An unadjusted pay gap is the observed difference in pay between women and men before controlling for factors such as job, grade, location, experience or working pattern. Directive (EU) 2023/970 defines the gender pay gap using average pay levels and separately defines the median gender pay gap. Article 9 requires employers within scope to report those observed measures, together with variable-pay, quartile and category-level metrics. An unadjusted gap is therefore an important compliance and diagnostic measure, but it is not the same thing as a finding of pay discrimination. Employers should use it to identify where deeper category-level and equal-pay analysis is needed.
Jurisdiction: European Union
An Unadjusted Gap Is a Descriptive Measure
An unadjusted pay gap compares observed pay outcomes before the analyst controls for explanatory variables. In practical terms, it answers a simple question: how different are the pay levels of women and men in the population being measured? The result reflects the combined effect of workforce composition, occupation, grade, seniority, working time, variable pay, location and any other factor already embedded in the observed data. Because it is descriptive, the measure is useful for showing the size and direction of a disparity, but it does not by itself explain why that disparity exists.
The Directive Uses Average and Median Pay Gap Measures
Article 3 defines the gender pay gap as the difference in average pay levels between female and male workers of an employer, expressed as a percentage of the average pay level of male workers. It separately defines the median gender pay gap using the median pay levels of women and men. Article 9 requires both measures. This means an employer should not collapse the reporting framework into one preferred statistic. The mean-style headline gap and the median gap describe different features of the pay distribution and can move differently when high or low pay values are concentrated in one part of the workforce.
A Raw Gap Is Not Automatically Pay Discrimination
A large observed gap can arise from several sources. Women and men may be distributed differently across occupations, seniority levels, locations, working-time patterns or bonus-eligible roles. Some differences may be explainable by objective, gender-neutral factors, while others may reflect barriers, undervaluation or unequal pay that requires correction. The unadjusted metric does not resolve those questions. It identifies a disparity that should be examined. Employers should therefore avoid language suggesting that every raw gap equals discrimination, while also avoiding the opposite mistake of dismissing a gap merely because structural factors may contribute to it.
Small Headline Gaps Can Still Hide Category-Level Problems
An organisation-wide gap can be close to zero even when significant differences exist inside particular worker categories. Positive and negative gaps across departments or grades can offset one another in the overall average. Article 9 therefore goes beyond a single headline number and requires category-level gender pay gap information, broken down by ordinary basic wage or salary and complementary or variable components. Employers should use the overall result as a screening measure, then inspect the categories that correspond to the same work or work of equal value. That is where unexplained differences may become more visible.
Workforce Structure Often Explains Part of the Observed Gap
A useful diagnostic step is to decompose the observed pattern descriptively before moving to more complex statistical models. Analysts can compare the distribution of women and men by grade, job family, worker category, location, full-time or part-time status and access to complementary or variable pay. These views can reveal whether the headline gap is being driven mainly by representation at different pay levels or by differences within comparable groups. Descriptive decomposition does not replace the legal equal-pay test, but it can help the employer decide where to focus job-evaluation, remuneration and process review.
Document the Population, Period and Pay Definition
An unadjusted result is only reproducible if the methodology is clear. The employer should record which workers were included, the measurement period, the pay fields used, how working-time data were handled, how variable pay was treated and which records were excluded. The same label can otherwise hide materially different calculations. National implementation may prescribe more detailed rules, so the operational calculation should follow applicable national methodology. Internally, preserving source fields and transformation logic allows the organisation to explain why a published or audited result differs from another analysis using a different population or period.
Frequently Asked Questions
What is an unadjusted pay gap?
It is the observed difference in pay between groups before controlling for explanatory factors such as job, grade, location, experience or working pattern.
Does an unadjusted gender pay gap prove discrimination?
No. It identifies an observed disparity that may have multiple causes. Equal-pay compliance requires deeper analysis of comparable work, objective criteria and the reasons for differences.
Can an employer have a small overall gap and still have equal-pay issues?
Yes. Organisation-wide results can mask larger category-level differences, which is one reason Article 9 also requires information by categories of workers.
Related Guides
Official Sources
Requirements and practices differ by jurisdiction and organisation. Check current local law, official guidance and professional advice for a specific situation.