pay for occupations in the classification system should be available.
The examples in the pilot study demonstrate that the pay gap can vary depending on the measure used, but the patterns are not always the same for different occupations, justifying the use of multiple statistical measures.
Pay data should be available for each sector (private and public, public divided into central and local government) and the sectors combined.
Patterns in pay vary across different sectors of the labour market: for example, the pay gap in the private sector is often larger than in the public sector (Hoen et al., 2024). If the work performed in an occupation is of equal value across sectors, the overall average, for the sectors combined, should serve as the basis for pay comparisons. Analyses of the gender pay gap for work of equal value should, according to the EU Pay Transparency Directive, allow for comparisons across sectors.
National statistics should include easily accessible data on the number of women and men working full and part time in each occupation and on full- and part-time pay.
Given the importance of working hours in determining actual pay, and the fact that women work part time to a greater extent than men in the Nordic countries, it is important to analyse the impact of part-time work on pay and pay differentials between different occupations. In this context, it is particularly important to consider the way in which part-time and full-time pay are made comparable.
Data on the age structures of occupations in the same table as pay are needed.
Age structures can affect both the equivalence of occupations and their mean and median pay. A cursory check of whether otherwise comparable occupations have very different age structures may require that the overall pay gap be specified with respect to different age groups.
Statistics should include the number of native- and foreign-born persons within occupations, broken down into women and men, and their mean and median pay, at least for occupations with many foreign-born workers.
The EU Pay Transparency Directive recognises the importance of taking intersectional discrimination into account. Instances of possible structural discrimination at the national level can be detected by comparing the pay of native and foreign-born workers. In occupations with many foreign-born workers, their pay levels can affect the gender pay picture of the occupation and thus also gender pay comparisons with occupations of equal value.
Statistics on education should be easily linkable to data on gender and pay.
Data on education in the national statistical databases is relatively coarse-grained, although to different degrees in the three countries. The fact that education and pay are separate statistical domains means that the data on education and pay cannot be easily linked. Analyses of pay differentials based on education can highlight the pay gap between women with higher education and men with lower education. Many educational qualifications are directly linked to different occupations, and it would be beneficial for combined statistics on education and pay to be sufficiently detailed to enable occupations to be identified and compared on this basis.
It should be easy to follow developments over time.
Time series are needed at a minimum for data on mean and median pay and the number of men and women in occupations at the four-digit level of the classification systems, both by sector and for the sectors combined. The development of the current pay gap within and between occupations in which work of equal value is performed can be useful for making predictions about or influencing the future development of the pay gap. For example, changes in the gender pay gap may relate to changes in the gender composition of occupations or changes in the shares of the private and public sectors.