1. The Index’s coverage is very incomplete
Only 50 per cent of firms (44 per cent of SMEs and 93 per cent of firms with more than 1,000 employees) that are normally required to report do so, providing a score out of 100.
Only a quarter of private-sector employees are included in the calculation of the Index’s pay gap indicator.
2. The introduction of the Index has had no detectable effect on gender inequality in the firms concerned
Firms with fewer than 50 employees (not subject to the Index) and those with more than 50 employees (subject to the Index) show remarkably parallel trends in terms of gender inequality between 2010 and 2020. No change in trend is observed for firms subject to the Index following its introduction in 2018.
3. The Index tends to obscure the actual gender inequalities
The good results achieved by firms regarding pay gaps are linked to two questionable methodological choices made in calculating the Index’s pay gap indicator:
- the application of a tolerance threshold (all gaps of less than 5 per cent are set to 0)
- the use of men as the reference group to normalise the gaps (gaps within each group are divided by the average pay for men, which reduces them when women are paid less than men but increases them when the opposite is true).
4. Firms that report their Index are no more virtuous – in terms of their performance on gender equality – than those that do not
The administrative data used in the study make it possible to measure gender inequalities for both firms that report an Index and those that do not: firms that do not report their Index or state that they are unable to calculate it are no more unequal than the others. However, they report scores for the pay gap indicator that are higher than those recalculated from the administrative data.
It is observed that for the simple and transparent indicator—the proportion of women among the ten highest earners—firms’ reported figures match their reconstructed results, whereas this is no longer the case for the complex and adjustable pay gap indicator. This suggests that the use of complex and opaque indicators may enable firms to conceal their actual inequalities.